International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 Analytical Hierarchy Process to explore the influence of social networking on student’s academic performance Addisu Mesfin1, Prof.Shruti Patil2 1 MTech Student, Symbiosis Institute Of Technology (SIT), Pune , 2 Assistant Professor, Symbiosis Institute of Technology (SIT), Pune 412115, Maharashtra State, India . Abstract Social networks have penetrated all ages of internet users, becoming an efficient way of communication as well as entertainment, especially in the student community. This advent has not only impacted the human lives positively but also created a niche for some pessimistic activities and behavioral tendencies. The purpose of this paper is to provide a quantitative insight on the use of various social networking sites and their effects on the academic performance of students. An inverse ratio is observed between the overall well being of an individual and the amount of time one spends on the social networks. This observation is tested with the help of AHP as a mathematical tool. The analytical hierarchy process (AHP) is used as a technique for arranging and analyzing complicated problems by using mathematics and psychology. It can use both prejudiced individual judgments and objective assessment just by Eigen vector and examining the reliability of the assessment by Eigen value. I. Introduction The internet has given us the ability to connect with people from around the globe with a few clicks of a button. You can easily send information to a friend or get information. Social sites such as Whatsapp, Facebook, viber, imo etc, have attracted millions of users, many of whom have integrated these sites into their daily practices. College students spend a lot of time on these sites uploading or downloading, getting information concerning their career or academic work. Students are always online every second, chatting with friends, watching online movies, playing online games, doing research. Social site has become a habit for some students; they find it difficult to study for one hour without login to one social site. Some students have become very smart because of the information they get from this sites, why some have become very poor academically, because it easy to get almost any materials for school assignment. Fig 1: Show % of time spent on various networking sites ISSN: 2231-5381 http://www.ijettjournal.org Page 435 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 II. Literature review Different authors have conducted many researches about influence of social sites on college student’s academic performance. Most paper result shows: student higher openness spend more time on face book. Time spend on face book is strongly or significantly negative related to all overall GPA. Some paper shows sharing information is positively predictive and using face book socializing is negatively predictive. Face book users scored low grade than non face book users because face book users study low hour than non face book users. Time spend on face book has negative relationship with time spend to prepare class. Some result shows that personality is also related to use of social sites. The No of paper Paper title 1 The effects of personality traits, self-esteem, loneliness, and narcissism on Face book use among university students Methodology used 2 3 4 Too much face and not enough books: The relationship between multiple indices of Face book use and academic performance A tale of two sites: Twitter vs. Facebook and the personality predictors of social media usage Effect of online social networking on student academic performance ISSN: 2231-5381 majority user of online social networks is college students. Male were related to SNS than female to play online games. Females upload more photos and update their status than male. So, gender has direct relationship related to SNS use and viewing photos, comments are spend more time. SNS is decreasing both efficiency and productivity in academic settings. Some research concludes that different motivations are influence to use face book like, photos, videos, online games etc. There is also the negative relationship between time spend and use of laptop, tablet, mobile phone in class. Most college students use class tasks but they are tend to facebook. Commonly students use their electronic device to send text, photo. Advantage Disadvantage Use three of big five traits(Neuroticism (worry) extraversion (stay alone) openness (inner feeling) use of face book North-Eastern US sample 3866 online in survey monkey.com(on facebook) Reduce higher openness spend more time on face book. In this paper study technology use and academic performance Time spend on face book is strongly or significantly negative related to all over GPA Use big-five factors use of the two largest social network sites (face book and twitter ) online Collect survey all over US area Develop use of language(writing skill) openness(+ve correlations) neuroticism(emotional) Extraversion(quiet &shy) Most USA students use class tasks but they are tend to face book. Higher attentions span online social network sites http://www.ijettjournal.org Visit or comment friends photo, un friends photo on face book take more time . Page 436 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 5 6 7 The relationships among the Big Five Personality factors, self-esteem, narcissism, and sensationseeking