An Automatic Classification Approach to Business Stakeholder Analysis on the Web Wingyan Chung, Hsinchun Chen, Edna O. F. Reid January 16, 2003 Agenda • • • • • • • Introduction Literature Review Research Questions Research Approach and Testbed Evaluation Methodology Experimental Results and Discussion Conclusions and Future Directions 2 Introduction Current Business Environment • Networked business environment facilitates information sharing • Collaborative commerce integrates business processes among partners through electronic sharing of information – Sales support, vendor management, planning and scheduling, demand planning, etc. • Knowledge sharing about stakeholder relationships through a company’s Web sites and pages – Textual content or annotated hyperlinks 4 Problems • Information overload on the Web – Hinders analysis of stakeholder relationships • Knowledge hidden in interconnected Web resources – Posing challenges to identifying and classifying various business stakeholders • e.g., A company’s manager may not know who are using their company’s Web resources – Problem of traditional stakeholder analysis – The emergence of electronic commerce 5 An Automatic Classification Approach • Need better approaches to uncovering such knowledge – Enhance understanding of business stakeholders – Enhance understanding of competitive environments • We propose an automatic classification approach to business stakeholder analysis – Human knowledge + machine-learned information • We will review related areas in stakeholder analysis and Web page classification techniques 6 Literature Review Stakeholder Analysis • Stakeholder theories evolve over time while the view of firm changes – Production view (19th century): Suppliers and Customers – Managerial view (20th century): + Owners, Employees – Stakeholder view (1960-80s) (Freeman, 1984): + Competitors, Governments, News Media, Environmentalists, … – E-commerce view (1990s - now): + International partners, Online communities, Multinational employees, … 8 Summary of stakeholder types Research Stakeholder Types Reid, 2003 Partners/suppliers, customer, employee, investor, education institutions, media, portal, public, recruiter, reviewer, competitor, unknown Elias & Cavana, 2000 Owners, community, unions, employees, government, consumer advocates, competitors, financial community, media, customers, SIG, suppliers Agle et al., 1999 Shareholders, employees, customers, government, communities Donaldson & Preston, 1995 Investors, government, suppliers, trade associations, employees, communities, customers, political groups Clarkson, 1995 Employees, shareholders, customers, suppliers, public stakeholders • These types, ordered by their relevance to those appearing on the Web, are important for practical understanding of stakeholders of firms 9 Comparing Stakeholder Types* Used Research P E C S U M G R V O T F I N Reid, 2003 Elias & Cavana, 2000 Agle et al., 1999 Donaldson & Preston, 1995 Clarkson, 1995 * P = Partners/suppliers, E = Employees/Unions, C = Customers, S = Shareholders/investors, U = Education/research institutions, M=Media/Portals, G = Public/government, R = Recruiters, V = Reviewers, O = Competitors, T = Trade associations, F = Financial institutions, I = Political groups, N = SIG/Communities (Note that a class “Unknown” is not included here) 10 Comments on Stakeholder Research • Strong explanatory power but are weak at practical classification of stakeholders • Conclusions drawn from old data • Previous research rarely considers the many opportunities offered by the Web for stakeholder analysis, e.g., – Business intelligence, which is obtained from the business environment, is likely to help in stakeholder activities – Tools have been developed to exploit business intelligence but not yet applied to stakeholder analysis 11 BI and Stakeholder Analysis • Advanced BI tools often rely on Web mining techniques to discover patterns on the Web automatically (Etzioni 1996; Kosala & Blockeel 2000), e.g., – PageRank (Brin & Page 1998), HITS (Kleinberg 1999), Web IF (Ingwersen 1998) – External links mirror social communication phenomena (e.g., stakeholder relationships) • Tools and approaches exploit Web content and link structure information – Ong et al 2001; Tan et al. 2002; Reiterer et al. 2000; Chung et al. 2003; Reid 2003; Byrne 2003 12 Information on the Web • Structural and textual content • But commercial BI tools lack analysis capability (Fuld et al. 2002) • Need to automate stakeholder classification, a primary step in stakeholder analysis – Automatic classification of Web pages is a promising way to alleviate the problem 13 Web Page Classification • The process of assigning pages to predefined categories – Helps to discover companies’ stakeholders on the Web and enables companies to understand the competitive environment better • Major approaches include k-nearest