Publication Date

5-2017

Date of Final Oral Examination (Defense)

3-7-2017

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Computer Science

Department

Computer Science

Major Advisor

Dianxiang Xu, Ph.D.

Advisor

Francesca Spezzano, Ph.D.

Advisor

Edoardo Serra, Ph.D.

Abstract

Effective friend classification in Online Social Networks (OSN) has many benefits in privacy. Anything posted by the user in social networks like Facebook is distributed among all their friends. Although the user can select the manual option for their post-dissemination, it is not feasible every time. Since not all friends are the same in a social network, the visibility access for the post should be different for different strengths of friendship for privacy. We propose a model with 24 features for finding friendship strength in a social network like Facebook. Previous works in finding friendship strength in social networks have used interaction and similarity based features but none of them has considered using linguistic features as the driving factor to determine the strength. In this paper, we developed a supervised friendship strength model to estimate the friendship strength based upon 24 different features comprising of structure based, interaction based, homophily based and linguistic-based features. We evaluated our approach using a real-world Facebook dataset that has 680 user-friend pairs and obtained accuracy of 85% across close and acquaintance friend classification. Our experiments suggest that features like average comment length; likes, love, friend posts, mutual friends and closeness variable consistently perform better in predicting friendship strength across different classifiers. In addition, combining language-based features with homophilic, structural and interaction features produces more accurate and trustworthy models to evaluate friendship strength.

Share

COinS