to Chinese University students’ uses of social networking sites (SNSs) An exploration of social networking site use, multitasking, and academic paper was write in china Survey 265 The result shows that Narcissism and extraversion were positively related to post photos Male were related to SNS play more online games than female Females were more update their status than males This research give SNS is decrease both pause to efficiency and productivity university admin in academic settings and policy makers to think more about the use of SNS related to education Personality and Canada The research Psychological distress and motivations associated university of conclude that high level of associated with Facebook use south-western Different sensitive to treat Ontario students motivations are at all questioners influence to use were distribute face book like, online photos, videos, online games Table 1: show literature review which are written by different authors in different years First prepared Dataset, second classify and generate III. Methodology used for analyzing problem input/output parameters, third association model and under consideration the last verify the model by using AHP. We prepared Data Collection We have taken 172 number of students in 4 different questionnaires to display the main criteria (Attention departments(computer science ,IT, Civil, M-tech/cs). Span , Academic competence , Effective time Among this there were 122 male and 50 female management , General Awareness , Neuroticism ) , students in different batches. A printed hardcopy of sub-criteria (Questionnaires) and Alternatives the questionnaire was distributed to them which (Facebook , Whatsapp , Viber) then as individual consisted of total 26 questions. Based on the pairs ask respondents to compare the pair scale feedback received, a softcopy of dataset was prepared ranges from 1 to 9 (Saaty, 1980). which was then processed using AHP. Following were the steps taken: Sampled USA 451 and Europe 406 online from survey hosting website Fig 2: Workflow of Methodology ISSN: 2231-5381 http://www.ijettjournal.org Page 437 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 WEKA WEKA (Waikato Environment for Knowledge Analysis): is a popular suite of machine learning software written in Java, developed at the University of Waikato, New Zealand. It is free software licensed under the GNU General Public License. Weka is a workbench that contains a collection of visualization tools and algorithms for data analysis and predictive modeling. Advantages of WEKA a) Free availability under the GNU General Public License. b) Portability, since it is fully implemented in the Java programming language and thus runs on almost any modern computing platform. c) A comprehensive collection of data preprocessing and modeling techniques. d) Ease of use due to its graphical user interfaces. Weka supports several standard data mining tasks, more specifically, data preprocessing, clustering, classification, regression, visualization, and feature selection. All of Weka's techniques are predicated on the assumption that the data is available as one flat file or relation, where each data point is described by a fixed number of attributes (normally, numeric or nominal attributes, but some other attribute types are also supported). Weka provides access to Excel(CSV format) SQL databases using Java Database Connectivity and can process the result returned by a database query. It is not capable of multi-relational data mining, but there is separate software for converting a collection of linked database tables into a single table that is suitable for processing using Weka. Another important area that is currently not covered by the algorithms included in the Weka distribution is sequence modeling. Fig 3: Process of Association Model Association Rule and Predictive Apriori Association Association Rule of data mining Association rules are used to find the frequent Rule. pattern, association or correlation in transaction There are 16 questions which were distribute to SIT database. Association rule mining can be used in students . Depends on the weka result we got the Basket Data Analysis. Association Rule algorithms above results in association Algorithm. Each are Apriori, Sampling, Partitioning & Parallel questions use as input, weka uses to combine this Algorithm. This section describes the Apriori input questions and process after processing we got six outputs this are Attention Span, Academic ISSN: 2231-5381 http://www.ijettjournal.org Page 438 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 Competency, Effective time management, General awareness, Neuroticism and Physical health. This six outputs are not got directly from the questionnaire but all questions are talk about this six out puts(Attention Span, Academic Competency, Effective time management, General awareness, Neuroticism and Physical health). Attention Span: is the length of time which a person is able to concentrate mentally on a particular activity. And it is the amount of concentrated on can spend on task without becoming distracted. Your attention Span can have a major impact at your performance at work. Most Students Spend their time on social site networks this is show that the attention span of college students in social media is more than attention span of college students on their academic activity. Academic competence: It includes