基于主题模型的有向社交网络链接预测方法

被引:13
作者
吴梦蝶
唐雁
机构
[1] 西南大学计算机与信息科学学院
关键词
社交网络; 链接预测; 主题模型;
D O I
暂无
中图分类号
TP391.1 [文字信息处理]; TP393.02 [];
学科分类号
120506 [数字人文];
摘要
社交网络链接预测方法通常针对简单网络,且只考虑单一的网络结构特征,难以适应不断复杂化的社交网络.针对此问题,提出一种有向网络的链接预测方法,基于主题模型分析节点语义信息,综合节点属性特征和网络结构特征进行链接预测.实验证明可以更真实地还原社交网络的用户关系,提高链接预测的精度.
引用
收藏
页码:152 / 158
页数:7
相关论文
共 10 条
[1]
有向与加权网络的链路预测 [D]. 
张扬夫 .
湘潭大学,
2011
[2]
Proximity measures for link prediction based on temporal events[J] Paulo R.S. Soares;Ricardo B.C. Prudêncio Expert Systems With Applications 2013,
[3]
Supervised methods for multi-relational link prediction.[J] Darcy Davis;Ryan Lichtenwalter;Nitesh V. Chawla Social Netw. Analys. Mining 2013,
[4]
Link prediction in complex networks: A survey[J] Linyuan Lü;Tao Zhou Physica A: Statistical Mechanics and its Applications 2010,
[5]
Link prediction in weighted networks: The role of weak ties[J] Linyuan Lü;Tao Zhou EPL (Europhysics Letters) 2010,
[6]
Predicting missing links via local information[J] Tao Zhou;Linyuan Lü;Yi-Cheng Zhang The European Physical Journal B 2009,
[7]
Prediction and ranking algorithms for event-based network data[J] Joshua O'Madadhain;Jon Hutchins;Padhraic Smyth ACM SIGKDD Explorations Newsletter 2005,
[8]
Link mining[J] Lise Getoor;Christopher P. Diehl ACM SIGKDD Explorations Newsletter 2005,
[9]
Evolution of networks[J] S. N. Dorogovtsev;J. F. F. Mendes Advances in Physics 2002,
[10]
Using Markov chains for link prediction in adaptive web sites Zhu J;Hong J;Hughes J G; Proceedings of the International Conference on Soft-Ware:Computing in an Imperfect World 2002,