基于知识图谱的社交网络用户行为研究进展

社交网络包含复杂的结构信息与丰富的语义信息.互联的多类型数据,实体对象的行为关系等问题的研究面临极大的挑战.知识图谱旨在处理用户数据知识及行为信息,发现事物、概念与实体对象间的复杂联系,使事物间关联关系得到清晰说明.首先介绍知识图谱基本知识;其次基于知识图谱,在社交网络中,可视化表示用户的行为关系,对其中的行为知识抽取、行为知识表示、行为知识加工等3种关键技术和研究进展进行综述,实验分析与对比其中的技术模型,并介绍可视化识别技术,运用概率软逻辑识别候选行为知识,提高用户行为关系可视化的准确性;最后介绍用户行为关系可视化在用户信息检索、用户安全评测、行为关联推理等方面的应用,对当前研究存在的挑战进行讨论分析,并对其发展前景进行了展望.

A survey on user behavior of social network based on knowledge graph

Social networks contain complex structural information and rich semantic information. The research of multi-type data of interconnection and the behavioral relationship security of entity objects are facing great challenges. Knowledge graph is designed to process user data knowledge and behavior information, and to find the complex relationship between things, concepts and entity objects, so that the relationship between things can be clearly explained. Firstly, basic knowledge of knowledge graph was introduced. Secondly, based on knowledge graph, the behavioral relationship of users is visualized in the social network, and the key technologies and research progress of the three modules of behavioral knowledge extraction, behavioral knowledge expression, and behavioral knowledge processing are summarized. By experimental analysis and comparison of the technical models, the visual identification technology was introduced. By using probability soft logic to identify candidate behavioral knowledge, the accuracy of user behavior relationship visualization was improved. Finally, the application of user behavior relation visualization in user information retrieval, user security evaluation and behavior correlation analysis is introduced, the challenges of current research are discussed and analyzed, and its development is prospected.