many of the academic activities like reading, writing, calculating, solving problems, attending, questioning, and studying needed for academic success. The academic performance of college students is decrease because of the social site networks like facebook, Whatsapp, Viber etc. Students are stay more than 4 hours in a day in this sites so they did not give more time to their academic activities. Effective time management: It is very important to college students that you develop effective strategies for managing your time to balance the conflicting demands of time for study, leisure, playing and access internet. Time management skills are valuable in job hunting, but also in many other aspects of life: from revising for examinations to working in a vacation job. Some of these skills including setting clear goals, breaking your goals down into discreet steps, and reviewing your progress towards your goals are covered in Action Planning. But because of social sites college students did not use this Effective time management effectively so, college students are weak in their academic activities. General Awareness: It is knowledge or perception of situation or facts to concern about and wellinformed interest in a particular situation or development. Having knowledge or discernment of something that aware of about the difference social site networks are the main reason to spend our time which is use to academic activities. Aware implies ISSN: 2231-5381 knowledge gained through one's own perceptions or by means of outside information. All most all college students have general awareness about the social site networks but they use this social sites everyday more than their academic studies. Neuroticism : Persons are often self-conscious and shy, and they may have trouble controlling urges and delaying gratification. Neuroticism is a prospective risk factor for most "common mental disorders", such as depression, phobia, panic disorder, other anxiety disorders, and substance use disorder symptoms that traditionally have been called neuroses. So , most social sites like Facebook ,Whatsapp, Viber etc are the fundamental cause of neuroticism on college students. Physical Health : It is critical for overall well-being and is the most visible of the various dimensions of health, which also include social, intellectual, emotional, spiritual and environmental health. Some of the most obvious and serious signs that we are unhealthy appear physically. For instance social site networks are cause of physical health problem like, Eye problem, Skin problem, Finger problem and cause of ergonomics. IV. AHP model for influence of social sites in student’s academic performance Users of AHP first decompose their decision problem in to a hierarchy of more easily sub-problems. Each can be analysis independently .The elements of hierarchy can related to any aspect of the decision problem tangible or intangible. Once hierarchy is built the decision makers systematically evaluate its various elements by comparing them to each other two at a time, with respect to their impact on an element which are listed . For example facebook , Whatsapp , Viber Etc. The decision makers can use correct data about the elements, but they typically use their judgments about the elements relative meaning and importance. AHP uses to convert the evaluation into numerical value that can be processed and compared the entire range of the problem. A numerical weight or priority is derived for each element of the hierarchy, allowing diverse and elements to be compared to one another in a rational and consistent way. In the final step of the process, numerical priorities are calculated for http://www.ijettjournal.org Page 439 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 each of the decision alternatives. These numbers represent the alternatives' relative ability to achieve the decision goal, so they allow a straightforward consideration of the various courses of action. Fig 4: AHP Model Influence of social sites in student’s academic performance AS/Q6 : Where do you browse ETM/Q7 : When browse ETM/Q9 : How many hours go online AS/Q8 : What do you browse ETM/Q11 : Why do you choose AS/Q10 : Most your choose ETM/Q16: Which Spend more time AS/Q16 : Spend more time GA/Q2 : Are you aware this sites GA/Q11 : Why do you choose AC/Q7 : When browse GA/12 : What are effects AC/Q 9 : How many hours you go online GA/Q13 : Influence student performance AC/Q13 : Influence student performance N/Q12 : What are effects AC/Q14 : Related to your GPA N/Q13 : Influence student performance AC/Q16 : Which Spend more your time N/Q15 : How to satisfied All his the above questionnaires are solved in AHP . The results are show in below. The following three-step procedure provides a good V. AHP Implementation approximation of the synthesized priorities. AHP implementation uses matrix which the Step 1: Sum the values in each column of the pair questionnaires are more complex or has more than or wise comparison matrix. equal to three alternatives. Around twelve questions Step 2: Divide each element in the pair wise matrix have three and more than three chooses. So, AHP by its column total. The resulting matrix is referred to implementation is must. The result is show in below: as the normalized pair wise comparison matrix. Procedure for Synthesizing Judgments ISSN: 2231-5381 http://www.ijettjournal.org Page 440 