• 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 ›› 2022, Vol. 44 ›› Issue (10): 1762-1770.

• 计算机网络与信息安全 • 上一篇    下一篇

基于AFP的有向加权注意力流网络链路预测

马满福1,姜璐娟1,李勇1,张强1,范颜军2,邓晓飞1   

  1. (1.西北师范大学计算机科学与工程学院,甘肃 兰州 730070;
    2.西北工业大学计算机学院,陕西 西安 710129)

  • 收稿日期:2021-11-25 修回日期:2022-03-29 接受日期:2022-10-25 出版日期:2022-10-25 发布日期:2022-10-28
  • 基金资助:
    国家自然科学基金(172161034,61863032);全国高等院校计算机基础教育教学研究(2020-AFCEC-355);甘肃省教育科学规划课题研究(GS[2018]GHBBKZ021);西北师范大学重大科研项目(NWNU-LKZD2021-06)

AFP-based link prediction of directed weighted attention flow network

MA Man-fu1,JIANG Lu-juan1,LI Yong1,ZHANG Qiang1,FAN Yan-jun2,DENG Xiao-fei1   

  1. (1.School of Computer Science and Engineering,Northwest Normal University,Lanzhou 730070;
    2.School of Computer Science,Northwestern Polytechnical University,Xi’an 710129,China)
  • Received:2021-11-25 Revised:2022-03-29 Accepted:2022-10-25 Online:2022-10-25 Published:2022-10-28

摘要: 个性化推荐系统在减轻信息超载、提供个性化服务和辅助用户决策等方面应用广泛,链路预测是个性化推荐的重要方法之一。传统启发式链路预测方法仅考虑网络的图结构特征,缺乏对显式特征和隐式特征信息的应用,且大多数方法基于无向无权网络。针对传统链路预测方法存在的不足,基于集体注意力流网络和R-GCN方法,提出了链路预测算法AFP,将注意力流网络中2节点间不同的边方向抽象为2种边关系类型,并引入注意力机制学习网络中的节点属性和边属性,还综合考虑了网络的图结构特征、显式特征和隐式特征,最后通过评分函数得到三元组成立与否的概率,将链路预测问题转化为一个二分类问题,预测节点间的边属于某个关系类型的可能性。实验结果表明,相比于GCN、GAT等6个基准算法,该算法在准确度、精度和召回率等多个评价指标上均有提升。

关键词: 链路预测, 有向加权图, 注意力机制, R-GCN

Abstract: Personalized recommendation systems are widely used in reducing information overload, providing personalized services, and assisting users in decision-making. Link prediction is one of the important methods of personalized recommendation. Traditional heuristic link prediction methods only consider the graph structure characteristics of the network, and lack the application of explicit and implicit feature information, and most methods are based on undirected and unweighted networks. Aiming at the shortcomings of traditional link prediction methods, this paper proposes a link prediction method AFP based on the collective attention flow network and the R-GCN method. The different edge directions between the two nodes in the attention flow network are abstracted into two types of edge relations. The attention mechanism is introduced to learn the node attributes and edge attributes in the network, and the network's graph structure characteristics, implicit characteristics and explicit features are comprehensively considered. The scoring function is used to get the probability of the establishment of the triple, and the link prediction problem is transformed into a two-category problem, thus predicting the possibility that the edges between nodes belong to a certain type of relationship. Experiments show that, compared with 6 benchmark models such as GCN and GAT, this method improves the accuracy, precision, recall and other evaluation indicators.

Key words: link prediction, directed weighted graph, attention mechanism, R-GCN