融合多源数据的移动社交网络用户行为感知研究

Research on the User Behavior Perception Based on Multisource Data Fusion Under the Environment of Mobile Social Network

用户行为感知是移动社交网络中的关键前沿技术,对服务的提供者和被提供者都有着至关重要的影响。论文提出了移动互联网环境下多源数据的收集和融合的方式,使用高斯混合模型,并引入概率密度函数等多种算法建立多源信息融合模型来对多源信息进行融合、聚类及特征提取,进而提出了基于多源数据的移动社交网络用户行为感知方案,从而使根据移动社交网络用户的特征性完成用户的个性化服务定制和服务推送成为可能。

User＇s behavior perception is the key technology in the mobile social network, which has fatal influence on the service provider and the user. In this paper, there is a way to collect multi-source and fusion methods under the environ- ment of mobile Internet. Using the Gaussian mixture model, establishing multi-source fusion models to integrate these infor- mation, gather them and extract their characteristics through introducing a lot of algorithm, such as the probability density function. Then proposing a lot of plans, many of which are based on the users＂ behaviors from the multi-source information, meanwhile, providing the user＇s personal service design and service push which rely on the mobile website user＇s character- istic becomes possible.