一种基于相似社团和节点角色划分的社交网络用户推荐方案

针对现有的社交网络用户推荐方案中主要考虑个体相似性问题以及节点角色无层次差别的问题，提出一种基于相似社团和节点角色划分的推荐方案。在传统的用户相似度计算基础上，从社团结构和属性两方面，综合考虑社团间联系的紧密程度和社团用户兴趣爱好相似程度，提出一种社团相似度的计算方法；其次，从用户节点所在的社团内部和外部2个维度度量节点间紧密度，并据此度量节点的社会影响力，进而将它们划分成不同角色，实现用户推荐的差异化。通过新浪微博真实社交数据对方案进行验证，实验结果表明，该方案适用于存在社团现象的社交网络层次化用户推荐，并具有良好的推荐效果。

A user recommendation scheme based on similar community and node role division in social network

In view of current research on user recommendation mainly considering similarity of node pair and ignoring role level difference of users, we introduce a new way based on similar community and node role division. At first, relying on the similarity of node pair, we put forward a method to calculate the similarity of community pairs from two perspectives:community structure and attributes of users. Secondly, according to the measurement to external and internal tightness of every node in community, we analyze the users''social influence and divide them into different roles in order to recommend friends discriminatively. Finally, we select Sina Weibo data to verify our method. Experiment results show that the scheme performs well and is suitable for user recommendation where there are communities in the social network and users can be divided into roles of different levels.