基于上下文相似度和社会网络的移动服务推荐方法

发布时间：2019-04-26 14:02

【摘要】：针对传统的基于协同过滤的移动服务推荐方法存在的数据稀疏性和用户冷启动问题,提出一种基于上下文相似度和社会网络的移动服务推荐方法(Context-similarity and Social-network based Mobile Service Recommendation,CSMSR).该方法将基于用户的上下文相似度引入个性化服务推荐过程,并挖掘由移动用户虚拟交互构成的社会关系网络,按照信任度选取信任用户;然后结合基于用户评分相似度计算发现的近邻,分别从相似用户和信任用户中选择相应的邻居用户,对目标用户进行偏好预测和推荐.实验表明,与已有的服务推荐方法 TNCF、SRMTC及CF-DNC相比,CSMSR方法有效地缓解数据稀疏性并提高推荐准确率,有利于发现用户感兴趣的服务,提升用户个性化服务体验.
[Abstract]:Aiming at the problems of data sparsity and user cold start in traditional mobile service recommendation methods based on collaborative filtering, a mobile service recommendation method based on context similarity and social network (Context-similarity and Social-network based Mobile Service Recommendation,) is proposed. CSMSR). In this method, the context similarity of users is introduced into the personalized service recommendation process, and the social network composed of virtual interaction of mobile users is mined, and the trusted users are selected according to the degree of trust. Then, combining the nearest neighbors based on the similarity calculation of the user score, the corresponding neighbor users are selected from the similar users and trusted users, and the preference prediction and recommendation of the target users are carried out. The experimental results show that compared with the existing service recommendation methods TNCF,SRMTC and CF-DNC, the CSMSR method can effectively alleviate the data sparsity and improve the recommendation accuracy, which is beneficial to the discovery of services of interest to users and the improvement of user personalized service experience.
【作者单位】：南京工业大学计算机科学与技术学院;复旦大学;中国人民解放军73677部队;
【基金】：国家自然科学基金(No.61203072) 江苏省重点研发计划(社会发展)(No.BE2015697)
【分类号】：TP391.3

【相似文献】