基于Attention机制的链接预测算法
|
摘要
针对社会网络中链接预测问题,提出了基于注意力(Attention)机制的链接表示及其预测算法.基于待预测节点的共邻关系构建其链接局部网络,设计了基于紧密游走的网络拓扑序列化方法.采用双向循环神经网络(Bi-RNN)对链接序列进行向量编码,以充分挖掘序列相关节点间的上下文依赖信息.通过Attention机制对链接中的节点进行关注和加权,强化重要节点对链接预测任务的贡献,实现链接拓扑特征的自动提取与准确分类预测.实验结果表明,在4种不同类型的社会网络数据集中,该算法的准确率和运算效率都有较大提高且普适性较强.
Aiming at the link prediction task in social networks,a link representation method and a link prediction algorithm based on attention mechanism were proposed.A link local network was designed which contained a node pair to be link predicted and their co-neighbors,and was serialized by multiple closely walks.Links sequences were encoded by bi-directional recurrent neural network(Bi-RNN) for its ability on mining the context information from the sequential nodes.The attention mechanism,used to focus and weight the nodes in the link,strengthens the contribution of important nodes,which can promote the accuracy of link prediction.Experiments on four types of social network datasets show that the algorithm has significant improvement in accuracy and computational efficiency,and is suitable for multiple social networks.
Aiming at the link prediction task in social networks,a link representation method and a link prediction algorithm based on attention mechanism were proposed.A link local network was designed which contained a node pair to be link predicted and their co-neighbors,and was serialized by multiple closely walks.Links sequences were encoded by bi-directional recurrent neural network(Bi-RNN) for its ability on mining the context information from the sequential nodes.The attention mechanism,used to focus and weight the nodes in the link,strengthens the contribution of important nodes,which can promote the accuracy of link prediction.Experiments on four types of social network datasets show that the algorithm has significant improvement in accuracy and computational efficiency,and is suitable for multiple social networks.
引文
[1]LIBEN-NOWELL D,KLEINBERG J.The link prediction problem for social networks[J].Journal of the Association for Information Science and Technology,2007,58(7):1019-1031.
[2]LICHTENWALTER R N,LUSSIER J T,CHAWLA NV.New perspectives and methods in link prediction[C]//Proc of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Washington:ACM,2010:243-252.
[3]HASAN M,MOHAMMAD A.Link prediction using supervised learning[J].Workshop on Link Analysis Counterterrorism and Security,2006,30(9):798-805.
[4]LI K,GAO J,GUO S,et al.LRBM:a restricted boltzmann machine based approach for representation learning on linked data[C]//Proc of IEEE International Conference on Data Mining.Shenzhen:IEEE,2015:300-309.
[5]DAS R,NEELAKANTAN A,BELA NGER D,et al.Chains of reasoning over entities,relations,and text using recurrent neural networks[EB/OL].[2016-07-05].http://cn.arxiv.org/abs/1607.01426.
[6]XIONG W,HOANG T,WANG W Y.Deeppath:a reinforcement learning method for knowledge graph reasoning[C]//Proc of Conference on Empirical Methods in Natural Language Processing.Copenhagen:IEEE,2017:564-573.
[7]SCHUSTER M,PALIWAL K K.Bidirectional recurrent neural networks[J].IEEE Transactions on Signal Processing,1997,45(11):2673-2681.
[8]NEWMA M E.Clustering and preferential attachment in growing networks[J].Physical Review E:Statistical Nonlinear and Soft Matter Physics,2001,64(2):1-4.
[9]ZHOU T,LYU L Y,ZHANG Y C.Predicting missing links via local information[J].The European Physical Journal:B,2009,71(4):623-630.
[10]ADAMIC L A,ADAR E.Friends and neighbors on the web[J].Social Networks,2003,25(3):211-230.
[11]ZHANG L,CHENG H,FANG Y Q.Research on link representation and prediction method based on convolution neural network[J].Journal of Zhejiang University:Engineering Edition,2018,52(3):552-559.
[12]YANG Z,YANG D,DYER C,et al.Hierarchical attention networks for document classification[C]//Proc of Conference on the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.San Diego:IOS Press,2017:1480-1489.
[13]LESKOVEC J.Stanford large network dataset collection[EB/OL].[2018-03-01].http://snap.Stanford.edu/data/.
[14]PAPADIMITRIOU A,SYMEONIDIS P,MANOLOPOU-LOS Y.Fast and accurate link prediction in social networking systems[J].Journal of Systems and Software,2012,85(9):2119-2132.
[2]LICHTENWALTER R N,LUSSIER J T,CHAWLA NV.New perspectives and methods in link prediction[C]//Proc of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Washington:ACM,2010:243-252.
[3]HASAN M,MOHAMMAD A.Link prediction using supervised learning[J].Workshop on Link Analysis Counterterrorism and Security,2006,30(9):798-805.
[4]LI K,GAO J,GUO S,et al.LRBM:a restricted boltzmann machine based approach for representation learning on linked data[C]//Proc of IEEE International Conference on Data Mining.Shenzhen:IEEE,2015:300-309.
[5]DAS R,NEELAKANTAN A,BELA NGER D,et al.Chains of reasoning over entities,relations,and text using recurrent neural networks[EB/OL].[2016-07-05].http://cn.arxiv.org/abs/1607.01426.
[6]XIONG W,HOANG T,WANG W Y.Deeppath:a reinforcement learning method for knowledge graph reasoning[C]//Proc of Conference on Empirical Methods in Natural Language Processing.Copenhagen:IEEE,2017:564-573.
[7]SCHUSTER M,PALIWAL K K.Bidirectional recurrent neural networks[J].IEEE Transactions on Signal Processing,1997,45(11):2673-2681.
[8]NEWMA M E.Clustering and preferential attachment in growing networks[J].Physical Review E:Statistical Nonlinear and Soft Matter Physics,2001,64(2):1-4.
[9]ZHOU T,LYU L Y,ZHANG Y C.Predicting missing links via local information[J].The European Physical Journal:B,2009,71(4):623-630.
[10]ADAMIC L A,ADAR E.Friends and neighbors on the web[J].Social Networks,2003,25(3):211-230.
[11]ZHANG L,CHENG H,FANG Y Q.Research on link representation and prediction method based on convolution neural network[J].Journal of Zhejiang University:Engineering Edition,2018,52(3):552-559.
[12]YANG Z,YANG D,DYER C,et al.Hierarchical attention networks for document classification[C]//Proc of Conference on the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.San Diego:IOS Press,2017:1480-1489.
[13]LESKOVEC J.Stanford large network dataset collection[EB/OL].[2018-03-01].http://snap.Stanford.edu/data/.
[14]PAPADIMITRIOU A,SYMEONIDIS P,MANOLOPOU-LOS Y.Fast and accurate link prediction in social networking systems[J].Journal of Systems and Software,2012,85(9):2119-2132.