neighbor, neural network, Support Vector Machines, and Naïve Bayesian network (Chen & Chau 2004) • Previous work – Kwon and Lee 2003; Mladenic 1998; Furnkranz 1999; Lee et al. 2002; Glover et al. 2002 14 Feature selection in Web Page Classification • Features considered – Page textual content: full text, page title, headings – Link related textual content: anchor text, extended anchor text, URL strings – Page structural information: #words, #page outlinks, inbound outlinks (i.e., links that point to its own company), outbound outlinks (i.e., links that point to external Web site) • Methods for selection – Human judgment / Use of domain lexicon – Feature ratios and thresholding – Frequency counting / MI 15 Research Questions Research Gaps • Stakeholder research provides rich theoretical background but rarely considers the tremendous opportunities offered by the Web for stakeholder analysis – Conclusions drawn from old data may not reflect rapid development in e-commerce • Existing BI tools lack stakeholder analysis capability • Automatic Web page classification techniques are well developed but have not yet been applied to business stakeholder classification 17 Research Questions • How can we develop an automated approach to business stakeholder analysis on the Web? • How can Web page textual content and structural information be used in such an approach? • What are the effectiveness (measured by accuracy) and efficiency (measured by time requirement) of such an approach for business stakeholder classification on the Web? 18 Research Approach and Testbed Automatic Classification Approach • Purpose: To automatically classify the stakeholders of businesses on the Web in order to facilitate stakeholder analysis • Rationale – Business stakeholders should have identifiable clues that can be used to distinguish their types – The Web content and structural information is important for understanding the clues for stakeholder classification • Two generic steps: – Creation of a domain lexicon that contains key textual attributes for identifying stakeholders – Automatic classification of Web pages (stakeholders) linking to selected companies based on textual and structural content of Web pages 20 Building a Research Testbed • Business stakeholders of the KM World top 100 KM companies (McKellar 2003) • Used backlink search function of the Google search engine to search for Web pages having hyperlinks pointing to the companies’ Web sites • For each host company, we considered only the first 100 results returned – Removed self links and extra links from same sites – After filtering, we obtained 3,713 results in total – Randomly selected the results of 9 companies as training examples (414 283 pages stored in DB) 21 Creation of a Domain Lexicon • Manually read through all the Web pages of the nine companies’ business stakeholders to identify one-, two-, and three-word terms that were indicative of business stakeholder types • Extracted a total of 329 terms (67 one-word terms, 84 two-word terms, and 178 three-word terms), e.g., 22 Automatic Stakeholder Classification • Three steps: Manual Tagging Feature selection Automatic classification 23 Manual tagging Feature selection Automatic classification Manual Tagging • Manually classified each of the stakeholder pages of the nine selected companies into one of the 11 stakeholder types (based on our review on slides 9-10) 24 Manual tagging Feature selection Automatic classification Feature Selection • Structural content features: binary variables indicating whether certain lexicon terms are present in the structural content – A term could be a one-, two-, or three-word long – Considered occurrences in title, extended anchor text, and full text • Textual content features: frequencies of occurrences of the extracted features – The first set of features was selected based on human knowledge, while the second was selected based on statistical aggregation, thereby combining both kinds of knowledge 25 An Example (a media type) Link to the host company (ClearForest) <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" /> HTML hyperlink and extended anchor text <title>David Schatsky: Search and Discovery in the Post-Cold War Era</title> ... <p>I just saw a demo by <a href = "http://www.clearforest.com"> ClearForest, </a> a company that provides tools for analyzing unstructured textual information. It's truly amazing, and truly the search tool for the post-Cold War era. ... </p> ... </body> </html> 26 Manual tagging Feature selection Automatic classification Automatic Classification • A feedforward/backpropagation neural network (Lippman 1987) and SVM (Joachims, 1998) were used due to their robustness in automatic classification – Train the algorithms using the stakeholder pages of the 9 training companies and obtain a model or sets of weights for classification – Test the algorithms on