Fardad FAROKHI, Saghar BABAKHANBAK, Kaveh KAVOUSI

Application of a time delay neural network for predicting positive and negative links in social networks

The structure of online social networks cannot in most cases be defined only by the relationships among its members. The interuser relations on the website of social networks are often a mixture of positive and friendly interactions, such as trust and interest, and at the same time negative interactions, such as mistrust and lack of interest. One of the issues with signed social networks is the prediction of edge signs. Some of the most recent studies have tried to extract features of the users and their relations with their neighbors to solve this problem. The results of such works demonstrate a relative success for such efforts. The present paper is an effort to offer a rather new approach to solve the problem that has not been addressed yet. The approach uses the distributed time delay neural network to present a model capable of predicting the sign of hidden or unknown edges. The goal is to have the neural network trained by available data and then assess its efficiency of the network. Our implementation on the actual datasets of Slashdot media shows that the algorithm, while simple, has a higher rate of precision in comparison to other existing methods.

PDF

___

[1] Ye J, Cheng H, Zhu Z, Chen M. Predicting positive and negative links in signed social networks by transfer learning. In: Proceedings of the 22nd International Conference on World Wide Web; 1317 May 2013. pp. 1477-1488.
[2] Leskovec J, Huttenlocher D, Kleinberg J. Predicting positive and negative links in online social networks. In: Proceedings of the 19th International Conference on World Wide Web; 2630 April 2010. pp. 641-650.
[3] Newman MEJ. The structure and function of complex networks. SIAM Rev 2003; 45: 167-256.
[4] Burke M, Kraut R. Mopping up: modeling Wikipedia promotion decisions. In: Proceedings of the 2008 ACM conference on Computer Supported Cooperative Work; 812 November 2008. pp. 27-36.
[5] Guha R, Kumar R, Raghavan P, Tomkins A. Propagation of trust and distrust. In: Proceedings of the 13th International Conference on World Wide Web; 1722 May 2004. pp. 403-412.
[6] Massa P, Avesani P. Controversial users demand local trust metrics: an experimental study on epinions.com community. In: Proceedings of the 20th National Conference on Artificial Intelligence; 2005. pp. 121-126.
[7] Brzozowski MJ, Hogg T, Szabo G. Friends and foes: ideological social networking. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems; 510 April 2008. pp. 817-820.
[8] Kunegis J, Lommatzsch A, Bauckhage C. The Slashdot Zoo: mining a social network with negative edges. In: Proceedings of the 18th International Conference on World Wide Web; 2024 April 2009. pp. 741-750.
[9] Lampe CAC, Johnston E, Resnick P. Follow the reader: filtering comments on Slashdot. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems; 2007. pp. 1253-1262.
[10] Leskovec J, Huttenlocher D, Kleinberg J. Signed networks in social media. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems; 2010. pp. 1361-1370.
[11] Dubois T, Golbeck J, Srinivasan A. Predicting trust and distrust in social networks. Privacy, security, risk and trust (PASSAT). In: 2011 IEEE Third International Conference on Social Computing; 2011. pp. 418-424.
[12] Wang C, Bulatov AA. Inferring attitude in online social networks based on quadratic correlation. In: Yu JXU, Ravindran B, Pudi V, editors. Advances in Knowledge Discovery and Data Mining. Cambridge, MA, USA: AAAI Press, 2014. pp. 139-150.
[13] Liu F, Liu B, Wang X, Liu M, Wang B. Features for link prediction in social networks: a comprehensive study. In: IEEE International Conference on Systems, Man, and Cybernetics; 2012. pp. 1706-1711.
[14] Chiang KY, Natarajan N, Tewari A, Dhillon IS. Exploiting longer cycles for link prediction in signed networks. In: Proceedings of the 20th ACM International Conference on Information and Knowledge Management; 2011. pp. 1157-1162.
[15] Malekzadeh M, Fazli M, Khalilabadi PJ, Rabiee H, Safari M. Social balance and signed network formation games. In: Proceedings of KDD Workshop on Social Network Analysis; 2011.
[16] Haykin S. Neural Networks: A Comprehensive Foundation. New York, NY, USA: Pearson Education.
[17] Waibel A, Hanazawa T, Hinton G, Shikano K, Lang KJ. Phoneme recognition using time-delay neural networks. IEEE T Acoust Speech 1989; 37: 328-339.
[18] Reese MG. Application of a time-delay neural network to promoter annotation in the Drosophila melanogaster genome. Comput Chem 2001; 26: 51-56.
[19] More JJ The Levenberg-Marquardt algorithm: implementation and theory In: Watson GA, editor Numerical Analysis New York, NY, USA: Springer, 1978 pp 105-116.
[20] Dohnal IJ. Using of Levenberg-Marquardt method in identification by neural networks. In: Student EEICT; 2004. pp. 361-365.
[21] Zouhal LM, Denoeux T. An evidence-theoretic K-NN rule with parameter optimization. IEEE T Syst Man Cy C 1998; 28: 263-271.
[22] Kavousi K, Sadeghi M, Moshiri B, Araabi BN, Moosavi-Movahedi AA. Evidence theoretic protein fold classification based on the concept of hyperfold. Math Biosci 2012; 240: 148-160.