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 Step 3: Compute the average of the elements in each row of the normalized matrix. λmax = Column sum of 1st * weight of 1st column + Column sum of 2nd * weight of 2nd column + Column sum of 3rd * weight of 3rd + Column sum of 4th * weight of 4th column... ------------column x * weight n----(1) CI = λmax – n / n-1-----------------(2) CR = CI / RI--------------------------(3) students spend their time on facebook, 10 students spend their time on (facebook , whatsapp and viber) and 4 students spend their time on other social sites. Many students choose whatsapp, their reason is fast and cheap. Most students browse this sites in hostel, home, college, café and class. Most students browse at night. It has influence students performance 121 students said yes, 27 students said no and 24 students not respond. Most students go online daily. Around 93 students spend 4 or 6 hours in everyday. Most students use online to chatting, sports(news), entertainment(game) and education. The following tables show the summery of the usage of social sites among respondents or students by using AHP implementation. VI. Results All students have awareness about social sites. 79 students spend their time on (facebook and whatsapp) ,66 students spend their time on whatsapp, 13 Table 2: Random Consistency Index (RI) n 1 2 3 4 5 6 7 8 9 10 RI 0 0 0.58 0.9 1.12 1.24 1.32 1.41 1.45 1.49 All questionnaires have its own results and recommend which one is best among each questions chooses What is your Mobile or laptop type? frequent percentage Laptop , android phone 131 76.162 others 41 23.84 total 172 100% Around 76.16% of SIT students are use Laptop and android phones. But 23.83% of students use other devices like iphone, Apple and some are use either Laptop or Android phone. Are you aware of social network sites? frequent percentage yes 172 100 no 0 0 total 172 100 % All SIT students which are answered my questionnaires in all batches and departments have awareness to social sites. Do you have access to the internet? frequent percent yes 167 97.09 no 5 2.91 Total 172 100% 97.09% of SIT students in have internet access. But around 2.91 % of students have not Internet access because of different reasons like economically, lack of device which students use to access internet. How long you have social site profile? ISSN: 2231-5381 http://www.ijettjournal.org Page 441 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 more than 3 years Weight (W) 0.53296088 Vector(V) 100 % 53.29608796 more than 2.5 years more than 2 years more than 1 year 0.22856992 0.15016238 0.04415341 22.85699188 15.01623834 4.415340914 More than 6 month 0.04415341 4.415340914 Sum 1 100 λ max CI CR 5.231988188 0.057997047 0.051783078 For this implementation, we get the results that more than 3 years is the best choice, followed by more than 2.5 years as the second choice, followed by more than 2 years as the third choice, followed by more than 1 year as the fourth choice and the worst choice is More than 6 month. The composite weights are ratio scale. We can say that more than 3 years is 12.07 times more preferable than More than 6 month and more than 1 year , more than 3 years is 3.55 times more preferable than more than 2 years and more than 3 years is 2.33 times more preferable than more than 2.5 years . The pie chart shown in figure 5 gives details about different alternatives in terms of their proportion in how long use social sites. Fig 5: shows percentage of how long use social sites The above values can also be verified through MATLAB as shown in figure 6. The snippet in figure 6 shows the process For the same as excel implementation result. ISSN: 2231-5381 http://www.ijettjournal.org Page 442 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 Fig 6: Use of MATLAB to verify AHP results for proposed analysis Most Effects of Social sites Weight (W) 0.395268962 0.272550306 0.138332298 0.075292037 0.075292037 0.043264361 1 Time consume Addictive Eye problem Discipline Sleepless Expensive Sum λ max 6.067509105 CI 0.013501821 CR 0.010888565 Vector (V) % 39.5268962 27.25503059 13.83322976 7.529203692 7.529203692 4.326436068 100 For this implementation, we get the results that Time consume is the best choice, followed by Addictive as the second choice, followed by Eye problem as the third choice followed by Discipline as the fourth choice, followed by Sleepless as the fifth choice and the worst choice is Expensive. The composite weights are ratio scale. We can say that Time consume is 9.14 times more preferable than Expensive , Time consume is 5.25 times more preferable than (Sleepless and Discipline) , Time consume is 2.86 times more preferable than Eye problem and Time consume is 1.45 times more preferable than Addictive. The pie chart shown in figure 9 gives details about different alternatives in terms of their proportion in most effects of social sites in students performance. ISSN: 2231-5381 http://www.ijettjournal.org Page 443 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 Fig 7 : Show percentages the most effects of social sites. The above values can also be verified through MATLAB as shown in figure 10. The snippet in figure 10 shows the process For the same as excel implementation result. Fig 8 : Use of MATLAB to verify AHP results for proposed analysis Which one is spend more your time? Facebook Others Whatsapp Facebook, Whatsapp Weight (W) 0.456936795 0.25564231 