Predict Edges in Fliker Social Network Using Data Mining Method

Amir Hossein Rasekh, Zeinab Liaghat, Ala Mahdavi

Sharif University of Technology, Tehran, Iran.
Young Researchers Club, Sepidan Branch, Islamic Azad University, Sepidan, Iran.
Young Researchers Club, Shiraz Branch, Islamic Azad University, Shiraz, Iran.

DOI: 10.4236/iim.2012.43009                

Abstract

Using social networking services is becoming more popular day by day. The websites of the social networks like face-book currently are among the most popular internet services just after giant portals such as Yahoo, MSN and search engines like Google. One of the main problems in analyzing these networks is the prediction of relationships between people in the network. The purpose of this paper is to forecast the friendship of a person with a new person using existing data on Flickr website accurately. In this paper, we achieved about 90% percent correct prediction with regards to the results which are obtained by using data mining methods.

Keywords

Social Network; Data Mining; Link Prediction; Regression

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]	A. H. Rasekh and Z. Liaghat, “Predictive Relationship of Friendship in Social Networks,” The Fifth 2011 Interna- tional Conference on Data Mining, 14-15 December 2011, pp. 47-53.
[2]	Bruce Hoppe and Claire Reinelt, “Social Network Analysis and the Evaluation of Leadership Networks,” The Leader-ship Quarterly, Vol. 21, No. 4, 2010, pp. 600-619.
[3]	B. Hoppe and C. Reinelt, “Social Network Analysis and the Evaluation of Leadership Networks,” The Leadership Quarterly, Vol. 21, No. 4, 2010, pp. 600-619. doi:10.1016/j.leaqua.2010.06.004
[4]	D. M. Boyd and N. B. Ellison, “Social Network Sites: Definition, History and Scholarship,” Journal of Computer-Mediated Communication, Vol. 13, No. 1, 2008, pp. 210-230.
[5]	M. Safaei, M. Sahan and M. Ilkan, “Social Graph Generation & Forecasting Using Social Network Mining,” 33rd Annual IEEE International Computer Software and Applications Conference, Seattle, 22-24 July 2009, pp. 31-35. doi:10.1109/COMPSAC.2009.110
[6]	H. Witten and Eibe Frank, “Data Mining,” Morgan Kaufmann Publishers, San Francisco, 2005.
[7]	C. Faloutsos, K. McCurley and A. Tom-kins, “Fast Discovery of Connection Subgraphs,” International Conference on Knowledge Discovery and Data Mining, New York, 22-25 August 2004, pp. 118-127. doi:10.1145/1014052.1014068
[8]	J. Cranshaw, E. Toch, J. Hong, A. Kittur and N. Sadeh, “Bridging the Gap Between Physical Location and Online Social Networks,” Proceedings of UBICOMP’10, New York, 26-29 September 2010, pp. 119-128.
[9]	Kaggle.com. IJCNN Social Network Challenge. http://www.kaggle.com/socialNetwork, 2011.
[10]	K. H. Zou, A. James, O. Malley and L. Mauri, “Receiver- Operating Characteris-tic Analysis for Evaluating Diagnostic Tests and Predictive Models,” Circulation Journal of the American Heart Association, Vol. 115, No. 15, 6 February 2007, pp. 654-657.
[11]	D. W. Hosmer and Lemeshow, “Applied Logistic Regression,” 2nd Edition, Wiley-Interscience, 2000.
[12]	Y. Wang, “A Multinomial Logistic Regression Modeling Approach for Anomaly Intrusion Detection,” Computer & Security, Vol. 24, No. 8, 2005, pp. 662-647.
[13]	J. Zhang, R. Jin, Y. Yang and Hauptmann, “A. G. Modified Logistic Regression: An Approximation to SVM and It Is Applications in Large-Scale Text Categorization,” Proceedings of the 20th International Conference on Machine Learning, Washington, 21-24 August 2003, pp. 888-895.
[14]	D. G. Kleinbaum, M. Klein and E. R. Pryor, “Logistic Regression: A Self Learning Text,” Springer Science + Business Media LLC, 2010.