基于知识图谱和图数据库Neo4j的图书推荐系统研究

Research on Book Recommendation System based on Knowledge Graph and Graph Database of Neo4j

[目的/意义]在图书馆大数据技术快速发展的背景下,为了应对信息冗余和信息过载产生的问题,本文对面向读者的图书推荐系统展开深入研究。[方法/过程]分析当前图书馆信息推荐工作的现状,并就图数据库用于推荐系统的可行性进行了分析。然后本文借助Protégé软件构建了推荐模型的知识图谱结构,形成知识关联模型;并完成RDF数据到图数据库Neo4j的数据映射,对数据结构内容予以可视化呈现。最后结合实体模型提出推荐算法并进行实例分析。[结果/结论]本文拓展了图书馆推荐系统的研究范畴,所提出的建模思路、技术路线和实现方法对于业内图书推荐系统的研发和推广具有一定的借鉴和参考意义。

[Purpose/significance]Under the background of the rapid development of library big data technology,in order to deal with the problem of information overload,this study conducts an further study of the book recommendation system for readers.[Method/process]This study analyzes the situation of current library information recommendation work,and analyzes the feasibility of using graph database in recommendation system.This study constructs the knowledge graph structure of the recommendation model with the help of Protégésoftware to form the knowledge association model;then the data mapping of RDF data to graph database Neo4j is completed,and the content of data structure is visualized.Finally,recommendation algorithms are proposed and analyzed based on the model.[Result/conclusion]This study expands the research scope of library recommendation system,and the proposed model ideas,technical routes and implementation methods have certain reference significance for the research and development and promotion of book recommendation system in the industry.