i
IMPROVED CREDIT EVALUATION SYSTEM OF TAOBAO COMPANY
KUANG YUNZHU
A project report submitted in partial fulfillment of the
requirements for the award of the degree of
Master of Science (Information Technology - Management)
Faculty of Computer Science and Information Systems
Universiti Teknologi Malaysia
APRIL 2010
ii
I declare that this project entitled “Improved Credit Evaluation System for TaoBao
Comapny” is the result of my own research except as cited in the references. The
project has not been accepted for any degree and is not concurrently submitted in
candidature of any other degree.
Signature :
Name : KUANG YUNZHU
Date :
iii
To my beloved family and friends
To my respected supervisor
iv
ACKNOWLEDGEMENTS
First of all, I will like to be thankful to University Technology of Malaysia for
giving us the opportunity to accomplish this work.
I acknowledge and deeply appreciate the worthwhile and limitless efforts of
my diligent and able supervisor, Dr. Othman Ibrahim to the successful
accomplishment of this project work.
I also acknowledge the limitless effort of the head of department, Dr. Mohd
Zaidi Abd Rozan. I wish to acknowledge and deeply appreciate Dr. Ab Razak Che
Hussin who has imparted some knowledge onto me at some time.
My appreciation goes to my parents and friend for their contributions and
support in all areas of life.
Finally, my profound gratitude and appreciation goes to all my course and
school mates, and who so ever help me in one way or the other to ensure the success
of this work. I thank you all, and the best wish to everyone.
v
ABSTRACT
Electronic commerce (E-commerce) has been very popular in recent years.
The society of the world is becoming to the information age; and the e-commerce
become to the new growth spot in the new internet industry. However, because the
impersonal nature of online-trading and untruth information, it makes the
e-commerce more difficult to build a trust between sellers and buyers than in
traditional market. One of the e-commerce development restrict is the lacking
trading-credit. The components of this thesis are including marketing theories, risk
theories, morality theories and credit evaluation system to research the basic notion
of the credit risk, the effect of credit risk in e-commerce on online-trading, and credit
evaluation system. Then, creating a credit evaluate model to analyze the advantages
and compare with TaoBao’s credit system to make sure the model has more effective
credit evaluation information for the TaoBao company.
vi
ABSTRAK
Electronic perdagangan (e-perdagangan) sangat popular pada masa sekarang.
Masyarakat dunia menjadi ke era maklumat dan e-commerce menjadi ke titik
pertumbuhan baru di industri internet yang baru. Namun, kerana sifat impersonal
online-trading dan kepalsuan maklumat, itu membuat e-commerce lebih sukar untuk
membina kepercayaan antara penjual dan pembeli dari pasar tradisional. Salah satu
perkembangan e-commerce adalah kurang mengehadkan perdagangan kredit.
Komponen-komponen daripada tesis ini meliputi teori-teori pemasaran, teori risiko,
moral teori dan sistem evaluasi kredit untuk kajian pengertian dasar dari risiko kredit,
kesan daripada risiko kredit melalui e-commerce di talian-perniagaan, dan sistem
evaluasi risiko kredit. Jika tidak, mencipta model menilai kredit menganalisis
keuntungan dan membandingkan dengan yang benar-benar ada sistem kredit untuk
memastikan model mempunyai lebih berkesan maklumat evaluasi kredit untuk
e-commerce.
vii
TABLE OF CONTENTS
CHAPTER
1
TITLE
PAGE
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENT
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
xii
LIST OF FIGURES
xiv
LIST OF APPENDICES
xvi
INTRODUCTION
1.1
Introduction
1
1.2
Background of Problem
2
1.3
Statement of Problem
4
1.4
Objectives
5
1.5
Scopes
6
1.6
Summary
7
viii
2
LITERATURE REVIEW
2.1
Introduction
8
2.2
E-commerce credit
9
2.2.1 Credit
10
2.2.2 E-commerce
11
2.2.3 E-commerce Credit
15
2.3
E-commerce Credit Risk
17
2.4
Online-trading in E-commerce
18
2.4.1 The Characteristics of Online-trading
18
2.4.2 Online-consuming Model
21
C2C Credit Risk Analysis
24
2.5.1 C2C System Structure
25
2.5.2 C2C Characteristics
27
2.5.3 The Origin of C2C Credit Risk
29
The Construction of C2C Credit Evaluation System
32
2.6.1 The Analysis of C2C credit evaluation system
33
2.6.2 Case Study of TaoBao and E-bay
34
2.6.3 The Lacks of Current C2C Credit Evaluation
38
2.5
2.6
System
2.7
3
Summary
39
RESEARCH METHDOLOGY
3.1
Introduction
40
3.2
Project Methodology
41
3.2.1 Feasibility and Planning Phase
46
3.2.2 Requirement Analysis Phase
48
3.2.3 System Design Phase
49
3.2.4 System Build Phase
50
ix
3.3
3.4
4
3.2.5 System Testing and Evaluation Phase
51
Hardware and Software Requirements
52
3.3.1 Hardware Requirements
52
3.3.2 Software Requirements
53
Project Plan
55
DATA ANALYSIS
4.1
Introduction
56
4.2
Current system of TaoBao Company
57
4.3
Problem of TaoBao’s Current System
65
4.4
TaoBao
Credit
Evaluation
System
Related
66
Questionnaire
4.4.1 Section A: Personal Information Questions
67
4.4.2 Section B: Optional Questions of TaoBao
68
Credit System
4.4.3 Section
C:
Optional
Questions
of
71
4.4.4 Section D: Questions of Credit Evaluation of
73
Authentication Part of TaoBao System
TaoBao System
4.4.5 Conclusion
of
TaoBao
Credit
System
81
Questionnaire
5
4.5
Proposed Solution
82
4.6
Summary
84
SYSTEM DESIGN
5.1
Introduction
85
5.2
Multi-Authentication
86
5.3
Buyer Authentication
87
5.4
Credit Evaluation Method
88
x
5.5
6
88
5.4.2 Evaluation Indicator
94
5.4.3 Evaluation Model design
105
Summary
116
SYSTEM DEVELOPMENT AND TESTING
6.1
Introduction
117
6.2
Database Creation
118
6.3
Authentication Module
112
6.4
Credit Evaluation Module
124
6.4.1 Original Credit Module
125
6.4.2 Dynamic Credit Module
126
User Satisfaction Test
129
6.5.1 New User’s Credit Situation Test
130
6.5.2 User’s Dynamic Credit Evaluation Test
131
6.5.3 System Integrative Test
132
Summary
133
6.5
6.6
7
5.4.1 Fuzzy Method
DISCUSSION AND CONCLUSION
7.1
Introduction
134
7.2
Overall Achievement
135
7.3
Outcome
137
7.4
Recommendation and Future Research
139
7.5
Chapter Summary
140
141
REFERENCE
APPENDICE
A
146
xi
LIST OF TABLES
TABLE NO.
TITLE
PAGE
2.1
The comparing of E-bay and TaoBao credit system
36
3.1
Advantages and disadvantages for Waterfall Model Spiral
42
Model Prototype model
4.1
The Frequency of User
68
4.2
General Comment of TaoBao Current Credit System
69
4.3
The General Problems Analysis
69
4.4
User’s Behavior Check
70
4.5
The General Opinion of TaoBao’s Authentication Method
71
4.6
The Question of Buyer Authentication
72
4.7
Question of New Member’s Credit Situation
72
4.8
Question of The Current Credit Evaluation System
73
4.9
Data Collection of Importance of Turnover
74
4.10
Data Collection of Importance of User’s Credit Rank
75
4.11
Question of Seller’s Evaluation Indicators Confirming
76
4.12
Question of Buyer’s Evaluation Indicators Confirming
77
4.13
Question of The Priority of Age Level
78
4.14
Question of The Priority of Education Level
78
4.15
Question of The Priority of Marital Status
79
xii
4.16
Question of Gender Confirm
80
4.17
Question of Salary level Confirm
81
5.1
The Scoring Function of Initial Credit Evaluation
97
5.2
The Scoring Function of Seller Dynamic Credit Evaluation
100
5.3
The Scoring Function of Buyer Dynamic Credit Evaluation
102
5.4
Initial Credit Evaluation Indicators Weight
105
5.5
Dynamic Credit Evaluation Indicators Weight for Seller
106
5.6
Dynamic Credit Evaluation Indicators Weight for Buyer
107
6.1
User’s Dynamic Credit Evaluation Test
131
6.2
System Integrative Test
132
xiii
LIST OF FIGURES
FIGURE NO.
TITLE
1.1
Distribution of the reasons of having not attempted the
PAGE
4
network Shopping
2.1
Online-consuming model
21
2.2
E-commerce system constructions
25
2.3
The general transaction process of e-commerce
35
3.1
Project methodology
45
4.1
TaoBao online-trading system
58
4.2
Seller’s credit ranks
62
4.3
Buyer’s credit ranks
62
4.4
TaoBao credit evaluation system
63
4.5
Age Distribution
67
4.6
The Distribution of Role
68
4.7
Question Result of Career Priority Confirm
80
4.8
The Process of Improved System
84
5.1
Fuzzy Credit Evaluation Framework
91
5.2
Credit Evaluation Indicators Structure
95
6.1
Database of Bank System
119
6.2
Database of User’s Initial Credit
120
xiv
6.3
Database of User’s Initial Credit
120
6.4
Database of Buyer’s Dynamic Credit
121
6.5
Database of Seller’s Dynamic Credit
122
6.6
Buyer’s Interface of Before Authentication
123
6.7
Authentication Interface
124
6.8
Granting Initial Credit Interface
125
6.9
Buyer’s Interface of After Authentication
126
6.10
Buyer Evaluation Interface
127
6.11
Interface of Seller’s Credit
127
6.12
Seller Evaluation Interface
128
6.13
Interface of Buyer’s Credit
129
6.14
Credit Situation of New User Test
130
xv
LIST OF APPENDICES
APPENDICES
A
TITLE
QUESTIONNAIRE
PAGE
147
1
CHAPTER 1
INTRODUCTION
1.1
Introduction
E-commerce means a serious of business and commerce that use modern
electronic information technology, especially use internet technology. It is the new
trading model that grown up in western world in ninetieth of twenty century. The
feature of e-commerce is purposing online-trading base on internet environment. In
the developing global economy, e-commerce has become to a necessary element of
business strategy and a strong catalyst for economic development; the integration of
information and communications technology (ICT) in business has revolutionized
relationships within organizations and those between and among organizations and
individuals (Chen Zhihao, 2003). Specifically, the use of ICT in business has
enhanced productivity, encouraged greater customer participation, and enabled mass
customization, besides reducing costs (Zhen Wenhong, 2000). Following the
lawmaking and consummation of e-commerce, and the development of safe-key of
internet, can build a good business environment of e-commerce, and also can create a
2
“level playing field” for internet enterprise of any size (Wei Mingxia, 2001).
However, throughout the practice in so many years, e-commerce has taken the value
that should be expected. The mostly point is the main stakeholders in e-commerce
cannot trust each other.
1.2
Background of Problem
According to the estimates of International Data Corp (IDC) (2004) we can
see the value of e-commerce has increased very fast, in 2000 the value of global
e-commerce is US$350.38 billion; and then was projected to climb to as high as
US$3.14 trillion by 2004. IDC also predicts an increase in Asia’s percentage share in
worldwide e-commerce revenue from 5% in 2000 to 10% in 2004. Asia-Pacific
e-commerce revenues are projected to increase from $76.8 billion at year-end of
2001 to $338.5 billion by the end of 2004 (Zorayda Ruth Andam, 2003).
However, recently, the development of e-commerce faces to some threats,
those threats result in the e-commerce developed very slowly, is not because absence
the model of payoff, but it is the loss of the trust. The e-commerce credit risk is more
and more important; credit risk is becoming to the key point of baffling consumer’s
online-trading (JvarePnaaetal., 2000; Baetal., 2000). Some enterprises that spent a
tremendous financing in built an e-commerce establishment have try to find the way
to build the credit and force the online-trading for receive the cost, make payoff and
advance the development of the enterprise. But the results are not satisfaction,
because the e-commerce activities are through the space-time, the credit risk is more
3
sense to the organizations’ and individual’s online-activity and performance (Tadesli,
1999).
Dummy market (or electronic market) is the market that basing internet and
relative to the traditional market; it is the typical “stranger” social. Its universality,
changing records, complex entities and dummy are make its credit problem is more
important than traditional market; the trader easier to trend to lie each other, because
this losing credit action has low risk, but the income is satisfactory. Following the
cheat is more popular in the e-commerce, the losing credit; the credit conjuncture and
the problem of deteriorating the situation of trust are more and more graveness
(Zorayda Ruth Andam, 2003).
According to the eighteenth research report (2006) from China Network
Information Centre (CNNIC), China has 300 millions internet consumers, as a
quarter of the netters. Compare with 2005, there were raising 50% of the internet
consumers. But there were 71.1% netters who do not have the experience of
online-trading do not trust the online-trading. From the “Hotspot research report” of
CNNIC (2004.11), it surveyed the netters’ online-trading. In the multi-select question
of “Distribution of the reasons of having not attempted the network shopping”, the
main results of the netters who do not have the experience of online-trading are:
don’t trust the web, worry about the quality of commodity and the after service,
oppugn the safety, and worry the way of payment, it show in the Figure.1.1 (CNNIC,
2006). However, some netters had already to try online-shopping, but they gave up
because the too fussy process to balance and have to fill too much information
(CNNIC, 2006).
4
70
60
50
40
30
20
10
0
others
no suitable commodity
do not want to buy, no needs
unabundance commodity
unsatisfation price
commodity information not complete
do not know how to buy
worry the commodity send
worry the way to pay
complex
after service
safety
worry quality
do not trust
Figure 1.1 Distribution of the reasons of having not attempted the network shopping
(CNNIC, 2006)
The main point of the problem of online-trading is lacking an effective
e-commerce credit system, especially the personal credit system, to control the
online-trading credit risk. So, it is very urgent to investigate the e-commerce credit
evaluating system, keeping advance the current credit situation.
5
1.3
Statement of Problem
Credit system exits in the West developed countries has more than 160 years
history, the importance of credit value in depth people’s lives. Herbig, Meliwicz
(1994) had discussed the concept of credit model in 1994s. Tadelis, in 2001, said that
credit is the exchangeable assets and also is promoting.
But after so many years of development, credit issues remain a major
problem that affecting the development of e-commerce. The e-commerce credit risk
is also not very clear, people do not know where it happens and how to avoid;
because the current credit evaluating system is still too vague to analysis the
stakeholders who do the transaction in e-commerce credit.
Due to the above problem, it is urgency to have a clear credit evaluating
system that can support a good credit environment of online-trading, furthermore, to
advance the development of e-commerce.
1.4
Objectives
The objective of this project is providing useful information for reducing the
online-trading credit risk, and to help the people or organization to enjoy the
online-trading; consequently advance the development of e-commerce.
6
To achieve the above objectives, there sub-objectives should be formulated
are:
1.5
i.
To analyze the e-commerce credit risk.
ii.
To evaluate the C2C credit evaluation system.
iii.
To design system for reduce credit risk of TaoBao system.
Scope
This project represents a systematic analysis approach for evaluate the credit
risk of online-trading. It is the oriented can help consumer sensing and avoiding the
credit risk of online-trading, for advance the development of e-commerce in some
aspects. This research covers:
i.
E-commerce credit risks mechanism.
ii.
C2C operating system and credit evaluating system.
iii.
Case study of TaoBao credit evaluating system.
iv.
This credit evaluating system is developed based on web-sites design
application and windows environment.
7
1.6
Summary
This chapter focused to introduce the current situation of development of
e-commerce and online-trading and important of credit risk in e-commerce. Firstly, it
defined the connotation and the characteristic of credit. Then, introduced the nature
of e-commerce, and analyzed the current situation of e-commerce. Laid out the credit
risk has been the very important element in the e-commerce. In addition, it
introduced the online-trading, which is the main action in e-commerce, and analyzed
the mechanism of online-trading. At the end, get the conclusion that is when the
credit risk in online-trading can be improve, it will advance the e-commerce move on.
Otherwise, this is the main objective of this project.
8
CHAPTER 2
LITERATURE REVIEW
2.1
Introduction
C2C as a major transaction form of e-commerce has been growing rapidly. In
2006, the China Network Information Center (CNNIC) at the first time did the
investigation of China C2C online-trading and issued the “China 2006 C2C
online-trading report” (CNNIC, 2006). The report show that C2C online-trading is
more and more important in the city netizens, the number and frequency of C2C
behavior are very active. As of march 2006, in Beijing, Shanghai and Guangzhou,
there have online-trading consumer s 2million people, the rate in internet users
reached 16.2% (CNNIC, 2006).
C2C online-trading problems caused by a variety of causes are many, but in
general is the lack of an effective e-commerce credit system, especially the personal
credit system, cannot get an effectively of the credit risk control.
9
In this chapter, first introduce the concept of e-commerce credit risk, it
includes that presenting the class of e-commerce and analyzing the credit risk
mechanism in e-commerce. Then, continue to introduce the characteristics of
online-trading, analyzing the online consumer’s requirements, mentality and
activities, according these theories summarize the e-commerce credit risk how to
impact the online consumers.
In addition, focus on the C2C e-commerce, and introduce its credit evaluating
system; use the case study of E-bay and TaoBao to find out the general problems of
current e-commerce credit evaluating system.
2.2
E-commerce credit
Since the emergence of network on the 60s of 20th century, the theory of
e-commerce, with the developing of dummy market activities, is getting enrichment
content, constant innovation, deepening of the concept, growing areas of research;
e-commerce credit risk is the product of e-commerce theory and practice keep to
innovation (Kolloek, 1999).
10
2.2.1
Credit
“Credit” is the static state attribute of people or organization; otherwise, it is
the degree of faithful and recognized by other people, almost defined in the
economics (Jin Xiaohong, 2007). In the economics area, the concept of credit can
define in a broad and narrow. In the narrow sense, the credit is the activities of debit
and prepay between the different owners of goods and money, reflecting the debt
relationship of owner and debtor. In the broad sense, credit is the unity of the
subjective faith and objective solvent. The development of credit, lead credit
becomes a means to a way of trading, occurring the credit transactions; otherwise, it
is the concept of narrow sense which defined above. It actually is the “credit”
concept in the abroad area using and extending in the economy area.
There the natures of the credit characteristics are: firstly, it is the moral
character to fulfilling their obligations and commitment in the subjective; and in the
objective, it has the ability to delivering or paying. Credit is the diathesis and also is
the capability, as Muller said: “to judge one’s credit, look for how many solvencies
he has (Fukuyama,F, 1995).” As a result, the credit quality should include two levels
credit as character credit and assets credit. Secondly, in the essential, credit reflects
the relationship between right and debt. Thirdly, reflecting to the expectations and
requirements with the future benefit, as Macleod said: “every expectation is credit,
merchandise credit to be sold in exchange for monetary credit.” Therefore, credit can
be defined that one of the psychology base on trust, and represent a lot of forms, the
final goal is to obtain “the payment value”. (Jiang Lingming, 2003)
11
2.2.2
E-commerce
The e-commerce and e-business are not the same in some aspects. In
e-commerce, information and communications technology (ICT) is used to
inter-business (B2B) and in business-to-consumer (B2C) transactions. In e-business,
on the other hand, ICT is used to enhance one’s business. It includes any process that
a business organization (either a for-profit, governmental or non-profit entity)
conducts over a computer-mediated network. A more comprehensive definition of
e-business is: “The transformation of an organization’s processes to deliver
additional customer value through the application of technologies, philosophies and
computing paradigm of the new economy (Zorayda Ruth Andam, 2003).” From the
view of philology e-business can be defined to be a general commercial activities or
industry behavior; it includes the specific action and operations. E-commerce
normally can be defined to the actions of trade and purchase in internet; it is usually
implied that the value of property exchange. A more systemic definition is:
E-commerce is the use of electronic communications and digital information
processing technology in business transactions to create, transform, and redefine
relationships for value creation between or among organizations, and between
organizations and individuals (Emmanuel Lallana, Rudy Quimbo, Zorayda Ruth