0.244772745 0.04264815 Vector (V)100 % 45.69367948 25.56423104 24.47727452 4.264814961 Sum 1 100 λ max 4.024527796 CI 0.008175932 CR 0.009084369 ISSN: 2231-5381 http://www.ijettjournal.org Page 444 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 For this implementation, we get the results that Facebook is the best choice, followed by Others as the second choice, followed by Whatsapp as the third choice and the worst choice is Facebook, Whatsapp. The composite weights are ratio scale. We can say that Facebook is 10.71 times more preferable than (Facebook, Whatsapp) , Facebook is 1.86 times more preferable than Whatsapp and Facebook is 1.78 times more preferable than Others. The Total Final Results are: Weight (W) Vector(V) 100 % Academic Competence 0.393134702 39.31347018 General Awareness 0.287317351 28.73173509 Attention Span 0.130690291 13.06902907 Neuroticism 0.130690291 13.06902907 Effective Time mgt 0.058167366 5.81673659 Sum 1 100 λ max 5.164594451 CI 0.041148613 CR 0.036739833 For this implementation, we get the results that Academic Competence is the best choice, followed by General Awareness as the second choice, followed by Attention Span as the third choice, followed by Neuroticism as the fourth choice and the worst choice is Effective Time mgt. The composite weights are ratio scale. We can say that Academic Competence is 6.76 times more preferable than Effective Time mgt, Academic Competence is 3 times more preferable than (Neuroticism and Attention Span) and Academic Competence is 1.37 times more preferable than General Awareness. The pie chart shown in figure 11 gives details about different alternatives in terms of their proportion in which the final result of effect of social sites. Fig 9: Show percentages that the final result uses of influence of social sites. The above values can also be verified through MATLAB as shown in figure 12. The snippet in figure 12 how the process For the same as excel implementation result. ISSN: 2231-5381 http://www.ijettjournal.org Page 445 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 Fig 10: Use of MATLAB to verify AHP results for proposed analysis VIII. CONCLUSSION Identify why most students use social sites specially facebook and whatsapp. Most researchers are focus only on Face book. But we included other social sites like Whatsapp, Viber also .Most the papers shows only sample results manually. They didn’t use any data mining techniques. Prediction or AHP implementation is necessary because the data is very large. Without data mining prediction and AHP, the output is not correct or it shows only highlight results. Then we gave recommendation effect of the use of social sites and what is the solution one to minimize our time spend. The final result shows facebook is most time consuming as compared to whatsapp and viber because facebook has different pages, student visit their friends photo, video , play online games which their friends invite etc. Finally this are the main AHP implementation results related to students academic performance and other events. 1. The first composite weights ratio scale are : We can say that Academic Competence is 6.76 times more preferable than effective time mgt, academic competence is 3 times more preferable than (Neuroticism and Attention span) and academic ISSN: 2231-5381 competence is 1.37 times more preferable than aeneral awareness. 2. The second composite weights ratio scale is: We can say that time consume is 9.14 times more preferable than expensive, time consume is 5.25 times more preferable than (Sleepless and Discipline), time consume is 2.86 times more preferable than Eye problem and time consume is 1.45 times more preferable than addictive. IX. Reference [1] [2] [3] [4] [5] Aryn C. Karpinski et al (2013) An exploration of social networking site use, multitasking, and academic performance among United States and European university students. Craig Ross *, Emily S. Orr, Mia Sisic, Jaime M. Arseneault, Mary G. Simmering, R. Robert Orr (2009) “Personality and motivations associated with Facebook use”. David John Hughes a et al (2012) A tale of two sites: Twitter vs. Facebook and the personality predictors of social media usage Heyam A. Al-Tarawneh (2014) The Influence of Social Networks on Students’ Performance. Jason L. Skues et al (2012) The effects of personality traits, self-esteem, loneliness, and narcissism on Facebook use among university students. Jin-Liang Wanga,b et al (2012)The relationships among the Big Five Personality factors, self-esteem, narcissism, and sensation-seeking to Chinese University students’ uses of social networking sites (SNSs). http://www.ijettjournal.org Page 446 International Journal of Engineering Trends and Technology (IJETT) – Volume 33 Number 9- March 2016 [6] [7] [8] Jomon Aliyas Paul et al (2012) Effect of online social networking on student academic performance. Napoleon, Egedegbe (29 may 2013) The Effect of Social Networking Sites on Students' Academic Performance in Girne American University, North Cyprus Reynol Junco (2012)Too much face and not enough books: The relationship between multiple indices of Facebook use and academic performance. ISSN: 2231-5381 [9] Dr. Saaty, Book ,AHP Tutorials .Doc. [10] Suhas Machhindra Gaikwad et al ( December 2015) Analytical Hierarchy Process to Recommend an Ice Cream to a Diabetic Patient Based on Sugar Content in it. [11] Wikipedia http://www.ijettjournal.org Page 447