期刊
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
卷 8, 期 2, 页码 308-318出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCSS.2021.3050397
关键词
Clustering algorithms; Social networking (online); Topology; Network topology; Machine learning algorithms; Natural language processing; Training; Community embedding; node belonging community information; node embedding; node influence information; overlapping community detection
资金
- National Natural Science Foundation of China [61802258, 61572326, 61702333]
- Natural Science Foundation of Shanghai [18ZR1428300]
- Shanghai Committee of Science and Technology [17070502800]
This article proposes a more effective node representation strategy by combining local node embedding and global community embedding, improving the effectiveness of learning node representation and community representation. The proposed model can also effectively detect overlapping communities.
Most community detection methods focus on the similarities between detection nodes to achieve community partitioning. Traditional network representation learning methods are also limited to the local context of the central nodes, which results in less truly representative results. This article examines nodes' influence information, nodes' community affiliating information, and similarity of community topologies and proposes a more effective node representation strategy. According to the local node information and global topology in the social network graph, a method of combining local node embedding and global community embedding is also designed. The effectiveness of learning node representation and community representation is improved by our approach. The proposed model can also effectively detect overlapping communities.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据