一种基于二部图和节点角色划分的社交网络推荐方案

发布时间：2018-10-15 16:48

【摘要】：针对现有社交网络用户推荐方案中大规模网络个体相似性计算复杂度高以及个体节点无差异对待的问题,本文提出一种基于二部图和节点角色划分的推荐方案.首先,通过划分重叠群体简化原生社交网络结构,并进一步构建群体-个体二部图模型;其次,通过群体-个体二部图所反映的拓扑特征,结合节点自身属性特征,对个体进行角色划分,提出一种基于群体-个体二部图的角色划分模型;最后,针对大规模网络中计算个体相似性复杂度高的问题,构建基于角色差异下的个体-个体二部图模型,实现层次化、个性化的推荐.实验表明,该方案适用于对社交网络中兴趣广泛度存在差异的个体间进行好友推荐,并在较小规模的二部图上生成目标个体推荐列表,降低了计算个体相似性的复杂度.
[Abstract]:In order to solve the problem of high computational complexity of individual similarity and no difference treatment of individual nodes in existing user recommendation schemes of social networks, this paper proposes a recommendation scheme based on bipartite graph and node role partition. Firstly, the primary social network structure is simplified by dividing overlapping groups, and the bipartite graph model of group-individual is further constructed. A role partitioning model based on group-individual bipartite graph is proposed. Finally, the complexity of computing individual similarity in large-scale networks is high. An individual-individual bipartite graph model based on role difference is constructed to realize hierarchical and personalized recommendation. Experiments show that the proposed scheme is suitable for friend recommendation among individuals with different interests in social networks, and generates a list of recommendations for target individuals on a small scale bipartite graph, which reduces the complexity of calculating individual similarity.
【作者单位】：重庆邮电大学网络与信息安全技术重庆市工程实验室;
【基金】：国家973重点基础研究发展计划(No.2013CB329606) 国家自然科学基金(No.61272400) 重庆市青年人才项目(No.cstc2013kjrc-qnrc40004) 教育部-中国移动研究基金(No.MCM20130351) 重庆市研究生研究与创新项目(No.CYS14146) 重庆市教委科学计划项目(No.KJ1500425) 重庆邮电大学文峰基金(No.WF201403)
【分类号】：TP391.3

【相似文献】