A Hybrid Friend Recommendation Technique Using SVM Based on the Users' Attributes

• Liu Yang University of Brasília
• Deborah M. Ferreira University of Brasília
• Jianya Zheng University of Brasília
• Li Weigang University of Brasília

Abstract

Online social networks have attracted millions of users to integrate their daily life into these social media. As an important e-activity, the Friend Recommendation System (FRS) has been designed to help the users exploring new friends with common interests. However, most existing FRS are using simple methods, such as mutual friends, location-based information etc. This paper proposes a hybrid technique utilizing Support Vector Machine (SVM), recommending people in social networks based on users' attributes. With the case study from Tencent Weibo, the proposed method has improved the accuracy of recommendation comparing with two classic algorithms, Naïve Bayes and Random Forests. Furthermore, considering nine attributes in FRS, we identified that the acceptance of followers is the most important factor to influence the users' decision.

Keywords: Friend Recommendation, Hybrid Approach, SVM

References

Swamynathan, G., Wilson, C., Boe, B., Almeroth, K., & Zhao, B. Y. (2008, August). Do social networks improve e-commerce?: a study on social marketplaces. In Proceedings of the first workshop on Online social networks (pp. 1-6). ACM.

Ma, H., Zhou, D., Liu, C., Lyu, M. R., & King, I. (2011, February). Recommender systems with social regularization. In Proceedings of the fourth ACM international conference on Web search and data mining (pp. 287-296). ACM.

Ting, I. H., Chang, P. S., & Wang, S. L. (2012). Understanding Microblog Users for Social Recommendation Based on Social Networks Analysis. J. UCS, 18(4), 554-576.

Tian, D., & Zheng, W. M. (2012, December). Chinese Microblogger character analysis using SVM. In Wavelet Active Media Technology and Information Processing (ICWAMTIP), 2012 International Conference on (pp. 385-389). IEEE.

Sandes, E., Weigang, L and Melo, A. (2012) “Logical model of relationship for online social networks and performance optimizing of queries”. In Proceedings of Web Information Systems Engineering - WISE 2012, X.S. Wang, I. Cruz, A. Delis, et al. Springer Berlin Heidelberg, LNCS: 726–736.

Li, X., Guo, L., & Zhao, Y. E. (2008, April). Tag-based social interest discovery. In Proceedings of the 17th international conference on World Wide Web (pp. 675-684). ACM.

Shechtman, O. (2013). The Coefficient of Variation as an Index of Measurement Reliability. In Methods of Clinical Epidemiology (pp. 39-49). Springer Berlin Heidelberg.

Deng, N. Y., & Tian, Y. J. (2004). A new method of data mining: support vector machines. Science Publication, Beijing City.