Andam, 2000).
According to a scholars’ definition (Resnick.P, and Zeckhauser,R., 2001), we
can recognize that e-commerce is the exchange value between stakeholder base on
the network and electronic information technology.
12
2.2.2.1 The Classes of E-commerce
E-commerce is not only built on the basis of business activities, but also
based on their creation maintenance and improvement on the existing and potential
relationships. There major different types of e-commerce are: business to business
(B2B), business to consumer (B2C), business to government (B2G), consumer to
consumer (C2C), system to system (S2S), consumer to government (C2G) and
mobile commerce (m-commerce) (Zorayda Ruth Andam, 2003).
i.
B2B e-commerce
B2B e-commerce is simply defined as e-commerce between companies. This
is the type of e-commerce that deals with relationships between and among
businesses. About 80% of e-commerce is of this type, and most experts predict that
B2B ecommerce will continue to grow faster than the B2C segment (Zorayda Ruth
Andam, 2003).
ii.
B2C e-commerce
B2C is the model of e-commerce between business and consumer; it is the
second largest and the earliest form of e-commerce. Its origins can be traced to
online retailing (or e-tailing) (Zorayda Ruth Andam, 2003). It involves customers
gathering information; purchasing physical goods (i.e., tangibles such as books or
consumer products) or information goods (or goods of electronic material or
digitized content, such as software, or e-books); and, for information goods,
receiving products over an electronic network (Goldman Sachs Investment Research,
November 1999). The common B2C applications are in the areas of purchasing
13
products and information, and personal finance management, which pertain to the
management of personal investments and finances with the use of online banking
tools (e.g., Quicken) (Gefen. D., 2000).
iii.
B2G e-commerce
B2G is generally defined as commerce between companies and the public
sector. It refers to the use of the Internet for public procurement, licensing procedures,
and other government-related operations. This kind of e-commerce has two features:
first, the public sector assumes a pilot/leading role in establishing e-commerce; and
second, it is assumed that the public sector has the greatest need for making its
procurement system more effective (TA Project, 2002).
iv.
C2C e-commerce
Consumer-to-consumer e-commerce or C2C is simply commerce between
private individuals or consumers (Zorayda Ruth Andam, 2003). This type of
e-commerce is characterized by the growth of electronic marketplaces and online
auctions, particularly in vertical industries where firms/businesses can bid for what
they want from among multiple suppliers (Traderinasia.com, 2002). It perhaps has
the greatest potential for developing new markets.
v.
M-commerce
M-commerce (mobile commerce) is the buying and selling of goods and
services through wireless technology-i.e., handheld devices such as cellular
telephones and personal digital assistants (PDAs). Japan is seen as a global leader in
14
m-commerce (Zorayda Ruth Andam, 2003).
vi.
S2S e-commerce
S2S is during the organization of e-commerce, such as between different
departments within the organization e-commerce through an enterprise network
(internet) and enterprises extranet, such as the e-governance in the same department
of government. This is the current e-commerce has been widely carried out (Zorayda
Ruth Andam, 2003).
vii.
C2G e-commerce
C2G is the e-commerce between government and individuals. For example,
the providing of social welfare funds or personal tax returns. This model has not
really formed. However, as the development of business-to-consumer and
business-government e-commerce, governments would be better to implement the
electronic services for individuals (Zorayda Ruth Andam, 2003).
The above definition of e-commerce almost based the internet, but we can
define it in more extensive coverage, including other technology such as radio, phone
shopping, private network (such as the banking system) and EDI (electronic data
interchange), etc.
15
2.2.3
E-commerce Credit
In the e-commerce researches, credit has been to the key point of stimulating
the internet purchase (Kolloek, 1999).
In the actual dealing, all the relationship need the credit factor, particularly
the relationship that occurring in the uncertain environment of e-commerce (Fung
and Lee, 1999). Thus establishing the credit relationship between the stakeholders is
the key point of e-commerce development (Whitmeyer, Joseph M., 2000). In the
relationship between consumer and marketers, the most important element also is
credit (Stewart, et al.1999), credit has impacted to consumers, especially to online
consumers (Swan, J.E. and Nolan, J.J., 1985). Therefore, credit is the key element of
e-commerce and can be a strategic mechanism that advancing the relationship
between consumer to marketers (Kotkin,J., 1998).
E-commerce credit can be defined that the consumers believe that a particular
transaction in some way with their own subjective expected probability of consistent
(Stewart, 1999). In the e-commerce transactions, the definition of credit includes a
number of distinct but inseparable factors: first, it includes a marketer’s specific
traditional credit concept. Second, it is tightly around the reliability and honesty of
transacting media (Fombrun,C. and Shanley, M., 1990). This is mean when consumer
estimate a particular transaction that will according their own expect to be occur,
consumer will consider the features of marketer and transaction media.
Compared with the traditional credit, interpersonal trust, e-commerce credit
first has the impersonal feature (Li Huibing, Yang Dongxue, 2000). Impersonal credit
16
can be occurred where the relationship between non-personalized, the relationship
between all stakeholders is “episodic”, in other words, in short-term, the information
is asymmetric and uncertain, and from there are including a number of important
inter-authorized between consumers and marketers (Dorre, R, 1987). The features of
dummy market and e-commerce decide that the e-commerce non-personalized credit
is necessary, Doney and Cannon putted forward the cognitive or perceptual process,
and format credit through these process (Danny and Cannon, 1997).
Danny and Cannon (1997) believe that credit can be completed by estimating
an entity’s ability of achieving tasks. According to this view, for consumer in order to
establish the credit of a specific transaction, they will use a competent procedure to
ensure that marketer and media can complete transaction in the specific ways, and
consistent with its expectation. In addition, credit also involves another procedure
which the consumer can have a subjective with specific situation base on the limit
information. There is the logic relationship between consumer and specific
transaction, if consumer estimate the personal private and safe is consistent with their
expectation, they will establish the credit. In short, when the collection of consumer’s
information and the subsequent visit, use and disclosure consistent with its
expectations, and consumer believe that their information on the way of transmission
and storage will not be browsed, stored and manipulated by inappropriate
organizations, the consumer will establish the credit.
17
2.3
E-Commerce Credit Risk
Risk can be defined that “the threat or probability that an action or event will
adversely or beneficially affect an organization's ability to achieve its objectives”
(Liu Youcai 1996). Generally, when we speak risk, it is only definite of those event
which are highly uncertain or hazardous (Liao Chenglin, 2003). Symbolically, risk
can be as: (Shi Liangping, 2003).
Risk = Uncertainty * Potential loss/gain
In the e-commerce credit, this uncertainty can be known that the variable
moral of traders, and the uncertainty of media platform.
E-commerce credit risk refers to the uncertain of credit in e-commerce or
dummy market (Zeng Yong, 2004). According to the concept of e-commerce,
e-commerce credit risk can be defined that the stakeholders of dummy market or
e-commerce process comply with the market contract (implicit or explicit) the degree
of uncertainty; it because of the uncertainty of stakeholders’ credit philosophy
(ethical, cultural or moral) and credit capacity in dummy market (Bian Xiaohong,
2006).
18
2.4
Online-trading in E-commerce
The basic features of e-commerce decide the e-commerce should be different
to the traditional market. It makes the online-trading more complex.
2.4.1
The Characteristics of Online-trading
Consumers have the special activities in the dummy market; in the dummy
market, the online-trading activities can consist of the online-consumer’s consuming
demands and consuming trend, and from those to bring out the activities.
2.4.1.1 Online-consumer’s Requirements
Online-trading is the new way of consume, it different to the traditional trade.
In this way, the consumer is facing to a virtual platform; it can be exempted from the
restrictions of time and space. The “Maslow's hierarchy of needs” can explain almost
online-trading activities, but the virtual society, after all, is different to the traditional
society; so it must have some new features (Wen Xing, 2007).
19
Firstly, all the demand hierarchy of online-trading has the close relationship
(Wen Xing, 2007). For example, in the same purchase orders, the consumer can buy
the most common and the expensive jewelry to meet the physiological needs and the
demand for respect together. So in the online-trading, company can supply wider
area product, and show the different hierarchy products in the same time.
Secondly, in the online-trading, consumer’s demands not only represent to the
normal demand, but more as the potential demand, because almost the online
consumer are the young people, they have advanced awareness and responsive to
new things, speeder receivable (Wen Xing, 2007). So in the online-trading, company
should through constant innovation to attract consumer, at the same time, use the
own strengths, stimulate the new demands of online consumers.
2.4.1.2 Online-consumer’s Mentality
Consumer is always been the hot spot in the online-trading (Wen Xing, 2007).
Catching the characteristic of online consumers, it is the very important to the
enterprise’s decisions and performance. One enterprise want to attract customers, get
constant competition, they need to analyze the online consumer, and understand their
characteristic, then make the decision.
20
2.4.1.3 Online-consumer’s Activities
The network environment provides broader selection space. In this space,
consumers can have a wider selection, base on the huge network and e-commerce
system; but too much selections to bring a negative impact on the consumer, it makes
the consumers do not know how to make the decisions (Wen Xing, 2007). Therefore,
almost websites often set up the products or services columns; and there have been
some comparison sites, analysis model and evaluate software to guide consumer
behavior; thus, consumers can make a purchase more sensibly. In addition, the
consumer will easy to compare the price, and then make a very rational purchase
decision. At the last, online consumers are according to their own needs to search the
suit products, but not longer passively accept the vendors to provide products or
services (Wen Xing, 2007). However, if they cannot find the product they want,
consumers will to show the desire of the products base on the network. Those should
affect the production and business processes.
2.4.2
Online-consuming Model
Consumers’ online purchase process consists by a series of closely
interrelated activities. The online-trading can be representing in the Figure.2.1, when
consider e-commerce credit risk. It includes the five stages are confirm needs, gather
information, program evaluation, purchase decision-making and post-purchase
evaluation (Zhang Huifang, 2002).
21
Figure 2.1 Online-consuming model (Zhang Huifang, 2002)
From these five stages can be seen that the purchase process began before the
actual purchase occur, and the impact will be sustained for a long time after the
purchase (Zhang Huifang, 2002).
i.
Confirming needs
The starting of purchase process is an awareness of needs. Consumers are
aware that the different between the actual situation and the expectation to generate
demands. The demand can be stimulated by internal factors, such as hunger and thirst,
and can also be stimulated by external factors, such as seeing someone bought a new
computer.
22
ii.
Gathering information
A consumer who had been aroused a demand will seek more information. So,
gathering information becomes to the second part into the process of online-trading.
The role of this sector is to collect data for prepare the next step, program evaluation.
There are two main aspects of gathering information: the internal channels
and external channels. At the first, consumers search the product information in their
memory. If there is not enough information for the decision-making, consumer would
go to the external environment to look for information. In the online-trading, the
gathering information is almost basing to the internet.
iii.
Program evaluation
In this process, firstly, online consumers should evaluate the e-commerce
credit risk. Secondly, consumers should consider the own ability of pay. Consumers
have to compare analysis and understand the commodities’ characteristics in order to
balance the demands and their purchasing ability. Normally, consumers’ evaluation
almost considers the functionality, reliability, performance, style, and price and
after-sales service. However, the information that consumers have gotten can be too
large amount to analyze. Thus, a variety of network tools, comparison model came to
solute those problems. Because consumers cannot contact the product with the
physical in online-trading, so consumer evaluates the products relying on the
description of goods. Therefore, the network marketer should describe their product
by text and pictures, for facilitate customers to do an evaluation. However, network
marketer cannot do the product of false propaganda, or they might lose their
customers.
23
iv.
Purchase decision-making
Compared with the traditional way of buying, there are two characteristics of
online consumers’ decision-making: firstly, the decision-making of online-trading is
faster than the traditional. Secondly, online consumers have to face higher credit risk.
An enterprise to deal in a virtual environment is not easy; there are three
conditions should be confirmed when online consumer buying a commodity: first of
all, firms should have a sense of trust. Secondly, there is a sense of security for the
payment. Thirdly, the products have a favorable impression.
v.
Post-purchase evaluation
When consumers bought goods, they are often inspecting the options in order
to determine the accuracy of this purchase decision. If a product’s price, quality and
service are the same to the expected, consumer will feel the satisfaction, otherwise,
consumers will be generated a version mind. Post-evaluation are often determine the
consumer purchasing trends in the future.
This model mainly discusses the model of e-commerce credit risk how to
impact the online-trading process. However, this model is somewhat vague and
rough to interpret the e-commerce credit risk impact of online-trading.
24
2.5
C2C Credit Risk Analysis
According to the analyzed of e-commerce credit risk characteristic, and the
introduction of behavior of online consuming. After that, focused on analyze the
impact of C2C e-commerce credit risk.
2.5.1
C2C System Structure
There is no authoritative conclusion of which parts should be included in a
complete e-commerce system. On the whole, e-commerce system is a collection of
multi-technology; it should include the four structures (Che Chun, 2005):
Payment
platform
Security
platform
Payment central server, mail server and payment gateway
Security authentication system
Network
platform
Internet
Figure 2.2 E-commerce system constructions (Che Chun, 2005)
Others
Games
News
Distant medical
e-business
Online pay tax
Online payment
Online application
Video, music play
Distant education
Safe WWW sites
platform
Safe E-mail
system
Managing platform
Application
25
Whatever network platform, security platform, payment platform, or
e-commerce applications, they must provide the necessary supporting services to
maintain the normal operation of e-commerce system.
In the C2C e-commerce system, it should be able to reflect the real-world
business activities to achieve the online transactions between the online traders; and
also providing a viable web-based management programs and trading methods to
make the dissemination of commodities more timely and widely.
C2C e-commerce system is based on the general e-commerce system,
according to the transaction characteristics of C2C e-commerce, forming its unique
e-commerce system architecture. Now, according to the typical C2C e-commerce
system, auction web-site analyzes the C2C e-commerce system.
There are some functions of C2C e-commerce auction system (Zeng Yong, 2004).
i.
The basic data support module.
C2C e-commerce system needs a powerful data supporting system for storing
the user’s personal information, transaction information,
information.
and commodity
26
ii.
Service management module.
The main aim of this module is achieving the updating information and user’s
store management of web-site with regularly.
iii.
Credit intermediaries.
More and more C2C e-commerce sites on the internet played a third-party
intermediary role. It uses the own credit mechanism to achieve the management of
online trader’s credit; through the evaluation of online trader’s credit, so that
providing the credit reference of trader’s records of transaction, thus to ensure the
transactions.
2.5.2
C2C Characteristics
C2C e-commerce system not only has the characteristics of general
e-commerce, but also has a number of its own notable features (ZhuXiaozhong,
ZhangZongyi, GenHuadan, 2004).
i.
Wide range and numerous participants.
Almost transaction stakeholders in C2C e-commerce are the individual and
small business. Because the development of internet, online-trading has become to a
fashionable, therefore attracting more user’s to participating.
27
ii.
Numerous products.
Online traders just need through a simple registration as long as you can
make shopping online, no need pay for the store. The low operating costs attract a
wide range business to set their shop online, and providing a varied options.
iii.
Flexible transaction.
Because the number of money of transaction general too small, so it can use a
varying ways of transactions. With the growing awareness of credit and the
improving of credit mechanism, the current C2C e-commerce web-site to be the
third-party intermediary roles, to provide payment guarantees to ensure the
transaction be success.
With the growing number of online-trading users, the role of C2C
e-commerce transaction platform to become more and more important. How better to
evaluating the online trader’s credit, solving the credit problem of transaction
stakeholders on C2C e-commerce transaction platform, are become to the main point
of development of C2C e-commerce.
2.5.3
The Origin of C2C Credit Risk
Because the characteristics of C2C e-commerce, its credit risk is even more
prominent. There are three credit risk arising (Wen Xing, 2007):
28
From the buyer’s credit risk.
i.
For the individual consumers, may exist using a credit card to pay and
malicious overdraft on the network, or using the forged credit card to fraudulent the
seller’s goods. For a group, they possibility delay the payment, so the seller has to
assume the risk.
From the seller’s credit risk
ii.
Because the virtual of network, the buyer may be cannot pick up the real
samples. In the process of putting the commodity into a picture, some basic
information will be lost, so buyer cannot get the complete information from the
picture and descriptions. These will take the product identification risk, which can
impact the performance, quality and other aspects of product, to the consumers.
iii.
The denying existence in both of buyer and seller
Because the releasing of transaction information and the delivering of goods
have a certain lag, it should be different with the real situation.
In short, C2C e-commerce credit risk problem has its own internal reasons, in
other hand, because the network and there are many imperfections problem of C2C
e-commerce system. In addition, C2C e-commerce credit risk problem has the
problem from external, namely, a whole credit system of society is not perfect by
lack of credit. If want to control the C2C e-commerce credit risk, have to develop a
good credit risk control strategies, establish a sound credit evaluating system, that
need the joint efforts of the whole society, especially the help of government.
29
For evaluating people’s basic information and confirming people’s initial
credit, These seven indicators are easy to be confirm when user doing the online
register; in addition, these seven indicators have strongly close with user’s credit.
There are age, gender, marital status, education degree, occupation, salary, and bank
deposits. Zhao Xiaodong and Zhen Tao (2003) described the indicators as following:

Age: Age is a more important indicator in the credit evaluation. In the
general bank individual credit evaluation indicator, they all believe that the
older person has higher credit level than youth.

Gender: women’s credit generally is believed that higher than men’s.

Marital status: married better than unmarried, with children is better than no
children.

Educational level: A person who has the higher education level has better
quality.

Occupation: A person who has a relatively stable degree of career has higher
credibility.

Salary: It is a measure of the credibility of the user’s economic indicators;
the higher income has greater credibility.

Bank deposit: More bank deposits are more worthy of trust.
For evaluating people’s behaviors and according to the report of “China’s
C2C online shopping survey” (CNNIC, 2009) shows that when the buyer choose the
seller, 75% of buyers are most concerned about the price of goods or cost-effective,
more than 55% of buyers are concerned about the quality of goods, and others
important aspects that buyers concerned are the delivery time and service attitude.
These results are similar to the research of I did in China by questionnaire. In
addition, in order to prevent the problem of users use cheap goods to carry out the
high credit rank; and to make a more equitable trading environment, the turnover
30
indicator will be added. Furthermore, to prevent the malicious evaluation and to
make buyer’s evaluation more reliable, the buyers’ evaluation can evaluate higher
scores by the higher credit rank. Therefore, this thesis selected six indicators as a
dynamic credit evaluation indicator; there are: time, quality, service, price, buyer
credit, and turnover. The description of each indicator as follows:

Time: in the promised time, the faster can be more satisfied;

Quality: the quality of goods with the commitment of quality, the closer the
better;

Price: compare the price with other same goods, the cheaper the better;

Service: better service has more satisfied;

Credit rating: the higher credit rank user to make the evaluation more reliable;

Turnover: more turnover transaction, the proceeds of the credit score is higher;
2.6
The Construction of C2C Credit Evaluation System
For the e-commerce credit risk, David Friedman, from United States, have
said that the dispute of online-trading has two roots: the first is the time different, the
second is anonymous system (ShaoBingjia, LiRu, 2006).
To solve the time different, the best way is providing the third-party
guarantees and paying. In China, the earliest third-party payment platform is
“AnFuTong”, in 2000, which belongs E-bay. After e-Bay, TaoBao, YiPai and other
auction sites have continue to designed “ZhiFubao” and “YiPaitong” to be the new
31
payment platforms. Although, there are many detail problem should be improve, but
showing the efforts of online-trading sites in online payment.
For the anonymous issue, David (2004) said, allowing third parties to confirm
the identity of both the traders, but the traders also remain anonymous (David
Friedman, 2004). After E-bay created a real name certification program, almost
online-trading sites have been chosen this procedure. Another mechanism to offset
the anonymity risk is credit evaluating system. If web site was not supported by a
credit system, online-trading cannot development any longer. Therefore, the credit
evaluating system that evaluates the transaction of traders is the key word to solve
problem of online-trading.
At present, almost C2C e-commerce sites has their own credit evaluating
system, but generally the participants in C2C transaction are individuals or small
businesses, they are not visibility, lacking understand between the traders is the
problem to use the credit evaluation system.
In this chapter, will compare and analyze two typical sites, TaoBao and E-bay,
to point out the lacks of C2C e-commerce credit evaluation system.
32
2.6.1
The Analysis of C2C Credit Evaluation System
Credit evaluation system is the tool that use in e-commerce to create and
transfer information. The goal is using the past transaction information to judge the
quality of products and services of sellers and the payment situation of buyers. It is
purpose to reduce the credit risk of transaction (ShaoBingjia, LiRu, 2006).
The basic principles of credit evaluation system are: after a user finishes a
transaction, the traders can evaluate some aspects of transaction, such as products
quality, the timely of sending goods and payment. Formatting feedback of credit
information is reflecting the user’s credit situation and to be the reference for other
users.
Credit evaluation system mainly has following aspects:
i.
To constraint the traders’ behavior; reduce transaction risk, especial credit risk;
improve the success rate of transaction; in some extent, reduce transaction costs.
ii.
To facilitate traders to understand and judge other’s credit situation, for improve
the success rate of online-trading.
iii.
To the both of traders, credit evaluating can reduce the transaction costs. For
example, it can reduce the costs of advertising that seller try to get the trust of
buyers.
33
2.6.2
Case Study of TaoBao and E-bay
In the C2C e-commerce sites, the general transaction process is following
Figure 2.3:
Figure 2.3 The general transaction process of e-commerce (LuHongyan, WeiXin,
2007)
According to this process, comparing these two web sites credit evaluation
system from different aspects (LuHongyan, WeiXin, 2007):
34
i.
Registering members.
Whatever the buyer or seller, they can register at the two platforms for free,
and also easy to operating.
ii.
The way of authenticating.
For the seller, E-bay is through mobile phones, fixed phones and ID card to
complete the authentication. TaoBao is through ID card and bank accounts to
complete the user identity authenticating. For the buyer, no need identity
authenticating for both of two platforms.
iii.
Payment security.
In the payment security area, both of those platforms are the third-party role
of payment tool.
iv.
Credit evaluation system (LuHongyan, WeiXin, 2007)
Table 2.1 The Comparing of E-bay and TaoBao credit system (LuHongyan,
WeiXin,2007)
E-bay
Credit
The
The evaluation of sellers All the evaluation of sellers
evaluation evaluating
system
rules
TaoBao
that from the buyer who that even from the buyer who
of has
been
passed
the did not pass the identity
35
buyer
seller
to identity authentication will authentication will get the
get the scores in seller’s scores
in
seller’s
credit
credit system, otherwise system
will not
Evaluation When buyer has
been When
buyer
has
effecting
auctioned a commodity, auctioned
time
he/she can evaluate and he/she can evaluate, but the
immediately have effect
a
been
commodity,
results will show after both of
the
traders
have
been
evaluate.
Credit
As long as the evaluation Getting the score when have
scoring
finish gets the scoring, been evaluated and finish the
system
deals get one evaluation. transaction
through
“positive” get 1 score, “ZhiFuBao”
payment
“neutral” on score and platform,
“negative” get -1 score
deals
get
one
evaluation. “positive” get 1
score, “neutral” on score and
“negative” get -1 score
Buyer’s
Buyer’s and seller’s credit Buyer’s and seller’s credit use
credit and use one credit system, do different credit system
seller’s
not separate
credit
The
In the second (or more) To the same traders, they can
transaction transaction between the evaluate and get score sixth
between
same traders, traders can time in one month, over
some
make a evaluation but transaction cannot get score
users
cannot get score
36
2.6.3
The Lacks of Current C2C Credit Evaluation System
According to the above compared the credit system of E-bay and TaoBao,
obviously, these two credit system ensure the security of online-trading in some
deeps. But also there are existing some lacks, especially in the e-commerce credit
evaluation, both of them have some places should improve:
i.
Lacking a effective verification of both seller and buyer.
ii.
Lacking of oneness in the rules of credit evaluation, web sites spent a lot of
budget to establishing and maintaining credit evaluation system.
iii.
Credit evaluation model is too simple.
iv.
Lacking timely evaluation
In this aspect, the lack of E-bay’s mechanism is more obvious. Firstly, buyers
do not need to be waiting the transaction successful, they can evaluate. This behavior
supplies a chance to give “falsity credit”. In addition, in the E-bay system, the
evaluator who evaluate first may be face the retaliatory negative feedback, so some
buyers who faced unsatisfactory service have to give up to evaluate. The results,
making the credit of sellers are distortion.
v.
Buyers and sellers credit
In the E-bay system, it does not separate the credit of buyers and sellers, but
the nature of credit of buyers and sellers are different, as a buyer, concerned about
37
the other sellers’ credit only, and for a seller, also only concerned about their
customers’ credit.
2.7
Summary
This chapter has described the literature review of e-commerce credit
evaluation system. According the e-commerce credit risk mechanism analyze to find
out the impact of credit risk of online consumers. Besides, it used the C2C credit
evaluation system to find the general problems of e-commerce online-trading. Case
study of E-bay and TaoBao are also introduced in this study.
38
CHAPTER 3
RESEARCH METHODOLOGY
3.1
Introduction
Methodology refers to everything that can be encapsulated for a series of
process, activities and tasks, for plan the way of doing things, examples of this are
software
development,
project
management
and
business
process
fields
(http://en.wikipedia.org/wiki/Methodology).
In this chapter, the research methodology of this project will be presented.
The methodology involves finding the mechanism of e-commerce credit risk,
investing components of online-trading, analyzing the lack of current system, at the
final, designing a system to support e-commerce credit evaluation system.
For achieving these steps, some tools that used are Microsoft word 2007,
Microsoft Project 2007, Macromedia Dreamweaver 8.0, StarUML and MySQL.
39
3.2 Project Methodology
The System Development Life Cycle (SDLC) is a type of methodology used
to describe the process for building information systems, intended to develop
information systems in a very deliberate, structured and methodical way, reiterating
each stage of the life cycle (DSDM Consortium, 1990). The typical system
development life cycle models include waterfall model, spiral model and prototype
model.
The waterfall model is a sequential software development process, in which
progress is seen as flowing steadily downwards through the phases of conception,
initiation, analysis, design, construction, testing and maintenance. It has its origins in
the manufacturing and construction industries; highly structured physical
environments in which after-the fact changes are prohibitively costly, if not
impossible (Wikipedia web).
The spiral model was defined by Barry Boehm in his 1988 article “A Spiral
Model of Software Development and Enhancement” (Wikipedia web). It is a
software
development
process
combining
elements
of
both
design
and
prototyping-in-stages, in an effort to combine advantages of top-down and bottom-up
concepts (Danephilip, 2008).
The Prototyping Model is a systems development method (SDM) in which a
prototype (an early approximation of a final system or product) is built, tested, and
then reworked as necessary until an acceptable prototype is finally achieved from
which the complete system or product can now be developed. This model works best
40
in scenarios where not all of the project requirements are known in detail ahead of
time. It is an iterative, trial-and-error process that takes place between the developers
and the users (Boehm B, 1988).
The advantages and disadvantages for those two methodologies are presented
in Table 3.1 (Wikipedia web)
Table 3.1: Advantages and disadvantages for Waterfall Model Spiral Model
Prototype model (Wikipedia web, 2006)
Methodology
Waterfall Model
Advantages
Disadvantages
1.
Clear project objectives.
2.
Stable
3.
4.
1.
project 2.
Time consuming
Never
backward
requirements.
(Traditional)
Progress of system is 3.
Little room for iteration
measurable.
Difficulty responding to
Strict
4.
sign-off
changes
requirements.
Spiral Model
1.
2.
3.
Avoidance of Risk is 1.
Highly
enhanced.
limiting re-usability
Strong
approval
and 2.
Applied differently for
documentation control.
each application
Implementation
Risk
priority
has 3.
over
functionality.
4.
customized
of
not
meeting
budget or schedule
4.
Possibility to end up
Additional Functionality
implemented
as
can be added at a later
Waterfall framework
the
41
date.
Prototype model 1.
Strong
between
Dialogue 1.
users
and
Missing
functionality
non-functional elements
functions
difficult to document
can
be
3.
Requirements
application
may cause application
Quick
not to be used as the full
of,
system was designed
incomplete,
but 4.
Incomplete or inadequate
functional, application
problem analysis
May
Client
generate 5.
for
a
production application
Encourages
innovation
may
be
unknowledgeable
6.
Environment to resolve
unclear objectives
7.
Incomplete
implementation
specifications
6.
Identifying
Confusing or difficult
validation,
5.
without
Prototype
2.
identified
4.
be
rigorous evaluation of
can be identified easily
3.
may
awarded
developers
2.
Contract
Approval process and
requirement is not strict
7.
Requirements
frequently
may
change
42
and flexible designs
significantly
In this project, methodology will base the general technology, which named
System Development Life Cycle. This methodology consists of five phases, that
summary in Figure 3.1:
i.
Feasibility and planning phase
ii.
Function and requirement analysis phase
iii.
System design phase
iv.
System build phase
v.
System testing and evaluation phase
43
Phase 1
Feasibility and planning
Phase 2
Function and requirement
analysis
Phase 3
System design
Phase 4
System build
Phase 5
System testing and
evaluation
Figure 3.1 Project methodology
44
3.2.1
Feasibility and Planning Phase
Generally, the feasibility includes the three aspects of possibility,
effectiveness and necessity. Possibility includes the feasibility of technical, material,
financial and personnel support. Effectiveness includes the implementation of the
project can bring economic effective and social effective, and necessity is more
complex, including the social environment, personnel quality and cognitive level
various factors, etc (http://searchcio-midmarket.techtarget.com).
Before the investment decision-making in the project, through the
investigation of situations of engineering, economic, social and other aspects,
comparing the possible solutions and analyzing the economic and social effective
after the completion of investment project, for inspect the project advanced of
technically, the rationality and profitability of economic, and the possibility and
feasibility of the building; then, to determine the feasibility of project investment.
The feasibility phase includes many aspects, which can be summarized the
following aspects: technical aspect, economic aspect and operating environment
aspect (www.uml.org.cn).
Technical feasibility analysis is that in the technologies and products of
current market, possibility of using the current or as well as likely to have the
capacity of technical, product features and human resources to achieve the project
objectives, functionality and performance; and whether or not can complete the
project in the limit period of time.
Technical feasibility analysis should in general be considered (www.uml.org.cn):
45
i.
The risk of project
ii.
The effectiveness of human resources
iii.
The possibility of technological
iv.
The availability of product
In the process of analyzing operating environment feasibility, the main focus
is that whether it can build the environment for running the system, and the works
that involved in building the environment, for put these works into the project plan.
Generally, the feasibility phase can be summarized into several steps
(www.uml.org.cn):
i.
Finding the exiting problems
ii.
Analyzing issues
iii.
To determine project objectives
iv.
To export and evaluation of various options
v.
To understand
For this project, this phase is relating to the chapter 1 and 2, for finding the
problems of current evaluation system and the roots of those problems. And then,
determine the objectives of this project and make a general outcome of this project
for solving those problems. At the last, determine this scope of this project, such as,
which technologies should be used and which knowledge should be defined to
support this project. In this planning phase, at the first should understand all the
components of C2C credit risk and current evaluation system. It can be through
defining the concepts of credit, E-commerce, e-commerce credit, risk, credit risk, and
online-trading mechanism. When understand all of above components, it can be more
46
understand the C2C online-trading mechanism, and then based analyzing the C2C
system structure, characteristics and online-trading model to find out the current
problems.
3.2.2
Requirement Analysis Phase
After the feasibility phases, that ensure the feasible of system, the next step
should analysis the functions that the system should complete. Function and
requirement analysis phase is a very important phase. Only understanding the
functions of the new system should be having, then can understand the requirements
of new system. The success of this phase can lay a good foundation of the success of
the whole system. However, the requirement is also keeping develop and change
during the whole process of system development, so there need to make the
requirement changing plan to meet the changing, and protect the process. In this
phase, the main objective is understand the main functions of the new system, and
according this function to develop system requirements, that can be following several
steps (http://www.doh.state.fl.us/irm/Apps/sdstandards/isdm/Design.pdf):
i.
Collecting information
ii.
Defining the system requirements
iii.
Making the priority of requirements
iv.
Generate and evaluate the options
47
In this phase, it is basing the feasibility and planning phase to find out the
requirements of doing this project, such as the requirements of solving the problems
of current credit evaluation system, software requirements and hardware
requirements. And it is putting all of those requirements in priority to find out which
requirements are more important.
3.2.3
System Design Phase
During this phase, it is using the documented from the function and
requirements phase to define the physical implementation of the system. Prototyping
is a highly recommended technique for ironing out design issues, user interface
specifications, and otherwise providing a flow of communication between and
among developers and customers (www.uml.org.cn).
This phase is mainly through the results of functions and requirements phase
to design the credit evaluation system, which is based the current credit evaluation
system, such as system framework, database design and so on. The main outputs of
this phase are (www.uml.org.cn):
i.
Designing application structure
ii.
Designing system interface
iii.
Designing a database
iv.
Designing details
v.
Design system control
48
In this phase, the mainly outcomes are the new database that supporting the
TaoBao’s current credit evaluation system and improving it. It includes a new
database, new evaluation mechanism and new website of making evaluation.
3.2.4
System Build Phase
The main task of this phase is exchanging the results of last phase, system
design; to code that computer can recognize and run. So it demands are using a
unified internationally development standard, for ensure the readability, easing
maintenance, therefore, to improve the efficiently of the system.
Modular and subsystem programming code will be accomplished during this
stage (Wikipedia web). Unit testing and model testing is done in this stage for make
sure the credit evaluation system is working.
3.2.5
System Testing and Evaluation Phase
Thick in the software design is completed, to go through rigorous testing to
find the software in the whole process and rectify the problems. Entire testing
process is divided into unit testing, assembly testing and system testing three stages.
49
i.
Unit testing is a software verification and validation method in which a
programmer tests if individual units of source code are fit for use; a unit is the
smallest testable part of an application (Wikipedia web).
ii.
Integration testing (sometimes called Integration and Testing, abbreviated
"I&T") is the activity of software testing in which individual software modules
are combined and tested as a group; it occurs after unit testing and before system
testing (SoftwareTestingClub.com).
iii.
System testing of software or hardware is testing conducted on a complete,
integrated system to evaluate the system's compliance with its specified
requirements; system testing falls within the scope of black box testing, and as
such, should require no knowledge of the inner design of the code or logic (IEEE
Standard Computer Dictionary, 1990).
When all the tests have been done, this credit evaluation system also should
be satisfaction of users (online traders). It means this system should have high
quality standard of credit evaluating.
3.3
Hardware and Software Requirements
Hardware and software is the mostly components of computer. They are used
to support and design the computer system and applications. In this project, for
design a credit evaluation system, there are the lowest collocation of hardware and
some software that need to use.
50
3.3.1
Hardware Requirements
For supporting to use the software and windows system, such as Macromedia
Dreamweaver 8.0, StarUML and MySQL, there is the list of the lowest collocation of
hardware:

Intel® Pentium® 233MHz or higher

Windows® 2000, Windows XP™, or higher

Microsoft® Internet Explorer 5.0 or higher

128 MB RAM (256MB recommended)

110 MB hard disc space (150MB space recommended)

CD-ROM drive

SVGA or higher resolution monitor (1024x768 recommended)

Mouse or other pointing device
3.3.2
Software Requirements
Software is a general term used to describe the role that computer programs,
procedures and documentation play in a computer system (NJ. Retrieved, 2007). In
this project, there is some software need to be designed the credit evaluation system:
Microsoft word 2007, Microsoft Project 2007, Macromedia Dreamweaver 8.0,
StarUML 5.0.2, MySQL and Visible Analysis 7.6.
51
Microsoft Office 2007 (officially called 2007 Microsoft Office System) is the
most recent Windows version of the Microsoft Office System, Microsoft's
productivity suite.3.4 System environment identification (Wikipedia.com). It is used
to write report of this project.
Microsoft project 2007 is the project management software of the Microsoft
Office System. It provides all the tools that need for effective planning, tracking,
problem solving, sharing and completing a project in keeping with conventional
management principles and practices; and also can be used for a project of any size
(Cheltenham Courseware Pty. Ltd, 2008). This software is used to make a schedule
of doing this project.
Dreamweaver (formerly Macromedia Dreamweaver) is a web development
application originally created by Macromedia, and is now developed by Adobe
Systems, which acquired Macromedia in 2005 (Wikipedia.com). It includes PHP,
JAVA and many other computer languages that are used to design a web application
and design a meth for calculate credit ranks.
StarUML is the software that supplying the platform to use UML (Unified
Modeling Language). “The Unified Modeling Language (UML) is a standard
language for specifying, visualizing, constructing, and documenting the artifacts of
software systems, as well as for business modeling and other non-software systems
(David Braun, 2000). The UML represents a collection of best engineering practices
that have proven successful in the modeling of large and complex systems. The UML
is a very important part of developing objects oriented software and the software
development process; it uses mostly graphical notations to express the design of
software projects; it is also using the UML helps project teams communicate, explore
52
potential designs, and validate the architectural design of the software (David Braun,
Jeff Sivils, Alex Shapiro, Jerry Versteegh, 2000)”.
MySQL stands for "My Structured Query Language", it is the program runs
as a server providing multi-user access to a number of databases (Wikipedia.com). In
this credit evaluation system, MySQL is used to store the user credit information, and
supplies the current and history credit information of online traders.
3.4 Project Plan
In this project following the above methodology will have done the
feasibility and planning phase and function and requirement analysis phase. In the
chapter 1, it was focusing on the find the current situation and problem of
e-commerce credit risk. In the chapter 2, accorded to the literature review, represent
the concepts of e-commerce credit risk and the roots of the risk. In addition, analyzed
the C2C e-commerce credit risk and used the case study to present the lacks of
current general C2C credit evaluation system. In the chapter 4, going to analyze the
particular C2C Company, that called TaoBao online-trading system, find the main
problems and give some primary solutions. And all the solutions will be
consummated, designed and implemented in the project 2.
53
CHAPTER 4
DATA ANALYSIS
4.1
Introduction
TaoBao is the successful case of e-commerce sites; it is a typical
representative of C2C business model. It was founded by Alibaba that is one of the
best B2B Company of the world in May 2003, to be the largest C2C trading site of
the world. TaoBao that appertained Alibaba is now ranked the third largest
online-trading sites in Aisa, is becoming a representative of China C2C e-commerce
model.
With the development of internet in China, TaoBao has more than 10 million
members, and getting 1 million new members per month on average, there are 100
thousands full time shops, and average daily visits reached 60million. The number of
goods from 20000 that the original registration to reach 7 million. According to the
54
financial reports of first three quarters of 2005, TaoBao’s successful transactions are
over 50 million RMB by the total.
However, the credit issue has increasingly become an important factor that
baffling the development of TaoBao. One of the mainly aspect is that some sellers
use the bug of TaoBao’s credit system to get high credit ranks without real
transaction. Another credit issue is that the buyer’s credit is not reliable. The buyer
may give a negative evaluation to seller because personal moral problem.
In this chapter, the current TaoBao online-trading system will be presented.
According to the analysis the structure, processes and the underlying factors of
TaoBao’s current system, pointing the problems of TaoBao’s credit system, and
giving a preliminary solution.
4.2
Current System of TaoBao Company
The whole business model is: user registration – identity authentication –
issuing information – purchase – payment in ZhiFubao – send commodity – confirm
receive – payment in seller – credit evaluation. The structure of TaoBao
online-trading can be simply presented in the following Figure 4.1:
55
Figure 4.1 TaoBao online-trading system
We can see there are two main parts of credit system in TaoBao’s system
structure, one is the ID identify, another is user credit evaluation system.
ID identifies part:
56
i.
TaoBao’s registration authentication mechanism
For buyer, through a virtual member name and e-mail to register, following
these steps: filling information, activate accounts and registration successful. In order
to prevent malicious registry, there are two ways of activation process: e-mail and
cell phone (a phone number only can activate one user account). User that logging in
with the activated account can choose purchase commodity or issue purchase
information that let seller to find, in the “my TaoBao” web site.
For seller, asked to approve the real-name authentication, and then released
10 commodities before they can shop in TaoBao. TaoBao provides an e-shop home
page for seller.
ii.
TaoBao real-name authentication
Logging in TaoBao, clicking “real-name authentication” button, can enter the
certification application page, there will be two selections box: “personal certificate”
and “business certification”. Filling the required information and providing the
validity documents and fixed telephone number to registration. Users who under the
age of 18 years cannot become certified members of TaoBao; and the members that
already certified members cannot change their name and ID number.
There are two ways for submit documents: electronic ID card photos or the
proving certificate of personal authentication, such as identity card, passport or
driver’s license.
57
iii.
The legal analysis of TaoBao’s real-name authentication
Firstly, although many countries of the world already have the e-signature
law or have relevant laws that recognized the legal validity of e-signature, but in the
e-commerce especially in the transaction of B2B and C2C, there are few to use the
e-signature. Secondly, TaoBao’s real name authentication only for the sellers.
Thirdly, e-mail plays an important role in the trader’s identity authentication
mechanism.
User credit evaluation part:
TaoBao members successfully take the deal each time; they can make a credit
evaluation to their trader. They have three selections when they make an evaluation
are: “positive”, “neutral” and “negative”. The basic rules are showing that:
i.
Each selection of evaluation has different scores, “positive” gets “1” score,
“neutral” is “0” score and “negative” is “-1” score.
ii.
For each month, the evaluation between same buyer and seller scoring no
more than 6 times, the extra evaluation scoring will not be scoring.
iii.
If there are a number of transaction between same buyer and seller within 14
days, all the “positive” evaluation only given 1 score, and all the “negative”
evaluation given -1 score.
iv.
Evaluation of whether get scores generally not indicated in the contents of
evaluation, only the evaluation does not through the ZhiFubao, will indicate
no score.
v.
If one of the trader has evaluated but another trader does not, and the
transaction that through ZhiFubao is success, the system will give a positive
58
evaluation in default after 45 days.
In the transaction as a seller’s role, its credit is divided in 15 ranks (shows in Figure
4.2):
Figure 4.2 Seller’s credit ranks (http://www.nb-020.com, 2007)
59
In the transaction as a buyer’s role, its credit is divided in 15 ranks (shows in Figure
4.3):
Figure 4.3 buyer’s credit ranks (http://www.nb-020.com, 2007)
There is the TaoBao current credit evaluation system diagram:
60
Figure 4.4 TaoBao credit evaluation system
According to the above structure, TaoBao’s credit evaluation system is
following these steps:
61
i.
User registration, to provide basic registration credit and confirm identity
and other information, if pass ten the next step, otherwise have to re-register.
ii.
According to the user’s basic registration information, using credit
evaluation model, give the basic credit to user.
iii.
To determine whether the user go to online transaction if yes, then record
the situation of the transaction, and use the current credit evaluation model
to give the scores, and store to the user credit files.
iv.
To determine whether the user queries other user’s credit, if yes, display the
required user’s credit and transaction history.
v.
To determine whether to browse, if yes, will display the rules and
regulations of the credit system.
vi.
4.2
Exit system.
Problem of TaoBao’s Current System
According to the above introduction of current system, there are some exiting
problems in the system that should be solved.
i.
At the first, there important thing is the way to identity authentication.
Obviously, the current system uses virtual member name and e-mail to
register, and using cell phone to activation are not safely enough, such as use
the virtual information to authentication, the result also is unreliable.
ii.
The current system just authenticates to the sellers, it makes buyers credit
unreliable, so the buyers may evaluate credit unbending.
62
iii.
In the TaoBao credit evaluation system, a high credit rank can get more
customers coming; therefore, many traders try to improve their credit rank,
but from low-grade to high grade require a number of transaction and the
need spend long period of time, so some traders have used the system bugs,
cheating to get the high ranks.
iv.
For the new user, their credit rank is less than the older user. They have to
face the less number of customers, and the few transactions cannot support a
good credit rank, so there is the negative feedback for new users to improve
their credit ranks.
v.
Credit evaluation model is too simple. Credit evaluation is only set up
“positive”, “neutral” and “negative” ranks. But in the actual transaction a
“positive” evaluation cannot show all the aspects of the trader, it cannot
reflect to the transaction history.
vi.
In the TaoBao credit evaluation system, whatever the turnover of transaction
is more or less the evaluation only give the same scores. But obviously, the
huge number of the turnover should face to more risks, so the credit system is
not fair for the traders who have more turnover transactions.
4.4
TaoBao Credit Evaluation System Related Questionnaire
In this part, a questionnaire of TaoBao credit evaluation system will be
designed and analyzed. This questionnaire is used to get the opinion and suggestion
from TaoBao system user; and those information will be used to design a new
framework of credit evaluation process. This questionnaire can be divided in two
section A, B, C and D. Mostly results of those questions will be used for the build the
63
new credit evaluation mechanism or method in chapter 5.
4.4.1
Section A: Personal Information Questions
This section is for getting the personal information of sponsors. In the age
collection shows 73% of the age between 18 to 24, 24% of the age between 25 to 30,
and 3% of the age between 31 to 40, it shows in Figure 4.5. Based on the role of
transaction and the frequency of using TaoBao system collection, there are four
persons are the seller of the current system and using TaoBao system every day.
Others are the buyer of TaoBao system. There are 10 persons use TaoBao system at
least once during one week. Other 16 persons use TaoBao system once of around one
month, show in the following 5able 4.1 and Figure 4.6. We can obvious to see, based
on the section A, the main user of TaoBao system is stripling, who are very kind to
accept fresh idea.
25-30
24%
31-40
3%
18-24
73%
Figure 4.5: Age Distribution
64
30
25
20
15
10
5
0
Seller
Buyer
Figure 4.6: The Distribution of Role
Table 4.1: The Frequency of User
Items
Percentage
Everyday
4
13%
One week
10
34%
One month
16
53%
4.4.2
Quantity
Section B: Optional Questions of TaoBao Credit System
There are three questions are related to the TaoBao’s current credit system,
which include two single options and one multi-option. The first question is about
the general comment of TaoBao current credit system. The accounted number and the
percentage is showing the following Table 4.2.
65
Table 4.2 General Comment of TaoBao Current Credit System
Item
Quantity
Percentage
User’s credit rank is not trustable
1
3.5%
User’s credit is too general to understand
17
57%
I have no idea
2
6.5%
User’s credit can be presented in some degree
10
33%
User’s credit can be presented very clear
0
0%
From the results, 57% persons thought TaoBao’s current credit system cannot
show user’s credit very good; they cannot get the details of credit from the current
system.
It is because the current system only puts the credit to positive, neutral and
negative, which are too simple to describe user’s credit situation.
The second question is about the problems that user have been faced. This
question is a multi-option. All the options are from the general problems that user
may be faced. The results are showed in Table 4.3.
Table 4.3: The General Problems Analysis
Items
Quantity
Percentage
Default positive evaluation impacts the credit real
2
7%
Hard recognize the facticity of credit
10
33%
Evaluated by others with untrue and malice reason
4
14%
Forced to give a positive evaluation because others’ bully
4
14%
situation
66
Have to give a positive evaluation because no big reasons
25
83%
The problems appeared after evaluation
24
80%
The credit score is unfair, when the turnover is large
2
7%
According to the results, there are two problems are the user frequency to be
faced. The most important one is about that is that user give a positive evaluation
because no big problem. The current system only supplies the three options of
evaluation, so almost, if people does not face a big problem, they would like to give a
positive for keeping other credit situation, and can save many troubles.
The third question is to know whether people will check other’s credit before
they go to trade. It is to understand user’s behavior. The results shows in Table 4.4.
Table 4.4: User’s Behavior Check
Items
Quantity
Percentage
Never
0
0%
Sometimes
6
20%
All the time
24
80%
The result is fortunately. Almost people is going to check other’s credit
situation, this behavior can be very good to support this research.
67
4.4.3
Section C: Optional Questions of Authentication Part of TaoBao System
The question of this section is to collect the opinions of the current
authentication part of TaoBao system. The first question is to get the general opinion
of current authentication method. The result showed in the Table 4.5.
Table 4.5: The General Opinion of TaoBao’s Authentication Method
Items
Quantity
Percentage
It is too simple to authenticate the identity
6
20%
Cannot be good to ensure each people just have one user
12
40%
I have no idea
2
7%
It can ensure the identity of most user
6
20%
It can make sure each people only have one user id
4
13%
id
From the result we can easy to see, there are 60% persons think the current
authentication method cannot make sure one people just has one user id. Therefore,
the authentication method need to be improved, such as put more personal
information to authenticate.
The second question is to collect the opinions of whether a buyer should go
to the authentication. The results showed in Table 4.6.
68
Table 4.6: The Question of Buyer Authentication
Items
Quantity
Percentage
No, because their identity is not important
9
30%
I have no idea
0
0%
Yes, they need go to authenticate
21
70%
According the result, almost people think buyer also need go to
authenticate to ensure the identity of buyer. And forcing their evaluation is
responsible.
The third question is about the new member of TaoBao system. Because
when a new member register in the TaoBao system, they do not have any credit rank,
the question wants to know whether other people would like to trade with new
member. The result showed in Table 4.7.
Table 4.7: Question of New Member’s Credit Situation
Items
Quantity
Percentage
Yes, I can trust them
10
33%
I have no idea
5
17%
No, I cannot trust them
15
50%
Based on this result, there are 50% people do not want to trust new
member, because they cannot know the credit situation of new member. It needs a
way to support the new member’s credit situation, such as the make the
69
authentication more strict or give the new member a initial credit.
4.4.4
Section D: Questions of Credit Evaluation of TaoBao System
The questions of this section are about the credit evaluation system of TaoBao,
which include two parts of initial credit part and dynamic credit evaluation part. One
of the question in this section is single option. Other questions have some indicators
of each question and need to give a mark for each indicator.
The first question is a general question to collect the TaoBao user’s opinion of
the current credit evaluation system, it only uses positive, neutral and negative to
describe user’s credit situation. The result showed in Table 4.8.
Table 4.8: Question of The Current Credit Evaluation System
Items
Quantity
Percentage
It cannot show the user’s credit situation
5
17%
It is not good to show the details of user’s credit
16
53%
I have no idea
1
3%
It can show the situation of user’s credit in some degree
8
27%
It definitely can show the situation of user’s credit
0
0%
70
Based on this result, we can obviously see there is nobody think the current
credit evaluation system can definitely show the situation of user’s credit. However,
near 70% persons think the current credit evaluation system cannot or not be good to
show the user’s credit situation. It needs a more details evaluation process to describe
user’s credit situation.
The questions from 8 to 15 need to be given a evaluation for each indicator;
there are five options of evaluation, strong disagree, disagree, neutrally, agree and
strong agree. It can be divided into two parts, the first part, from 8 to 11 questions, is
about the dynamic credit indicators confirm. The first question of this part is about
importance of the turnover during a transaction. Because the different transaction has
a different turnover, the current credit evaluation system of TaoBao gives all the
transaction a same credit score; it does not care the impact of turnover. It is unfair to
the user who has a large turnover transaction, because they have to face a huge credit
responsibility. The result showed in Table 4.9.
Table 4.9: Data Collection of Importance of Turnover
Items
Quantity
Percentage
Strong disagree
0
0%
Disagree
1
3%
Neutrally
4
14%
Agree
16
53%
Strong agree
9
30%
71
Based on the result, more than 80% people think the turnover is very
importance indicator during the transaction, so a transaction, which has a higher
turnover, needs to get more credit score from the evaluation.
The second question of dynamic part is about whether the user’s credit rank is
important during a credit evaluation process. The result showed in Table 4.10.
