潘理, 吴鹏, 黄丹华. 在线社交网络群体发现研究进展[J]. 电子与信息学报, 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 引用本文: 潘理, 吴鹏, 黄丹华. 在线社交网络群体发现研究进展[J]. 电子与信息学报, 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 PAN Li, WU Peng, HUANG Danhua. Reviews on Group Detection in Online Social Networks[J]. Journal of Electronics & Information Technology, 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 Citation: PAN Li, WU Peng, HUANG Danhua. Reviews on Group Detection in Online Social Networks[J]. Journal of Electronics & Information Technology , 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 潘理, 吴鹏, 黄丹华. 在线社交网络群体发现研究进展[J]. 电子与信息学报, 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 引用本文: 潘理, 吴鹏, 黄丹华. 在线社交网络群体发现研究进展[J]. 电子与信息学报, 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 PAN Li, WU Peng, HUANG Danhua. Reviews on Group Detection in Online Social Networks[J]. Journal of Electronics & Information Technology, 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 Citation: PAN Li, WU Peng, HUANG Danhua. Reviews on Group Detection in Online Social Networks[J]. Journal of Electronics & Information Technology , 2017, 39(9): 2097-2107. doi: 10.11999/JEIT161192 群体是在线社交网络重要的中观组织。群体发现不仅有重要的理论意义，还推动了在线社交网络的应用与发展，有广泛的应用前景。该文总结论述了在线社交网络群体发现的研究进展。在分析群体形成机理的基础上定义在线社交网络群体，并介绍群体发现问题。根据挖掘群体时采用的不同特征，该文分别阐述基于个体属性特征的群体发现方法和综合属性与结构特征的群体发现方法。随后从特征选取和检测算法两个方面重点介绍了恶意行为群体的发现方法。最后，对群体发现进一步的研究方向进行展望。 在线社交网络 /  群体发现 /  恶意行为群体 Abstract: Groups are important mesoscopic organizations of Online Social Networks (OSNs). Group detection not only has important theoretical significance, but also has a wide range of applications. It promotes the application and development of online social networks. In this paper, group detection technology in online social networks is studied. Based on analyzing the formation mechanism of social groups, the online social network groups is defined and the group detection problem is introduced. According to different features adopted by group detection methods, the methods based on the attribute features only and those based on combination of attribute features and structure features are analyzed, respectively. Especially, it reviews the malicious behavior group detection methods by analyzing their feature selection mechanisms and detection models in detail. Finally, further research direction of group detection in online social networks is prospected. Key words: Online Social Networks (OSN) /  Group detection /  Malicious behavior group

中国科学院电子学研究所, 北京市2702信箱, 邮编：100190

电话：010-58887066 传真：021-64253812 Email： [email protected]

北京仁和汇智信息技术有限公司