基于社交团体和用户相似度的信息推荐方法

发布时间：2018-07-17 06:23

【摘要】：社交网站与电子商务网站逐步实现信息共享,电子商务网站利用社交网站信息可以增强信息推荐的准确性与可信性。文章提出一种利用社交网站的用户社交网络及博文信息实现基于社交团体和用户相似度的信息推荐方法,该方法利用CNM算法发现用户所处的社会团体,通过基于语义信息的文本相似计算方法计算微博文本相似度,最后,在社团发现和文本相似度计算的基础上计算用户对项目的预测评分,实现信息推荐,并通过线下模拟实现测试该方法的有效性。
[Abstract]:Social networking website and electronic commerce website realize information sharing step by step, and electronic commerce website can enhance the accuracy and credibility of information recommendation by using social network information. In this paper, we propose a method of information recommendation based on the similarity between social groups and users by using the social network and blog information of social networking sites. The CNM algorithm is used to discover the social groups in which users belong. The text similarity is calculated by semantic information based text similarity calculation method. Finally, based on community discovery and text similarity calculation, users' prediction scores on items are calculated, and information recommendation is realized. The effectiveness of the method is tested by offline simulation.
【作者单位】：南京工业大学经济与管理学院;
【基金】：国家自然科学基金项目“技术范式转换预警的理论与方法”的研究成果;项目编号:71473119
【分类号】：G358

【相似文献】