Table 4.10: Data Collection of Importance of User’s Credit Rank
Items
Quantity
Percentage
Strong disagree
0
0%
Disagree
4
13%
Neutrally
4
13%
Agree
14
47%
Strong agree
8
27%
Based on the result, mostly people think the user’s credit rank will impact the
evaluation result. Because the user, who has a lower credit, may have a unreliable
credit situation, even evaluate un-responsible. If adding the user’s credit rank to the
credit evaluation process, it can solve this problem in some degree.
The next question is for confirming the seller’s evaluated indicators. There
are six indicators, which from the expert’s resource (Zeng Yong, 2004). For giving
the marks of each indicator, 1 is strong disagree, 2 is disagree, 3 is neutrally, 4 is
agree and 5 is strong agree. The method of ensuring the indicators is according the
72
mark and calculating the average mark, then from the average mark to get the
percentage of the importance of indicators. If the percentage more the 50%, it means
this indicator is important, more details showed in Table 11.
Table 4.11: Question of Seller’s Evaluation Indicators Confirming
Quantity
Items
Average
Percentage
6
3.9
78%
12
18
4.6
92%
8
18
4
3.9
78%
0
5
21
4
4.0
80%
0
2
8
13
7
3.8
76%
0
6
16
7
1
3.1
62%
1
2
3
4
5
Deliver Time
0
0
10
14
Quality
0
0
0
Turnover
0
0
0
Seller service
Price
Buyer’ credit
rank
The result is very gratify and under the expect. The percentage of all the
indicators are more than 60%; it means the indicators are improved by users.
The next question is about the buyer’s evaluated indicators. It is similar with
the seller’s indicator; the way of confirm the importance of indicators is also same
with the above question, more details showed in Table 12.
73
Table 4.12: Question of Buyer’s Evaluation Indicators Confirming
Quantity
Items
Average
Percentage
3
3.4
68%
8
0
3.1
62%
8
18
4
3.9
78%
0
4
21
5
4.0
80%
0
4
15
9
2
3.3
66%
0
5
17
7
1
3.1
62%
1
2
3
4
5
0
3
15
9
0
4
18
Turnover
0
0
Seller’ credit rank
0
Response Time
Satisfaction
of
commodity
Communicate
with
seller
Price
The result of those indicators chosen is not good like the seller’s. However, it
also has more than 62% satisfaction of each indicator.
The following questions belong to the static indicators selection (Zeng Yong.
2004). It is to confirm which elements of each indicator of initial credit evaluation
has more credit priority. The first question is to decide the people’s age in which
level has higher credit. The result showed in Table 13.
74
Table 4.13: Question of The Priority of Age Level
Quantity
Items
Average
1
2
3
4
5
Children (age<16)
0
24
5
1
0
2.2
Youngster (16<age<30)
0
4
9
15
2
3.5
Midlife (31<age<55)
0
1
10
8
11
4.0
Senior (55<age<70)
0
3
10
13
4
3.6
Old people(age>70)
0
5
21
3
1
3.0
Based on the result, we can know most of people do not trust the children,
who’s age below 16, because normally they do not have financial ability and they
always no need to response to their behavior. The most trustable people is who
during the age of 31 to 55, because when people in this age, almost of them have a
steady job and income.
The next question relates to the education level of people. The result showed
in Table 14.
Table 4.14: Question of The Priority of Education Level
Quantity
Items
Average
1
2
3
4
5
Doctor and higher levels
0
0
4
16
10
4.2
Master
0
0
5
19
6
4.0
Bachelor
0
2
9
16
3
3.7
75
College
1
4
19
6
0
3.0
Middle school and lower levels
5
4
16
5
0
2.7
From the results, we can easy to see, the people, who has a education level
more than master, can get a high credit trust. Other hand, if the people, who has a
education level less than middle school, only can get a few people’s trust.
The next question is related to the marital status of people. It is to get the
opinions of which marry status of people has higher credit, more details showed in
Table 15.
Table 4.15: Question of The Priority of Marital Status
Quantity
Items
Average
1
2
3
4
5
Married with children
0
0
3
19
8
4.2
Married with no children
0
0
8
17
5
3.9
Single
0
3
8
14
5
3.7
Divorced with children
2
8
8
11
1
3.0
Divorced with no children
3
9
8
9
1
2.9
From this result, we can see that the people, who married and with children, is
the most trustable by others. Because we believe that this kind of people is more
understand what is the responsibility and they should be a example to their children.
76
Next question is about the gender, which has more trust from the user of
TaoBao system. The result showed in Table 16. This result shows the female has a
higher credit that be trusted by TaoBao’s user.
Table 4.16: Question of Gender Confirm
Items
Quantity
Percentage
Male
4
13%
Female
26
87%
The next question is for put the priority for the different career. According
the result to decide which career has more trustable from TaoBao user. The result
showed in Figure 4.7.
30
25
20
15
10
5
0
Figure 4.7: Question Result of Career Priority Confirm
77
Based on this result, the careers can be divided in four parts. The first level is
doctor and teacher, which have more than 25 people think these two careers have the
highest credit level. The second level is finance, power, post and telecom, which
have more than 15 people think them can be trusted. Other level is public clerk,
lawyer, soldier and reporter. The last level is business owner, employee and others.
The last question is the salary more than which level can be trusted. The
result showed in Table 17.
Table 4.17: Question of Salary level Confirm
Items
Quantity
Percentage
1000
0
0%
2000
2
7%
5000
4
14%
10000
24
79%
From the result, we can know that the higher salary level can be more trusted
by others.
4.4.5
Conclusion of TaoBao Credit System Questionnaire
According the analysis of TaoBao credit system questionnaire, the author
confirmed the general problems of TaoBao credit system and got the basic
information for the new credit evaluation system to support the mechanism of system
78
design in Chapter 5.
At the same time, based on those information, one solution of the TaoBao
current credit system will be supplied.
4.5
Proposed Solution
According to the above information of questionnaire, there are some
problems and solutions are appeared.
For solving those above problems, the following solutions should be performed:
i.
The register process can combine the identity number and bank account,
through these two steps to ensure the authenticity of user’s information.
ii.
Creating the authenticating system for buyers to force the authenticity of
buyer’s evaluation.
iii.
To improve the bugs of current system to prevent the credit ranks cheating,
such as the above two aspects that can reduce the cheating in some ways.
iv.
Giving the original credit scores to new users depend on the registering funds;
and should confirm by the bank account.
v.
Making the evaluation into details, such as from service, the effective of send
commodities and after-sales service, then get a final average scores.
vi.
Putting the turnover into credit evaluation system.
79
There is a framework to show how to solve those problems step by step. At
the first, a user authentication part should be completed, such as the improving of
seller authentication and force the buyer going to authentication. Secondly, after
authentication part, user need fill more information, which from the questionnaire, to
get the initial credit. The initial credit can help people get a basic credit rank to be a
reference when they doing the first transaction. Then, the new system goes to the
dynamic credit evaluation part. This part should confirm the dynamic credit
evaluation indicators based on the questionnaire results. The new system will use a
fuzzy evaluation mechanism to improve the current credit system, not only simple
use the positive, neutrally and negative to evaluate, but more details of the
transaction process. The framework of this process is showed in Figure 4.8.
Multi-authentication Structure
Seller's identity authentication
Buyer's identity authenticaion
Credit Evaluation Structure
Grant Initial Credit
Dynamic Evaluation
Seller's Credit Evaluation
Byer's Credit Evaluation
Get Credit Scores
Figure 4.8: The Process of Improved System
80
4.6
Summary
In this chapter, mainly introduced the TaoBao online-trading company,
according the analyzing of the current credit evaluation system and the basic
information from questionnaire to Figure out the current problems and get the
necessary information. Focusing on some lacks of current credit evaluation system
give the preliminary solutions for the system; and using the results of questionnaire
supports the improved credit evaluation mechanism in Chapter 5.
81
CHAPTER 5
SYSTEM DESIGN
5.1
Introduction
In a virtual e-commerce market, commodities are intangible; and the traders
are anonymous. Therefore, buyers are very difficult to know the true identity of
sellers and hard to understand the quality and usability of goods. At the same time,
sellers are facing the same situations and problems, which it cannot be sure whether
the buyer will pay the relevant amounts. Those are the information asymmetry (Choi
Sithal and whslotn, 1997), which means that the two sides of the transaction do not
have the same amount of information.
It is providing the chance to fraudulent traders for get more benefit that
through betray. For the traders who are credibility, because the media (internet)
space-time, which caused by the separation, will enhance it of being cheated
(Brynjolfsson and Smith, 2000). Therefore, the main purpose of this chapter is to
82
summary the current problems and shortcomings that earlier out of the current
systematic analysis; and according to those theories design and establish solution.
These include design multiple authentication methods; and it uses fuzzy methods to
evaluate user’s credit.
5.2
Multi-Authentication
Authentication is the most typical features of e-commerce credit. The people
who involved in the transaction must be registered and login, what means both of
buyers and sellers should have a certain identity. However, there two practices in
actual occurrence are: the true identity and anonymity. Because the current TaoBao’s
authentication and management system are unable to conduct effective of user
authentication, which led to the diffusion of a large number of false user information.
Computer systems and network is a virtual digital world. In this digital world,
all the information that including the user’s identity information is using a specific
set of data indicated. Computer only can recognize the user’s digital identity. The
entire user’s authorization is the authorization for the user digital identities. However,
the world we live in is a real physical world. Everyone has a unique physical identity.
How to ensure the digital identity operations is the legal owner of this digital number,
which means the operator’s physical identity and digital identity that corresponds.
However, how to make a person’s physical identity and digital identity is fully
consistent has become a very important issue.
83
As the trading platform have a large number of users. Each user had to pass
the registration and landing and with others who are supported by TaoBao system
and related by bank accounts can be doing transactions. The current way of
registration is a simple way of e-mail, ID number and phone number to be certified.
However, if the authentication can be related the ID information and bank account
information, it will be able to a certain extent large to ensure that each user just has
only one account. Thus, it can be avoiding the self-purchased and with the bank’s
real name system will be more integrated operability. At the same time, the content
of user’s authentication also should include the information of user’s age, gender,
marriage situation, education, career and salary.
The main purpose of this section is that when user doing authentication, it can
be combined with banking system to ensure the authenticity of filled banking
information. This paper will use the MySQL database to simulate the banking system
to confirm the user’s input information.
5.3
Buyer Authentication
The current credit system also lacks the considered of buyers, what means
buyers no need pass the identity authentication and can do the credit evaluation. It
makes buyers credit unreliable, so the buyers may evaluate credit unbending. So it is
urgent need an authentication system of buyers’ identity, which can be reference to
the sellers’ authentication mechanism.
84
5.4
Credit Evaluation Method
This part mainly introduces the credit evaluation method, which uses fuzzy
method from multi-aspects to evaluate user’s credit. Furthermore, it will also
introduce the factors of evaluation and its’ weight.
5.4.1
Fuzzy Method
Fuzzy comprehensive evaluation is the method of fuzzy mathematics method.
It is evaluating a number of factors that are difficult to use a precise number to
quantify; and it is according certain evaluation criteria to be a quantifiable form.
Using the fuzzy method to the stakeholders of C2C online-trading system can
consider many factors that will impact the transaction; and it according to the degree
of importance of each factor and the evaluation results can improve the original
qualitative evaluation results to quantitative evaluation. Therefore, it will be better
handling the problems of transaction; and it provides a comparison and
discrimination of basis for buyers and sellers.
The basic idea of fuzzy evaluation method is: it is based on the determining
the levels and weights of evaluation factors and sub-factors to use the transformation
of fuzzy theory and use membership of various factors and sub-factors to structure
fuzzy evaluation matrix. Then, it can be through the multi-layer composite operations
to determine the respective levels of object (He Zhongxiong, 1985).
85
This thesis uses fuzzy comprehensive evaluation method to do the evaluation
for TaoBao credit system. It is mainly based on the following reasons:
i.
C2C credit risk evaluation is a very complex process, which is affected by many
factors. Using only one or a few indicators alone cannot fully reflect the credit
risk of C2C transaction.
ii.
C2C credit risk itself has a certain degree of ambiguity. There is no strict
boundary of credit risk levels, but of degree. Using strict mathematical model
cannot accurately define the vague phenomenon.
According to the TaoBao system characteristics and requirements, when we
are addressing the problems of credit evaluation, we should consider the evaluation
process to be simple, convenient and practical. So this thesis, which is according the
defects of fuzzy evaluation method, carries out the appropriate modifications.
i.
In the selection of indicators, since this is mainly based on the credit problem of
TaoBao transactions process. It is from the roles of transaction stakeholders to
evaluate the C2C credit risk. Consider the limitations of capture information
during the transaction process, this thesis only select the basic information of
transaction stakeholders and a few indicators, which include the transaction
process, to measure the credit risk for easier access data.
ii.
For determining the weight of indicators, this thesis chose the expert method to
give empowerment of indicators. Because the indicators of C2C are chosen by
the factors that online user are more concerned. Thus, it is according the
investigations and certain methods, which determined by experts, to obtain the
indicator weight.
86
iii.
For establishing the model architecture, because the fuzzy evaluation method
need calculate the membership of each indicator when it is doing evaluation. The
workload is large and complex. Thus, this thesis simplifies the fuzzy membership
function to construct a new fuzzy evaluation model, which simplifying the
evaluation process. In practice, it is redesigning the membership that according to
the function of each indicator for achieving a better rating results.
The C2C credit evaluation model, which is based on the fuzzy evaluation
method, is showed in following Figure 5.1:
Credit Indicators Choose
Static Credit Indicators
Dynamic Credit Indicators
Build Credit Indicators Evaluation Structure
Set Scoring function of each credit indicators by
using fuzzy method
Using Expert Method to determine weight of
each indicators
Integrated Evaluation Credit Indicators
Initial Credit Evaluation
Dynamic Credit Evaluation for Once Transaction
Credit Evaluation Scores
Figure 5.1: Fuzzy Credit Evaluation Framework
87
In the process of individual credit evaluation, first of all, based on the
applicant’s personal information carry out an index score and build the credibility of
its indicators’ vector. Then, the applicant should be matched. At the last, it can get the
credit score by multiplying the credibility of the vector and the vector of the indicator
weight. The improved model as follows:
i.
The system of credit indicators can be combined with the two components of
initial static V1 and dynamic credit evaluation indicators V2.
ii.
Determining the various indicators’ fuzzy membership function to establish an
initial static credit scoring matrix R1 and dynamic credit scoring matrix R2.
iii.
Determining the weight of individual indicators, the weight set of, named W1,
initial static credit evaluation indicators and the weight set of, named W2, initial
dynamics credit evaluation indicators.
iv.
Integrating the various evaluation indicators to construct score vector; and
initialing static credit evaluation vector (r11, r12 … r1n) and dynamic credit
evaluation vector (r21, r22 … r2n).
v.
Calculating initial credit score S1 and dynamic credit score S2, formulas are
following:
𝑛
S1 =
r1i
𝑖=1
n
S2 =
𝑟2𝑖
i=1
vi.
Calculating recent credit evaluation scores S0
n
S0 = S1 × W1′ +
S2i × W2′
i=1
88
vii.
Calculating historic credit evaluation scores S1
m
S=
S0i
i=1
Note: the improved credit scoring models can be combined two parts. The
first part is an initial static credit evaluation, which is recording the user’s credit
scores in the registration; it can be the reference when others doing the first
transaction. Another part is a dynamic credit evaluation, which recording the each
evaluation result; and according two times evaluation can get there three evaluation
index:
i.
The initial credit evaluation index: it is completed in the user initial registration.
During the first transaction, user can use it to refer to other’s credit in order to
decide whether to deal with.
ii.
History credit evaluation index: it is recording user’s all credit scores of
transaction. If the accumulate points higher, indicating the history longer than
other shop, which means more worthy of trust.
iii.
New credit evaluation index: this index reflects the recent credit situation of users.
It is more true and accurate. At the same time, it can be compared between the
new user’s credit degree and old user’s credit degree, no because it is a new store,
which has lower credit rank, so other people do not want to trade with.
89
5.4.2
Evaluation Indicator
This thesis divides the credit evaluation indicators into two parts of static and
dynamic indicators, which are more comprehensively and reflecting the credit status
of the transaction stakeholders. For the buyer and seller, they have some initial static
indicators. However, because they have different role in the transaction, so they have
different dynamic indicators. This thesis only introduces the process of setting
seller’s credit indicator, for the buyer it is the same. The structure of credit evaluation
indicators as show in the following Figure 5.2:
Credit Evaluation Scores
Turnover
Credit rating
Service
Price
Quality
Time
Dynamic Credit Evaluation Scores
Bank deposit
Salary
Occupation
Educational level
Marital status
Gender
Age
Initial Credit Evaluation Scores
Figure 5.2: Credit Evaluation Indicators Structure
i.
Static credit indicators explain
At the first is an initial static credit evaluation indicator. According to the
resource from internet, the questionnaire research and the feature of C2C’s
90
stakeholders, there seven indicator are chosen for initial static credit evaluation
indicator (Zhao Xiaodong, Zhen Tao, 2003,6). These seven indicators are easy to be
confirm when user doing the online register; in addition, these seven indicators have
strongly close with user’s credit. There are age, gender, marital status, education
degree, occupation, salary, and bank deposits. The indicators described below:

Age: Age is a more important indicator in the credit evaluation. In the
general bank individual credit evaluation indicator, they all believe that the
older person has higher credit level than youth.

Gender: women’s credit generally is believed that higher than men’s.

Marital status: married better than unmarried, with children is better than no
children.

Educational level: A person who has the higher education level has better
quality.

Occupation: A person who has a relatively stable degree of career has higher
credibility.

Salary: It is a measure of the credibility of the user’s economic indicators;
the higher income has greater credibility.

Bank deposit: More bank deposits are more worthy of trust.
In the static index selection may still much to be desired. However, with the
improvement of personal credit system, the selection of indicators can be done to
further changes. The initial scores for static credit indicators can be completed by the
system with registered information; and it uses fuzzy scoring matrix automatically to
evaluate credit scores.
91
Because of the TaoBao’s standard of credit evaluation, this thesis uses the
membership function of fuzzy method to establish the credible indicators’ function,
in order to ensure the continuity of quantitative indicators score. The following Table
5.1 shows the membership of credit scoring for each indicator.
Table 5.1: The Scoring Function of Initial Credit Evaluation
Indicators
Age
r11
Gender
r12
Marital
Status
r13
Indicator Scoring Function
Indicator
Scoring
Indicator Creditability Formula
Introduction
Older people have
higher creditability
than youth: below
1
31 ≤ ma ≤ 55
16 is 0 score, 16-30
m1 /10
Others
is 5 score, 31-55 is
0
ma < 16
10 score, 55-70 is 6
score, above 70 is 2
score
Generally
women
have
higher
1
m2 = 2
creditability
than
m2 /2 m2 = 1
men: women is 2,
men is 1
Married better than
unmarried,
with
children is better
than no children:
Married
with
children get 10,
1
m3 = 10
m3 /10 0 < m3 < 10
Married with no
0
m3 = 0
children
get
8,
unmarried get 6,
divorced
with
children
get
4,
divorced with no
children get 2
92
Educational
r14
Level
Occupation
r15
Bank
Deposit
r16
Salary
r17
The
educational
level
has
the
potential for get
fortune: doctor and
above levels get 10,
1
m4 = 10
m4 /10 0 < m4 < 10
master
get
9,
0
m4 = 0
bachelor
get
5,
college get 3, middle
school
get
2,
primary get 1, others
get 0
A person who has a
relatively
stable
degree of career has
higher
credibility:
Teacher / Doctor get
10, Finance / Power
/ Post / Telecom get
1
m5 = 10
m5 /10 0 < m5 < 10
9, Public Clerk /
0
m5 = 0
lawyers / soldiers /
Reporter get 8,
Business
owners
(including
individuals) get 6,
employee get 4,
Others get 2
Bank deposit is the
form and result of
person’s ability of
1
m6 > 5000
m6 /5000 0 < m6 ≤ 5000
creating wealth. The
0
m6 = 0
credit rank can be
determined by bank
deposit
The salary can show
1
m7 ≥ 10000
the ability to repay
m7 − 500 1/2
500 ≤ m7 < 10000 the debt, higher
10000 − 500
salary has more
0
m7 ≤ 500
creditability
93
Based on the above formula can be known the function number of
membership when the evaluation indicators have a fixed value. However, the setting
of membership functions can be improved by the actual situation. These membership
functions are not very accurate; the main object is showing the way of evaluation.
ii.
Dynamic credit indicators description
The second part is dynamic credit evaluation indicators. It is according the
actual transaction status to evaluate the credit, which supports the continuously
updated credit of user. This thesis chooses six indicators to constitute a user’s
indicator system of dynamic credit evaluation.
According to the report of “China’s C2C online shopping survey” shows that
when the buyer choose the seller, 75% of buyers are most concerned about the price
of goods or cost-effective, more than 55% of buyers are concerned about the quality
of goods, and others important aspects that buyers concerned are the delivery time
and service attitude. These results are similar to the research of I did in China by
questionnaire. In addition, in order to prevent the problem of users use cheap goods
to carry out the high credit rank; and to make a more equitable trading environment,
the turnover indicator will be added. Furthermore, to prevent the malicious
evaluation and to make buyer’s evaluation more reliable, the buyers’ evaluation can
evaluate higher scores by the higher credit rank. Therefore, this thesis selected six
indicators as a dynamic credit evaluation indicator; there are: time, quality, service,
price, buyer credit, and turnover. The description of each indicator as follows:

Time: in the promised time, the faster can be more satisfied;

Quality: the quality of goods with the commitment of quality, the closer the
94
better;

Price: compare the price with other same goods, the cheaper the better;

Service: better service has more satisfied;

Credit rating: the higher credit rank user to make the evaluation more reliable;

Turnover: more turnover transaction, the proceeds of the credit score is higher;
In addition, the transaction and buyer credit rank will determine by the
system automatically according to the transaction status. And others will be
evaluated by users.
The scoring of these indicators is following the below Table 5.2:
Table 5.2: The Scoring Function of Seller Dynamic Credit Evaluation
Indicators
Buyer credit rating
Turnover
Time
Indicator Scoring Function
Indicator
Scoring
Indicator Creditability Formula
Introduction
According the current
TaoBao credit ranks,
can be depart to 5
1
m8 = 5
levels: 0-250 scores is
r21
m8 /5 0 < m8 < 5
1, 251-1000 is 2 ,
1001-5000 is 3, 500120000 is 4, >20000 is 5
According
to
the
turnover of transaction,
1
m9 ≥ 1000
r22
m9 /1000 0 < m9 < 1000 more than 1000 is more
trustable
The faster can be more
satisfied, Very satisfied
1
m10 = 5
with 5, satisfied with 4,
r23
m10 /5 0 < m10 < 5
common with 3, not
satisfied with 2, very
95
Quality
r24
Service
r25
Price
r26
1
m11 = 5
m11 /5 0 < m11 < 5
1
m12 = 5
m12 /5 0 < m12 < 5
1
m13 = 5
m13 /5 0 < m13 < 5
bad with 1
The quality of goods
with the commitment
of quality, the closer
the
better,
Very
satisfied
with
5,
satisfied
with
4,
common with 3, not
satisfied with 2, very
bad with 1
Better service has more
satisfied, Very satisfied
with 5, satisfied with 4,
common with 3, not
satisfied with 2, very
bad with 1
The cheaper the better,
Very satisfied with 5,
satisfied
with
4,
common with 3, not
satisfied with 2, very
bad with 1
For the buyer, the indicators are similar to seller’s. There are also selected six
indicators: Credit rating, Commodity Receive Time, Satisfaction degree of
commodity, Communication, price:

Time: after receive the commodity, the faster response can be more satisfied;

Satisfaction: the quality of goods with the initial promise, the closer the better;

Price: the price with the initial price, closer are more satisfaction;

Communication: better communication has more satisfied;

Credit rating: the higher credit rank user to make the evaluation more reliable;

