Behavior Based Social Dimensions Extraction for Multi-Label Classification

Le Li, Junyi Xu, Weidong Xiao, Bin Ge, Alejandro Raul Hernandez Montoya
2016 PLoS ONE  
Classification based on social dimensions is commonly used to handle the multi-label classification task in heterogeneous networks. However, traditional methods, which mostly rely on the community detection algorithms to extract the latent social dimensions, produce unsatisfactory performance when community detection algorithms fail. In this paper, we propose a novel behavior based social dimensions extraction method to improve the classification performance in multi-label heterogeneous
more » ... . In our method, nodes' behavior features, instead of community memberships, are used to extract social dimensions. By introducing Latent Dirichlet Allocation (LDA) to model the network generation process, nodes' connection behaviors with different communities can be extracted accurately, which are applied as latent social dimensions for classification. Experiments on various public datasets reveal that the proposed method can obtain satisfactory classification results in comparison to other state-of-the-art methods on smaller social dimensions. In heterogeneous networks, edges are driven by various reasons, but the related information is often difficult to be obtained. Therefore, only applying observed connections is not enough for handle multi-label classification task. Mining latent relationship is a kind of representative methods, which aim to find the latent relationship between labeled nodes and unknown nodes. Probabilistic relational models [16] [17] [18] can construct the dependence between connected nodes. The probability of an unknown node's label is conditioned not only on the labels of Behavior Based Social Dimensions Extraction PLOS ONE |
doi:10.1371/journal.pone.0152857 pmid:27049849 pmcid:PMC4822808 fatcat:pqwy6mbgrzf6bl64twphbusyua