论文部分内容阅读
当前群体发现研究主要利用通联关系挖掘用户群体,未能充分利用网络中所隐含的用户社交关系,致使挖掘的群体不能真实反映用户在社会生活中的群体关系。提出一种基于用户位置信息的群体发现方法,利用序列模式挖掘算法挖掘用户位置规律序列,建立位置序列相关性度量标准,以位置规律相关性揭示用户社会活动的群体关系;结合局部相似性度量方法计算用户通信距离指数,反映用户之间的相识程度;最后采用通信距离指数对位置相关性进行加权计算用户群体相关性,再利用分裂聚类算法挖掘具有通信关系和社交关系的用户群体。实验结果表明,该方法能够有效地挖掘用户社交活动中的通信相关性和位置相关性,体现用户在现实社会活动中的群体关系。
The current group found that the research mainly use the relationship between users to tap the community, failed to make full use of the social network implied in the social relations, resulting in mining groups can not truly reflect the user groups in social life. This paper proposes a group discovery method based on user location information. It uses the sequential pattern mining algorithm to mine the sequence of users ’location rules and establishes the measure of relevance of location sequences to reveal the group relationship of users’ social activities by the relevance of location rules. Combining the local similarity measures The distance index of users is calculated to reflect the degree of acquaintance between users. Finally, the communication distance index is used to calculate the relevance of user groups, and then the split cluster algorithm is used to mine user groups with communication and social relationships. Experimental results show that this method can effectively mine communication relevance and location relevance of users’ social activities, and reflect the group relationship of users in real social activities.