基于关联关系的微博用户可信度分析方法

发布时间：2018-03-30 20:08

本文选题：用户自评价　切入点：关系网络　出处：《计算机应用》2017年03期

【摘要】：随着微博研究的深入,对微博用户可信度的评价成为一个研究热点。针对微博用户可信度评价的问题,提出了一种基于关联关系的用户可信度分析方法。以新浪微博为研究对象,首先从用户的资料信息、交互信息和行为信息三个方面出发,分析了用户的7个相关特征,利用层次分析法(AHP),进而得到用户自评价可信度;然后以用户自评价作为基点,以用户关系网络作为载体,结合用户之间潜在的用户互评关系,通过改进PageRank算法,提出了用户可信度评价模型User-Rank,进而,利用关系网络中其他用户对待分析用户的可信度进行综合评价。大规模的微博真实数据的实验表明,所提方法能够取得良好的用户可信度评价效果。
[Abstract]:With the deepening of Weibo's research, the evaluation of the user credibility of Weibo has become a research hotspot. In this paper, a method of user reliability analysis based on correlation relationship is proposed. Taking Weibo of Sina as the research object, this paper first analyzes seven related characteristics of users from three aspects: user information, interactive information and behavior information. The reliability of user self-evaluation is obtained by using the analytic hierarchy process (AHP), and then the PageRank algorithm is improved by taking user self-evaluation as the basis point, taking the user relationship network as the carrier and combining the potential user evaluation relationship between users. The user credibility evaluation model User-Rankis proposed, and then the comprehensive evaluation is made by other users in the relational network to analyze the users' credibility. The experiments of Weibo's real data show that, The proposed method can achieve a good evaluation effect of user credibility.
【作者单位】：北京信息科技大学智能信息处理研究所;国家计算机网络应急技术处理协调中心;
【基金】：国家自然科学基金资助项目(61370139) 北京市属高等学校创新团队建设与教师职业发展计划项目(IDHT20130519)~~
【分类号】：TP393.092

【参考文献】