Weighted Local Naive Bayes Link Prediction
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Weighted network link prediction is a challenge issue in complex network analysis. Unsupervised methodsbased on local structure are widely used to handle the predictive task. However, the results are still far fromsatisfied as major literatures neglect two important points: common neighbors produce different influence onpotential links; weighted values associated with links in local structure are also different. In this paper, weadapt an effective link prediction model—local naive Bayes model into a weighted scenario to address thisissue. Correspondingly, we propose a weighted local naive Bayes (WLNB) probabilistic link predictionframework. The main contribution here is that a weighted cluster coefficient has been incorporated, allowingour model to inference the weighted contribution in the predicting stage. In addition, WLNB can extensivelybe applied to several classic similarity metrics. We evaluate WLNB on different kinds of real-world weighteddatasets. Experimental results show that our proposed approach performs better (by AUC and Prec) thanseveral alternative methods for link prediction in weighted complex networks.
Abstract
Weighted network link prediction is a challenge issue in complex network analysis. Unsupervised methodsbased on local structure are widely used to handle the predictive task. However, the results are still far fromsatisfied as major literatures neglect two important points: common neighbors produce different influence onpotential links; weighted values associated with links in local structure are also different. In this paper, weadapt an effective link prediction model—local naive Bayes model into a weighted scenario to address thisissue. Correspondingly, we propose a weighted local naive Bayes (WLNB) probabilistic link predictionframework. The main contribution here is that a weighted cluster coefficient has been incorporated, allowingour model to inference the weighted contribution in the predicting stage. In addition, WLNB can extensivelybe applied to several classic similarity metrics. We evaluate WLNB on different kinds of real-world weighteddatasets. Experimental results show that our proposed approach performs better (by AUC and Prec) thanseveral alternative methods for link prediction in weighted complex networks.
참고문헌 (25)
- A. L. Barabasi, "Linked: How Everything is Connected to Everything Else and What It Means for Business, Science, and Every Life" (2002), Penguin Group
- T. Zhou, "Predicting missing links via local information", The European Physical Journal B-Condensed Matter and Complex Systems, 71(4), pp.623~630 (2009)
- A. Clauset, "Hierarchical structure and the prediction of missing links in networks", Nature, 453(7191), pp.98~101 (2008)
- D. K. Wind, "Link prediction in weighted networks", Proceedings of 2012 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), pp.1~6 (2012)
- H. Gui, "Modeling topic diffusion in multi-relational bibliographic information networks", Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management, pp.649~658 (2014)
- D. Liben‐Nowell, "The link‐prediction problem for social networks", Journal of the Association for Information Science and Technology, 58(7), pp.1019~1031 (2007)
- L. A. Adamic, "Friends and neighbors on the web", Social Networks, 25(3), pp.211~230 (2003)
- S. Guha, "ROCK: a robust clustering algorithm for categorical attributes", Proceedings of 15th International Conference on Data Engineering, pp.512~521 (1999)
- M. E. Newman, "Analysis of weighted networks", Physical Review E, 70(5) (2004)
- Z. Yu, "Friend recommendation with content spread enhancement in social networks", Information Sciences, 309, pp.102~118 (2015)
- T. Murata, "Link prediction of social networks based on weighted proximity measures", Proceedings of the IEEE, ACM International Conference on Web Intelligence, pp.2007[DOI]
- Z. Liu, "Link prediction in complex networks : a local naïve Bayes model", EPL (Europhysics Letters), 96(4) (2011)
- "Correlated cascades: compete or cooperate" (2015)[URL]
- Y. Sun, "Community evolution detection in dynamic heterogeneous information networks", Proceedings of the 8th Workshop on Mining and Learning with Graphs, pp.137~146 (2010)
- L. Lu, "Link prediction in complex networks : a survey", Physica A : Statistical Mechanics and its Applications, 390(6), pp.1150~1170 (2011)
- M. Al Hasan, "Link prediction using supervised learning", Proceedings of the 4th Workshop on Link Analysis, Counterterrorism and Security (SDM06) (2006)
- W. Liu, "Link prediction based on local random walk", EPL (Europhysics Letters), 89(5) (2010)
- R. N. Lichtenwalter, "New perspectives and methods in link prediction", Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.243~252 (2010)
- A. Barrat, "The architecture of complex weighted networks", Proceedings of the National Academy of Sciences of the United States of America, 101(11), pp.3747~3752 (2004)
- Z. Yang, "AsiaSim 2012", pp.119~126 (2012), Springer
- H. R. De Sa, "Supervised link prediction in weighted networks", Proceedings of the 2011International Joint Conference on Neural Networks (IJCNN), pp.2281~2288 (2011)
- L. Lu, "Link prediction in weighted networks : The role of weak ties", EPL (Europhysics Letters), 89(1) (2010)
- M. E. Newman, "The structure of scientific collaboration networks", Proceedings of the National Academy of Sciences, 98(2), pp.404~409 (2001)
- S. Scellato, "Exploiting place features in link prediction on location-based social networks", Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.1046~1054 (2011)
- R. K. Pan, "World citation and collaboration networks : uncovering the role of geography in science", Scientific Reports, 2 (2012)