Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


17324.pdf (1.25 MB)

ETD Abstract Container

Abstract Header

Empirical Study On Key Attributes of Yelp dataset which Account for Susceptibility of a user to Social Influence

Alluri, Anjaneya Varma
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439281364

Abstract Details

2015, MS, University of Cincinnati, Engineering and Applied Science: Computer Science.
This thesis focuses on finding the key attributes of a social network of friends that help us to understand and measure the `Susceptibility’ of a user to `Influence’. Susceptibility of a user here refers to a factor that explains the state of being likely to get effected by a friend in a social network to perform a certain action. On the other hand, Influence refers to a force or phenomenon where a certain action performed by a user on a particular product or business impacts the activity of his friends on that same product. For this study I have selected Yelp, as it is one of the huge and active social rating network where users across the globe participate in providing meaningful reviews about various businesses. User activity here refers to rating or reviewing a business in a particular location. With such volume of data and such huge social network of friends, it makes a perfect dataset to perform our study. The common attributes that would help us to measure susceptibility in a social rating network were recognized and were used to generate a mathematical model for susceptibility. In the end, we were also able to recognize an efficient learning algorithm that fits Yelp dataset and may be used to help predict susceptibility of a user to influence. In our experiments the final susceptibility measures determined were higher when compared to a dataset model prior to considering the key attributes. With this study, we could clearly see that a review of an elite and active friend in Yelp has a significant impact on the susceptibility of a user to influence. Along with the number of reviews exposed to a user by his friends, a new statistic named “useful voting fraction” for each of their friends also helped to derive an efficient learning algorithm that fits Yelp Dataset. Finally, we also found that the star rating which each review gets, plays an important role in measuring susceptibility.
Fred Annexstein, Ph.D. (Committee Chair)
Raj Bhatnagar, Ph.D. (Committee Member)
Chia Han, Ph.D. (Committee Member)
51 p.
Computer Science
Susceptibility; Influence; Yelp; Logistic Regression; friend network; Data Attributes

Recommended Citations

Citations

  • Alluri, A. V. (2015). Empirical Study On Key Attributes of Yelp dataset which Account for Susceptibility of a user to Social Influence [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439281364

    APA Style (7th edition)

  • Alluri, Anjaneya Varma. Empirical Study On Key Attributes of Yelp dataset which Account for Susceptibility of a user to Social Influence. 2015. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439281364.

    MLA Style (8th edition)

  • Alluri, Anjaneya Varma. "Empirical Study On Key Attributes of Yelp dataset which Account for Susceptibility of a user to Social Influence." Master's thesis, University of Cincinnati, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1439281364

    Chicago Manual of Style (17th edition)

ucin1439281364
758
© 2015, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.