Turnover: more turnover transaction, the proceeds of the credit score is higher;
96
The scoring of these indicators is following the below Table 5.3:
Table 5.3: The Scoring Function of Buyer Dynamic Credit Evaluation
Indicator Scoring Function
Indicators
Indicator
Scoring
Indicator Creditability Formula
Introduction
According the current
TaoBao credit ranks,
can be depart to 5
Seller credit rating
r21
1
m8 /5
m8 = 5
0 < m8 < 5
levels: 0-250 scores is
1, 251-1000 is 2 ,
1001-5000 is 3, 500120000 is 4, >20000 is 5
According
Turnover
r22
1
m9 /1000
to
the
turnover of transaction,
m9 ≥ 1000
0 < m9 < 1000 more than 1000 is
more trustable
After
receive
the
commodity, the faster
response can be more
Time
r23
1
m10 = 5
m10 /5 0 < m10 < 5
satisfied, Very satisfied
with 5, satisfied with 4,
common with 3, not
satisfied with 2, very
bad with 1
Satisfaction
r24
1
m11 = 5
m11 /5 0 < m11 < 5
The quality of goods
with
the
initial
97
promise, the closer the
better, Very satisfied
with 5, satisfied with 4,
common with 3, not
satisfied with 2, very
bad with 1
Better communication
has
more
satisfied,
Very satisfied with 5,
Communication
r25
1
m12 = 5
m12 /5 0 < m12 < 5
satisfied
with
4,
common with 3, not
satisfied with 2, very
bad with 1
The price with the
initial price, closer are
more satisfaction, Very
Price
r26
1
m13 = 5
m13 /5 0 < m13 < 5
satisfied
with
5,
satisfied
with
4,
common with 3, not
satisfied with 2, very
bad with 1
Weight reflects the importance of various indicators. In the multi-indicators
evaluation, a reasonable distribution of weights is a key of good assessment. Because
weight reflects people’s values and shows the concentration of people. The setting of
weight is very significant to the evaluation results. Because the sum of weight of all
indicators should equal 1; when the weight of one indicator is large, it means that
other indicators of the weight are smaller.
98
Overall, there are two methods to distribute weight, one is the subjective
weighting method and another is the objective weighting method. The subjective
weighting method depends on the experts or experienced persons to determine the
weighting.
This thesis is from the roles of users of TaoBao system to evaluate the credit
risk. The main users of TaoBao system are individual and small company, which
have the features of crowd wide, small turnover and frequent transactions. So using
the current empowerment through expert advice has more persuasive.
Delphi method has been a mature approach, which is a method of setting the
weight of indicator and based on experts in various fields after long-term experience
and practice to sum up the right way to empowerment indicators. It can rule out
many existing problems of the indicators weight impact. In the different applications,
it is giving the same indicators with a reasonable weight assignment.
However, with the continuous advancement of technology and credit
mechanisms, there will have a better ways to empower the credit evaluation
indicators.
99
5.4.4
Evaluation Model design
According to the research of questionnaire and the reference of Tian
Shuanglin (2009, 11), the weight of static indicators and dynamic indicators are
following the Table 5.4, Table 5.5 and Table 5.6:
Table 5.4: Initial Credit Evaluation Indicators Weight (Tian Shuanglin, 2009, 11)
Indicators
Weight
Age
0.1
Gender
0.15
Marital status
0.1
Educational level
0.1
Occupation
0.25
Salary
0.15
Bank deposit
0.15
Explain
Older people have higher
creditability than youth
Generally women have
higher creditability than
men
Married
better
than
unmarried, with children
is better than no children
The educational level has
the potential for get
fortune
A person who has a
relatively stable degree of
career
has
higher
credibility
Bank deposit is the form
and result of person’s
ability of creating wealth.
The salary can show the
ability to repay the debt
100
Table 5.5: Dynamic Credit Evaluation Indicators Weight for Seller
(Tian Shuanglin, 2009, 11)
Indicators
Weight
Explain
The buyer’s credit rating
Credit rating
0.1
can
show
the
credit
situation in degree
The
Turnover
0.3
turnover
important
is
very
elements
of
credit indicators.
The faster deliver can be
Time
0.15
more satisfied
The quality of goods with
Quality
0.15
the commitment of quality,
the closer is better
Better service has more
Service
0.15
satisfied
Price
0.15
The cheaper is better
Table 5.6: Dynamic Credit Evaluation Indicators Weight for Buyer
(Tian Shuanglin, 2009, 11)
Indicators
Weight
Explain
The seller’s credit rating
Credit rating
0.1
can
show
the
credit
situation in degree
The
Turnover
turnover
is
very
0.3
important
elements
of
101
credit indicators.
After
receive
commodity,
Time
the
the
faster
0.15
response can be more
satisfied
The quality of goods with
Satisfaction
0.15
the initial promise, the
closer the better
Better communication has
Communication
0.15
more satisfied
The price with the initial
Price
0.15
price, closer are more
satisfaction
Based on the Figure 5.2, we can separate the credit evaluation model in two
parts of original credit model and dynamic credit model.
5.4.4.1 Initial Credit Model Design
The first part is to establish an initial static credit evaluation. According to the
evaluating indicators from the registration and give the initial credit scores. The main
purposes of establishing the initial credit scores are: in order to provide an initial
credit value to users. When the user is the first time to transaction, if there is no
initial credit value to be a reference, the user cannot make a basic judgment. So,
102
initial credit evaluation prevents the fraud of once transaction. The second is having a
balancing effect. At present, same users of TaoBao system are according to the
fraudulent transactions to improve their credit rating. However, initial credit
evaluation, which is a stable factor, can effectively be used to limit this action. The
processes of initial credit evaluation are following these steps:
According to the user register information can get the evaluation of matrix R1,
which is based the above seven indicators (He Zhongxiong, 1999):
R1 =
v11
v12
v13
v14
v15
v16
v17
u11 u12
s11 s12
s21 s22
s31 s32
s41 s42
s51 s52
s61 s62
s71 s72
u13
s13
s23
s33
s43
s53
s63
s73
…
…
…
…
…
…
…
…
u1n
s1n
s2n
s3n
s4n = (suv )7×n
s5n
s6n
s7n
Formula (5-1)
Explain:
R1: it is the initial credit indicators evaluation matrix;
V1: it is the initial credit indicators set;
V1 = (v11, v12, v13, v14, v15, v16, v17) = (Age, Gender, Marital Status, Educational
Level, Occupation, Bank Deposit, Salary);
103
U1: it is the level domain of each indicator. The different indicator has different value
of n; the empty area is 0;
U1 = {u11, u12, u13 … u1n}……………………………………………..Formula (5-2)
According to the weight Table 5.4, can get the weight matrix W1:
W1 = (w11, w12, w13, w14, w15, w16, w17) = (0.1, 0.15, 0.1, 0.1, 0.25, 0.15, 0.15)
Formula (5-3)
Based on R1 and W1 can get the credit value for each static indicator:
(r11, r12, r13, r14, r15) = W1R1
= w11
w12
w13
w14
w15
w16
w17
s11
s21
s31
× s41
s51
s61
s71
s12
s22
s32
s42
s52
s62
s72
s13
s23
s33
s43
s53
s63
s73
s14
s24
s34
s44
s54
s64
s74
s15
s25
s35
s45
s55
s65
s75
Formula (5-4)
So, adding r11, r12, r13, r14, r15, can get the initial credit score of user:
5
S1 =
r1i = r11 + r12 + r13 + r14 + r15
i=1
…………………Formula (5-5)
104
5.4.4.2 Dynamic Credit Model Design
The second part is dynamic credit model, according to the user’s transaction
evaluate to calculate the credit for user; and it is show the last credit situation. This
model, which is similar to the initial credit evaluation model, can be separated in two
parts seller and buyer. The seller part is:
R2 =
v11
v12
v13
v14
v15
v16
u21
m11
m21
m31
m41
m51
m61
u22
m12
m22
m32
m42
m52
m62
u23 … u2n
m13 … m1n
m23 … m2n
m33 … m3n
m43 … m4n
m53 … m5n
m63 … m6n
= (muv )6×n
Formula (5-6)
Explain:
R2: the credit indicators evaluation matrix for once transaction;
V2: V2 = (v21, v22, v23, v24, v25, v26) = (Buyer credit rating, Turnover, Time, Quality,
Service, Price);
U2: U2 = (u21, u22, u23 … u2n) = (Very satisfied, satisfied, common, not satisfied,
very bad); according to the different indicators, the elements of U2 may have some
different, but mainly based on these five levels.
muv: is the value that according to the different indicators.
According to the Table 5.4, can get the weight matrix of dynamic indicators:
105
W2= (w21, w22, w23, w24, w25, w26) = (0.1, 0.3, 0.15, 0.15, 0.15, Formula (5-7)
0.15);
The credit dynamic evaluation is based on the user’s evaluation, after the
transaction, to get the credit scores for this transaction. The process is following:
(r21, r22, r23, r24, r25) = W2R2
= w21
w22
w23
w24
w25
w26
m11
m21
m31
× m
41
m51
m61
m12
m22
m32
m42
m52
m62
m13
m23
m33
m43
m53
m63
m14
m24
m34
m44
m54
m64
m15
m25
m35
m45
m55
m65
m16
m26
m36
m46
m56
m66
Formula (5-8)
So adding r21, r22, r23, r24, r25, can get the credit score of this transaction.
5
S2 =
r2i = r21 + r22 + r23 + r24 + r25
………………Formula (5-9)
i=1
The buyer part is similar to the seller’s:
R3 =
v11
v12
v13
v14
v15
v16
u31
m11
m21
m31
m41
m51
m61
u32
m12
m22
m32
m42
m52
m62
u33 … u3n
m13 … m1n
m23 … m2n
m33 … m3n
m43 … m4n
m53 … m5n
m63 … m6n
= (muv )6×n
Explain:
R3: the credit indicators evaluation matrix for once transaction;
Formula (5-10)
106
V3: V3 = (v31, v32, v33, v34, v35, v36) = (Seller credit rating, Turnover, Time,
Satisfaction, Communication, Price);
U3: U3 = (u31, u32, u33 … u3n) = (Very satisfied, satisfied, common, not satisfied, very
bad); according to the different indicators, the elements of U2 may have some
different, but mainly based on these five levels.
muv: is the value that according to the different indicators.
According to the Table 5.5, can get the weight matrix of dynamic indicators:
W3= (w31, w32, w33, w34, w35, w36) = (0.1, 0.3, 0.15, 0.15, 0.15, 0.15)....Formula (5-11)
The credit dynamic evaluation is based on the user’s evaluation, after the
transaction, to get the credit scores for this transaction. The process is following:
(r31, r32, r33, r34, r35) = W3R3
= w31
w32
w33
w34
w35
w36
m11
m21
m31
× m
41
m51
m61
m12
m22
m32
m42
m52
m62
m13
m23
m33
m43
m53
m63
m14
m24
m34
m44
m54
m64
m15
m25
m35
m45
m55
m65
m16
m26
m36
m46
m56
m66
Formula (5-12)
So adding r21, r22, r23, r24, r25, can get the credit score of this transaction:
S3 =
5
i=1 r3i
= r31 + r32 + r33 + r34 + r35 ……………………………..Formula (5-13)
Combining the scores of initial credit evaluation and dynamic credit
evaluation can get two composite credit scores:
107
i.
The credit scores of past three months
According to the formula 5-9 and formula 5-13, can get the formula:
S0 =
n
i=1 S2i
× W2 ………………………………………………Formula (5-14)
S0 =
n
i=1 S3i
× W3 ……………………………………………...Formula (5-15)
Explain:
S0: the total credit scores of the past three months;
Sai: the credit dynamic scores for once transaction;
n: the total times of transaction in past three months;
ii.
The history credit scores
According to the sum of credit scores, calculate the history credit scores:
S=
m
i=1 S0i
……………………………………………………….Formula (5-15)
Explain:
S: it is the user’s credit scores of all the transaction. It is reflecting the user’s
credit situation of doing transaction;
m: it is the times of total credit evaluation.
108
5.5 Summary
In this chapter, it mainly carried out the whole process of new credit
evaluation system of TaoBao company. In the authentication part, author suggests
using multi-authentication mechanism to confirm seller’s and buyer’s identity. In
addition, author selected initial and dynamic credit indicators, which can show the
basic information and the credit situation of users; and based on those indicators to
support the mechanism of granting initial credit and dynamic credit. There into,
author used fuzzy method to calculate user’s dynamic credit by combined author’s
own research and expert’s opinion. Finally, author carried out the whole system
design and supports to the system development, called chapter 6.
109
CHAPTER 6
SYSTEM DEVELOPMENT AND TESTING
6.1
Introduction
The main goals of this chapter are develop the system that based on the above
fuzzy credit evaluation model. This system has two main modules, which called
authentication module and credit evaluation module. An authentication module is
using the multi-authentication mechanism to make user’s information more trustable.
It has two functions. At the first is multi-authentication part. In this part, the main
purpose is that when user doing authentication, it can be combined with banking
system to ensure the authenticity of filled banking information. Here is based on the
TaoBao’s current authentication system and simply use the MySQL database to
simulate the banking system to confirm the user’s input information. Another part is
buyer authentication. It is same with seller’s authentication part and based
multi-authentication mechanism. A credit evaluation module can be divided to the
static credit evaluation module, also called user initial credit module, and dynamic
credit evaluation module.
110
The software that used to develop this system are Dreamwear 8.0, which uses
PHP, JAVA Script and HTML program to build a web-based system and user
interface.
6.2
Database Creation
Database is used to store information and can be built by many ways. In this
thesis, a database is built by MySQL software. One of the function in this system is
simulating the bank system that includes user’s bank account information, such as
user’s real name, account number, identity number and the deposit, show in Figure
6.1. All the information of bank system table are fixed, just like the real bank system
that cannot be changed by user. When a user is trying to authenticate in this
simulated TaoBao system, the information of user filled should be exactly same with
bank system. Otherwise, they cannot pass the authentication.
111
Figure 6.1: Database of Bank System
In addition, other tables are about user’s credit information, which include the
user information table, buyer credit table and seller credit table. The user information
table stores the basic user information, such as name, user id, password, address and
phone number, show in Figure 6.2, and user’s initial credit scores, which is depended
on the user’s age, gender, salary, education, marry status, bank deposit and career,
show in Figure 6.3.
112
Figure 6.2: Database of User’s Initial Credit
Figure 6.3: Database of User’s Initial Credit
113
Buyer credit table, in Figure 6.4, and seller credit table, in Figure 6.5,
shows user’s dynamic credit scores, which includes historic scores and total score
that shows for each evaluation.
Figure 6.4: Database of Buyer’s Dynamic Credit
114
Figure 6.5: Database of Seller’s Dynamic Credit
6.3
Authentication Module
Once a user login the system with user name of buyer, the user interface
will bring out. This page shows that the user is not pass the authentication yet, and
supports a hyperlink to authenticate, shows in Figure 6.6. An authentication module,
shows in Figure 6.7, supports user fill their personal information of real name, bank
account and ID number, which should exactly same with the dummy bank system
database. Once user fills all the information, it will compare with the bank_system
database. If all the information is correct, user can pass the authentication and go to
115
next step, user initial credit part. For checking the correction of information, a “LIKE”
MySQL command should be used.
Figure 6.6: Buyer’s Interface of Before Authentication
116
Figure 6.7: Authentication Interface
6.4
Credit Evaluation Module
The credit evaluation module has two parts. One is the original credit
module that supports the interface to get the user the basic information, which
includes age, gender, career, education level, marry status, salary and deposit, to
calculate user’s initial credit. Another is dynamic credit module, to get the evaluation
chooses and calculate the dynamic credit for user. If a user login with the buyer user
id, he or she has to pass the authentication to do the evaluation, otherwise user login
with the seller id, which assumes already pass the authentication and get the initial
credit. So the seller id can director to see there is a evaluation waiting to do.
117
6.4.1
Original Credit Module
The original credit module supports the platform to get the information of
user and bases on those information to calculate the user’s initial credit by system
automatic, in Figure 6.8. The information includes age, gender, career, education
level, marry status, salary and deposit. The deposit information is get from the
bank_system database, need pass the authentication process; age and other
information are filled by user.
When user fill all the information and click submit, an initial credit will be
calculated and show in the user interface web page. In addition, a dummy transaction
has already be done and waiting user to evaluate. The hyperlink of credit evaluation
will be showed in user interface web page, show in Figure 6.9.
Figure 6.8: Granting Initial Credit Interface
118
Figure 6.9: Buyer’s Interface of After Authentication
6.4.2
Dynamic Credit Module
The dynamic credit module can be divided in two web pages. One is for
the buyer evaluation, another is for the seller evaluation. There are four elements of
seller and chosen by buyer with radio button, which are deliver time, quality, service
and price, show in the Figure 6.10. Buyer can image the transaction situation and
give the evaluation to seller and the result will be calculated by the system automatic;
except the four elements that decided by buyer, there two elements of turnover and
buyer current credit rank also will be included. Once buyer finish the evaluation, a
web page of showing seller’s credit will be brought out, which shows seller’s initial
119
credit, total credit and the credit of the last evaluation, show in the Figure 6.11.
Figure 6.10: Buyer Evaluation Interface
120
Figure 6.11: Interface of Seller’s Credit
In the seller evaluation part, it is similar with the buyer evaluation part.
Only the elements of evaluation are different, which are time, satisfaction,
communication and price; and the results will show on the buyer credit interface,
show in the Figure 6.12 and Figure 6.13.
121
Figure 6.12: Seller Evaluation Interface
Figure 6.13: Interface of Buyer’s Credit
122
6.5
User Satisfaction Test
Because the users of TaoBao system still not familiar with the new credit
evaluation module, so simply using the questionnaire to get the satisfaction and
feedback from the TaoBao system user.
This questionnaire focus on the comparing of the new evaluation system
and the current credit system to confirm the satisfaction of TaoBao users. Totally,
sixteen persons have answered this questionnaire.
6.5.1
New User’s Credit Situation Test
The first question is related to the authentication part, including the
granting of initial credit. The authentication part and the initial credit part is trying to
improve the credit situation of new users in TaoBao system. The result showed in
Figure 6.14.
123
70%
60%
50%
40%
30%
20%
10%
0%
Trust New User
No Idea
Don't Trust New User
Figure 6.14: Credit Situation of New User Test
From the result, there are more than 60% people think the new user’s credit
situation can be improved and they would like to trade with them. One of the
objective is to ensure the user’s identity; it must be a true and unique identity.
Another objective is to give a standard to judge the new user’s credit situation, then
improve the credit situation of new user. Based on this result, because the improved
authentication mechanism and new user’s initial credit, comparing the above
questionnaire in chapter 4, almost people think the new user will be more trustable
than the current system.
6.5.2
User’s Dynamic Credit Evaluation Test
This question is related to test the user’s dynamic credit evaluation module.
It is to test whether the new dynamic credit evaluation module can clearly show
124
TaoBao system user’s dynamic credit. The result showed in Table 6.1.
Table 6.1: User’s Dynamic Credit Evaluation Test
Items
Quantity
Percentage
Strong disagree
0
0%
Disagree
1
6%
Neutrally
4
25%
Agree
10
62.5%
Strong agree
1
6.25%
The result shows near 69% people think the new credit evaluation system
can be good to show TaoBao system user’s dynamic credit situation. The main
objective of this thesis is to improve the credit evaluation mechanism, making the
credit evaluation can show more user’s credit details for other users consult and
judge. The result is gratified, almost people think the new credit evaluation
mechanism can show user’s credit in more details. That means, TaoBao system users
can base on those details to analyze others credit situation in more trading aspect,
then reducing the transaction risk.
