• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 计算机网络与信息安全 • 上一篇    下一篇

基于最简子图的链接表示及预测

尚振浩,程华,房一泉   

  1. (华东理工大学信息科学与工程学院,上海 200237)
  • 收稿日期:2018-04-23 修回日期:2018-07-12 出版日期:2019-02-25 发布日期:2019-02-25
  • 基金资助:

    国家自然科学基金(61501187)

Link representation and prediction
based on the simplest subgraph

SHANG Zhenhao,CHENG Hua,FANG Yiquan   

  1. (School of Information Science and Engineering,East China University of Science and Technology,Shanghai 200237,China)
  • Received:2018-04-23 Revised:2018-07-12 Online:2019-02-25 Published:2019-02-25

摘要:

稀疏网络的传统链接预测准确率低,为了捕捉稀疏网络节点间建立链接的可能性,提出基于节点间最短路径的最简子图概念。最简子图反映了节点间的拓扑紧密关系,在采用node2vec节点向量化方法的基础之上,实现了基于最短路径的链接表示,并采取长短期记忆循环神经网络(LSTM)学习长链接节点序列的特征,最终实现链接的分类。实验结果表明,该方法与已有方法相比,在4种不同数据集上的预测AUC值平均提高了11.6%,AP值平均提高了13.3%。

 

关键词: 最短路径, 最简子图, 链接表示, 长短期记忆网络(LSTM), 链接预测

Abstract:

Traditional link prediction methods of sparse networks have low accuracy. In order to capture the possibility of link establishment among sparse network nodes, we propose the concept of simplest subgraph based on the shortest path between a pair of nodes, which reflects the tight topology relationship between nodes. Based on the node2vec node vectorization method, the link representation based on the shortest path is implemented. To complete the link classification task, we use the long short-term memory (LSTM) recurrent neural network to learn the characteristics of long-link node sequences. Compared with existing methods, the proposed method can increase the AUC value on 4 different datasets by 11.6% averagely, and the AP value by an average of 13.3%.


 

Key words: shortest path, simplest subgraph, link representation, LSTM, link prediction