sets of stakeholder pages of 10 companies different from training examples 27 Evaluation Methodology Experimental Design • Consisted of algorithm comparison, feature comparison, and a user evaluation study – Compared the performance of neural network (NN), SVM, baseline method (random classification), human judgment – Compared structural content features, textual content features, and a combination of the two sets of features – 36 Univ of Arizona business students performed manual stakeholder classification and provided comments on the approach 29 Performance Measures • Effectiveness: – Overall accuracy – Within-class accuracy • Efficiency: time used (in minutes) • User subjective ratings and comments User Study • Each subject was introduced to stakeholder analysis and was asked to use our system named “Business Stakeholder Analyzer (BSA)” to browse companies’ stakeholder lists • We randomly selected three companies (Intelliseek, Siebel, and WebMethods) from testing companies to be the targets of analysis 31 Hypotheses (1) • H1: NN and SVM would achieve similar effectiveness when the same set of features was used – Both techniques were robust – Procedure: created 30 sets of stakeholder pages by randomly selecting groups of 5 stakeholder pages of each of the 10 testing companies 32 Hypotheses (2) • H2: NN and SVM would perform better than the baseline method – Incorporated human knowledge and machine learning capability into the classification • H3: Human judgment in stakeholder classification would achieve effectiveness similar to that of machine learning, but that the former is less efficient – They could make use of the Web page’s textual and structural content in classifying stakeholders – Humans might spend more time on it 33 Hypotheses (3) • H4 & H5 examined the use of different types of features in automatic stakeholder classification – H4: structural = textual – H5: combined > structural or textual alone 34 Experimental Results and Discussion Algorithm Comparison • H1 not confirmed • NN performed significantly differently than SVM when the same set of features was used – NN performed significantly better than SVM when structural content features were used – SVM performed significantly better than NN when textual content features or a combination of both feature sets were used – More studies would be needed to identify optimal feature sets for each algorithm 36 Effectiveness of the Approach • H2 confirmed • The use of any combination of features and techniques in automatic stakeholder classification outperformed the baseline method significantly – Our approach has integrated human knowledge with machine-learned information related to stakeholder types … – and was significantly better than a random conjecture 37 Comparing with Human Judgment • H3b and H3d (efficiency) confirmed – Human: 22 minutes (average), varied – Algorithms: 1 – 30 seconds (average) – Showing high efficiency of using the automatic approach to facilitate stakeholder analysis • H3a and H3c (effectiveness) not confirmed – Humans were significantly more effective than NN or SVM – They could rely on more clues in performing classification – Experience in Internet browsing and searching helped narrow down choices 38 However, the algorithms achieved better within-class accuracies than humans in frequently occurring types … 39 Use of Features • To our surprise, hypotheses H4a-b, H5a-b, and H5d were not confirmed – Different feature sets yielded different performances of the algorithms • Structural features enabled NN to achieve better effectiveness than textual ones • Textual and combined features enabled SVM to achieve better effectiveness than structural ones – Do not know exactly why – Future research: studying the effect of features and the nature of algorithms • H5c was confirmed: structural content feature did not add value to the performance of SVM 40 Subjects’ Comments • Overwhelmingly positive • “It would be very helpful!” • “That’s cool!” • “I want to use it.” Conclusions and Future Directions Conclusions • Proposed an automatic classification approach to business stakeholder analysis on the Web – Integrated Human expert knowledge + machinelearned information – Promising in terms of effectiveness and efficiency • A strong potential to use the approach to augment traditional stakeholder classification • Could potentially facilitate business analysts’ interaction with automated stakeholder analysis systems in today’s networked enterprises 43 Future Directions • To automate the next steps of business stakeholder analysis – With more expert participation and more Web page data • Type-specific stakeholder analysis – e.g., partner relationships are often important in developing business strategies • Automating cross-regional business stakeholder analysis – Study multinational business partnerships and cooperation and related HCI issues 44