125
6.5.3
System Integrative Test
This question is to compare the current TaoBao credit evaluation system
and the improved credit evaluation system to get the general testing of user’s
satisfaction. The result shows in Table 6.2.
Table 6.2: System Integrative Test
Items
Quantity
Percentage
Strong disagree
0
0%
Disagree
0
0%
Neutrally
3
19%
Agree
13
81%
Strong agree
0
0%
The question is asking TaoBao system user: whether the new evaluation
system can reduce the credit risk of TaoBao system? Based on the result can know
that more than 80% people think the new credit evaluation mechanism can reduce the
credit risk of TaoBao system. Form the result, there are several problems of TaoBao
system to be improved and get the satisfaction of TaoBao system users.
126
6.6
Summary
Overall, chapter 6 has been developing the rough credit evaluation system
of TaoBao to show the new evaluation mechanism how to work. Through the
authentication module to credit evaluation module, using a dummy buyer identity to
describe the whole process of the new evaluation system.
In addition, the user satisfaction test part carried out and analyzed the
feedback from TaoBao users. Based on those results, it ensures the advantage of the
new credit evaluation system.
127
CHAPTER 7
DISCUSSION AND CONCLUSION
7.1
Introduction
This chapter will look for the achievement and outcome of this project; in
addition, describe the future blueprint of the continue research. It will conclusion the
whole chapter of this thesis to show the overall achievement; and also in this chapter
a final result of improving the TaoBao credit evaluation system will be carried out to
show the whole process and the mechanism of new credit evaluation system.
However, the future research still need to be considered and will be discussed in this
chapter.
128
7.2
Overall Achievement
The credit risk is the main issue that hinders the development of
e-commerce in good health, and this issue is even more outstanding in C2C
e-commerce. In this project, detailed analysis the credit risk problems in the
e-commerce, and the main reason of the existence of credit risk problem in C2C
e-commerce. Therefore, in order to be able to reduce the credit risk, it urgent to
improve the current credit evaluation system.
Most of the C2C e-commerce sites have been established its own credit
evaluation system, in a certain extent, reducing the credit risk. However there are
also many shortcomings. This project used the case study to analyze the general
current credit evaluation system and point out the main problems of TaoBao’s current
credit evaluation system. Focusing the exiting problems, this thesis gave the opinion
to improve the credit evaluation system and design the rough system to show how
the new credit evaluation system work.
At the first, in the “introduction” chapter, mainly introducing the current
situation of e-commerce credit risk, found out the problem and gave the objectives.
Second, the “literature review” chapter, introduced the concepts of
e-commerce and e-commerce credit risk, and then analyzed the characteristics of
online-trading and focused on the C2C online-trading, found the reasons of leading
C2C credit risk, and used case study pointing out the problems of credit evaluation
system.
129
Third, in the “methodology” chapter, laid out the way to do this project and
analyzed the environment of software and hardware that should be supporting this
project.
Fourth, the “data analysis” chapter mainly analyzed TaoBao’s current
credit evaluation system, used questionnaire to found out the limitations of current
system and the opinions of improving the current system from the user of TaoBao
system, then gave primary solutions to improve the current system.
Fifth, the “system design” chapter based on the data analysis result from
chapter 4 to design each module of improved credit evaluation process. It designed
the multi-authentication module for supporting user’s identity authentication, which
can improve the current authentication mechanism and make user’s identity more
reliable. Then it based on the selected initial credit indicators and dynamic credit
indicators to design the user’s initial credit granting module, which improves the new
user’s credit situation, and the dynamic credit evaluation mechanism, which
improves the user’s dynamic credit to show more details that makes user’s credit
situation clearer.
Sixth, the “system development and testing” chapter focus on the system
develop, based on the designed system process and mechanism, to rough show the
new system how to work. Then it gets the feedback from the user of TaoBao system
by using questionnaire. The result of user’s satisfaction test is gratified to show most
users of TaoBao system approve the new credit evaluation process and mechanism.
130
The improved credit evaluation system has the following advantages:
i.
This system will in large part to improve the current situation of authentication
of identity, to ensure the roots of credit evaluation is reliable. Furthermore,
effectively improve the current system to prevent the cheating of credit ranks
that using bugs.
ii.
Improving credit situation of new users, on other hands, it is make the
competitive of new users that can attract more personal business transactions.
iii.
Improving the credit evaluation system into detail and making the credit
evaluation more obvious.
7.3
Outcome
The final outcome of this project can be showed in a framework, which
combines the user authentication part, initial credit granting part and the dynamic
credit evaluation part, it is showed in Figure 7.1. It is also including the indicators
selected and the fuzzy evaluation mechanism.
131
User Identity Authentication
Seller Identity Multi-Authentication
Buyer Identity Multi-Authentication
Credit Indicators Choose
Static Credit Indicators
Dynamic Credit Indicators
Build Credit Indicators Evaluation Structure
Set Scoring function of each credit
indicators by using fuzzy method
Using Expert Method to determine weight
of each indicators
Integrated Evaluation Credit Indicators
Initial Credit Evaluation
Dynamic Credit Evaluation for Once
Transaction
Credit Evaluation Scores
Figure 7.1: The New Process of Identity Authentication and Credit Evaluation of
TaoBao System
7.4
Recommendation and Future Research
The credit risk is a new research topic in C2C. Currently, excepting a few
technology research, almost relate researches are in the initial layer. Author has done
132
some exploratory work. However, because the limited of knowledge and lack enough
time and effort, a number of specific issues have not solution yet. In this thesis, there
are some lacks of better understanding of the complexity and profundity of the issues
of credit, and combining the sociology and psychology of credit in the E-commerce
management still has a long way to go; it is waiting the further study.
i.
As the continuous improvement of credit mechanism of e-commerce, to
establish a unified platform for credit management will be reducing the credit
risk of e-commerce. Credit management theory research needs further
improving.
ii.
To establish the uniform credit evaluation mechanism is the core of protecting
credit risk. However, to establish a good credit evaluation system needs storage
technology, government, banks and various E-commerce sites to work together
to complete. This thesis suggested some improvements of credit evaluation
mechanism and design a framework of credit evaluation system. However, there
still have many factors not considered comprehensive, need for further study,
such as further design the credit evaluation system and improving the credit
evaluation mechanism.
iii.
Although, using the fuzzy method to establish the credit evaluation model can
solve some problems of the current credit evaluation system of TaoBao, but
following the development of e-commerce, there are many new issues will be
came out. Therefore, it should base on those new issues to improve the credit
evaluation model.
In general, due to the C2C is a new business model, its development is still
not mature, author cannot solve all the problems of TaoBao system. Credit is an
importance factor that baffling the development of C2C. In the research of credit of
133
C2C, not only the credit evaluation mechanism should be completed, but also the
theoretical study has to be further in-depth.
7.5
Chapter Summary
This chapter concludes all the chapters of the thesis, from chapter 1 to
chapter 6. It focus on the achievement and outcome of the research; and also fingers
out the lacks of the research and gives the suggestion of the further research.
134
REFERENCES
Boehm B (1988), "A Spiral Model of Software Development and Enhancement",
"Computer", "IEEE", 21(5):61-72.
Bian Xiaohong. (2006) “The research of online company credit risk analysis base on
fuzzy method”. Working paper.
Chen Zhihao. (2003) The analysis of e-commerce development factors〔J〕.economic
management.
Che Chun. (2005) “The research of e-commerce credit risk base B2B model”. Xi An
electronic technology university.
Cheltenham Courseware Pty. Ltd. (2008) Microsoft project 2007, manual foundation
level.
Chi Guotai, Xu wen, Sun Xiufeng. (2006) “The research of individual credit card
risk evaluation structure and model”. 34:4.
China Network Information Center (CNNIC). (2004) The fifteenth research report.
China Network Information Center (CNNIC). (2006) The eighteenth research report.
China Network Information Center (CNNIC). (2009) The twentieth research report.
Definition adapted and expanded from Emmanuel Lallana, Rudy Quimbo, Zorayda
Ruth Andam, ePrimer (2000): An Introduction to E-Commerce.
David Braun, Jeff Sivils, Alex Shapiro, Jerry Versteegh, (2000) Unified Modeling
Language (UML) Tutorial.
Dorre, R. (1987) Taking Japan seriously, Stanford university, stanford.
Doney, P.M. and Cannon, J.P. (1997) An examination of the nature of trust in
135
buyer-seller relationships. Journal of Marketing. 35-51.
Fei Jianping. (2009) The research of e-commerce credit risk impact of online-trading
activities mechanism. 31.0053:f830:A.
Fukuyama.F. (1995) Trust: The social virtues and the creation of prospeity. NewYork:
Free Press.
Fombrun,C. and Shanley, M. What's in a name? Ruputation building and corporate
strategy. Academy of management journal. 233-258.
Fung, R., and Lee, M. (1999) EC-Trust: Exploring the antecedent factors,
Proceedings of the 5 Americas Conference on Information Systems.
Goldman Sachs Investment Research, November (1999). “E-commerce/Internet:
B2B:
Gefen, D. (2000) E-commerce: The role of familiarity and trust.Omega: The
international Journal of management science, 28. 725-737.2B or Not 2B?” v.
1.1, 16, 68-71.
He Zhongxiong. (1985) Fuzzy method and application [M], TianJin University
publish.
Herbig, p. Milewicz, J. (1994) A model of reputation building and destruction.
Journal of Business Research, 31. 23-31.
IEEE Standard Computer Dictionary (1990): A Compilation of IEEE Standard
Computer Glossaries; IEEE; New York, NY.
Jin Xiaohong. (2007) The Study of Electronic Business Credit Evaluation Mode , 4.
Jiang Lingming. (2003) The current situation Search of China E-commerce. Business
Research. 38.
Kollock, P. (1999) The production of trust in online markets. pp. 99-123.
Kotkin,J. (1998) The mother of all malls. Forbes. 60-65.
LuHongyan, WeiXin. (2007) C2C network market credit system comparing [J].
Liu Youcai. (1996) Principle and Design of Fuzzy Expert system. BeiJing:
Publishing house of beijing university of aeronautics and astronautics.
Li Huibing, Yang Dongxue. (2000) Social economic and business.
136
Liao Chenglin. (2003) The mechanism of credit lose. 7-9.
Resnick.P, and Zeckhauser,R. (2001) Trust Among Strangers in Internet
Transactions:Empirical Analysis of eBay's Reputation System.
NJ. Retrieved. (2007) "Wordreference.com: WordNet 2.0". Princeton University.
ShaoBingjia, LiRu. (2006) Online-trading websites credit evaluation system
investigation [J].
Stewart, K.J. (1999) Transference as a means of building trust in world web sites.
The proceedings of ICIS. 12-15.
Swan, J.E. and Nolan, J.J. (1985) Gaining customer trust: A conceptual guide for the
salesperson. Journal of personal selling&sales management. 39-48.
Shi Liangping. (2003) Business bank credit risk management [J]. 15-16.
The Eighteenth Research Report from China Network Information Centre (CNNIC),
2006.
Tadelis. (2001) The Market for Reputation as an incentive mechanism. Journal of
Political Economy.
Tian Shuanglin. (2009) The weight setting of credit rating indicators structure [J].
12-14.
Zeng Yong. (2004) The research of E-commerce credit evaluation. Wu Han
university.
ZhuXiaozhong, ZhangZongyi, GenHuadan. (2004) Modern credit risk model
analysis [J]. financial research. 33-46
Zorayda Ruth Andam. (2003) E-commerce and e-business book.
Zhao Xiaodong, Zhen Tao, (2003) personal credit rank evaluation model based
fuzzy-AHP evaluation method [J], 6.
Zhang Huifang. (2002) The development of E-commerce in three parts. 33-36.
Zhen Wenhong, Li Jian. (2000) The impact of credit of market. economic
management.2
Wei Mingxia. (2001) The analysis of marketing game. 4
Whitmeyer, Joseph M., (2000) Effect of positive peputation systems. Social science
137
research. 188-207.
Wen Xing. (2007) Information technology and online-trading. NanJing university.
WEBSITES:
http://www.tab.fzk.de/en/projekt/skizze/e-commerce.htm; Internet; accessed 2009
September.
http://hi.baidu.com/chuhan1234/blog/item/9fc2f54f7b25243eaec3abe1.html Internet;
accessed 2009 September.
http://searchcio-midmarket.techtarget.com/sDefinition/0,,sid183_gci755441,00.html.
Internet; accessed 2009 September.
http://www.doh.state.fl.us/irm/Apps/sdstandards/isdm/Design.pdf. Internet; accessed
2009 September.
http://www.uml.org.cn/Test/200908206.asp. Internet; accessed 2009 September.
http://atlas.kennesaw.edu/~dbraun/csis4650/A&D/UML_tutorial/what_is_uml.htm;
Internet; accessed 2009 September.
http://www.nb-020.com/News_Detail.asp?id=30. Internet; accessed 2009 September.
http://www.softwaretestingclub.com/forum/topics/is-integration-a-phase.
Internet;
accessed 2009 September.
http://www.traderinasia.com/classifieds.html. Internet; accessed 2009 September.
TA Project, (2002) http://www.tab.fzk.de/en/projekt/skizze/ecommerce.htm; Internet;
accessed 2009 September.
Traderinasia.com; (2002) http://www.traderinasia.com/classifieds.html; Internet.
wikipedia.com. http://en.wikipedia.org/wiki/Methodology. Internet; accessed 2009
September.
wikipedia.com.http://en.wikipedia.org/wiki/Systems_Development_Life_Cycle#Histo
ry. Internet; accessed 2009 September.
wikipedia.com. http://en.wikipedia.org/wiki/Waterfall_model.
Internet;
accessed
2009 September.
wikipedia.com. http://en.wikipedia.org/wiki/Spiral_model. Internet; accessed 2009
138
September.
wikipedia.com.http://en.wikipedia.org/wiki/Systems_Development_Life_Cycle#Build
_or_coding. Internet; accessed 2009 September.
wikipedia.com. http://en.wikipedia.org/wiki/Unit_testing. Internet; accessed 2009
September.
wikipedia.com.http://en.wikipedia.org/wiki/Systems_Development_Life_Cycle#Build
_or_coding. Internet; accessed 2009 September.
wikipedia.com.
http://en.wikipedia.org/wiki/Microsoft_Office_2007.
Internet;
accessed 2009 September.
wikipedia.com. http://en.wikipedia.org/wiki/Adobe_Dreamweaver. Internet; accessed
2009 September.
wikipedia.com.
http://en.wikipedia.org/wiki/MySQL#Platforms_and_interfaces.
Internet; accessed 2009 September.
139
APPENDIX A
QUESTIONNAIRE
140
UNIVERSITI
TEKNOLOGI MALAYSIA
IT MANAGEMENT
Questionnaire form
Credit System of TaoBao Company
OBJECTIVE
The purpose of this questionnaire is to Collect the opinion of current TaoBao credit
evaluation system from the TaoBao system users and confirm the indicators selection
for the improved credit system.
Section A General Information Question
Age :
□ 18-24
Your Role in TaoBao :
□ Buyer
□ 25-30
□ Above 31
□ Seller
Frequency of using TaoBao :
□ Every day
□ At least once in a week
□ Around one month
141
Section B: Situation of the Current TaoBao Credit System
1.
□
□
□
□
□
What do you think of the current TaoBao credit evaluation system?
User’s credit rank is not trustable
User’s credit is too general to understand
I have no idea
User’s credit can be presented in some degree
User’s credit can be presented very clear
2. Have you faced the problems of credit evaluation of TaoBao system?
□ Default good evaluation that impact the credit real situation
□ Hard distinguish truly credit and illusive credit
□ Evaluated by others with untrue and malice reason
□ Forced to give a good evaluate by others bully
□ Have to give a good evaluate because no big reasons
□ The problems appeared after evaluation
□ The credit rank is not fair, when doing a large turnover transaction
Others________________________
3. Do you check other’s credit situation before you trade with them?
□ Never
□ Sometimes
□ All the time
Section C: Question of Authentication Part of TaoBao System
4.
□
□
□
□
□
What do you think the current authentication of TaoBao system?
It is too simple to authenticate the identity
It cannot be good to ensure each people just have one account
I have no idea
It can ensure the identity of most user
It can make sure each people only have one account
5.
□
□
□
Do you think buyer need to pass the authentication?
No, because their identity is not important
I have no idea
Yes, otherwise they will make the evaluation unresponsively
6.
□
□
□
Do you trust the new member?
No, please put reason___________________
I have no idea
Yes, please put reason___________________
142
Section D: Question of Initial and Dynamic Indicators Select
7. What do you think the current credit evaluation system of TaoBao, using positive,
normal, negative?
□ It cannot show the user’s credit situation
□ It is not good to show the details of user’s credit
□ I have no idea
□ It can show the situation of user’s credit in some degree
□ It definitely can show the situation of user’s credi
Please put mark for below questions, 1 is disagree, Scale of Agreement
5 is agree
1
2
3
4
Part 1: Dynamic indicators:
5
8. Do you think the turnover during the credit
evaluation is important?
□
□
□
□
□
9. Do you think the user’s credit rank is important for
the evaluation?
□
□
□
□
□
 Commodity Deliver Time
□
□
□
□
□
 Product Quality
□
□
□
□
□
 Seller service
□
□
□
□
□
 Credit rating
□
□
□
□
□
 Price
□
□
□
□
□
 Credit rating
□
□
□
□
□
 Commodity Receive Time
□
□
□
□
□
 Satisfaction degree of commodity
□
□
□
□
□
 Communication
□
□
□
□
□
 Price
□
□
□
□
□
10. To be a buyer, which indicators of seller you will
concern?
11. To be a seller, which indicators of buyer you will
concern?
143
Part 2: Static indicators
Scale of Agreement
1
2
3
4
5
12. According to your experience which age is more
trustable?
 Children (age<16)
□
□
□
□
□
 Youngster (16<age<30)
□
□
□
□
□
 Midlife (31<age<55)
□
□
□
□
□
 Senior (55<age<70)
□
□
□
□
□
 Old people(age>70)
□
□
□
□
□
 Male
□
□
□
□
□
 Female
□
□
□
□
□
 doctor and higher levels
□
□
□
□
□
 master
□
□
□
□
□
 bachelor
□
□
□
□
□
 college
□
□
□
□
□
 middle school
□
□
□
□
□
 primary
□
□
□
□
□
 others
□
□
□
□
□
Married with children
□
□
□
□
□
Married with no children
□
□
□
□
□
Unmarried
□
□
□
□
□
divorced with children
□
□
□
□
□
divorced with no children
□
□
□
□
□
13. According to your experience which sex is more
trustable?
14. According to your experience the people in which
education level has high credit?
15. According to your experience which marry status
is more trustable?
144
16. what do you think that which career is more trustable? (multi-select)
□Teacher
□Public Clerk
□Doctor
□Lawyers
□Finance
□Soldiers
□Power
□Reporter
□Post
□Business owners
□Telecom
□Employee
□Others
17. What do you think that which salary level can be trusted?
□1000
□2000
□5000
□10000
Thanks for your time. Any comment please tell us:
145
UNIVERSITI
TEKNOLOGI MALAYSIA
IT MANAGEMENT
Questionnaire form
Credit System of TaoBao Company
OBJECTIVE
The purpose of this questionnaire is to test the improved TaoBao credit evaluation
system and get the feedback from TaoBao users to compare with the current TaoBao
credit evaluation system.
Questionnaire of system testing
1. According to the initial credit, do you think the new user’s credit situation is better
than the current system of TaoBao?
□ Yes, because it clear shows the background of new user, I can trust them
□ I have no idea
□ No, I still cannot trust them
Please put mark for below questions, 1 is strong
disagree, 5 is strong agree
Scale of Agreement
1
2
3
2. Do you think whether the new system can reflect
user’s credit situation?
3. Whether the new credit evaluation system can
reduce the credit risk, when you using TaoBao
system?
Thanks for your time again, also appreciate to give me your suggestion:
4
5