基于2K-means算法的读者兴趣分类图书自动推荐系统设计

Design of 2K⁃means algorithm based book automatic recommendation system for readers′interest classification

为了能够满足读者的个人兴趣特点和应用需求,提出基于读者兴趣分类的图书自动推荐系统设计思路。介绍了读者兴趣需求的图书自动推荐系统设计理论技术基础,包括数学挖掘、2K-means算法及UML语言。详细分析了基于读者兴趣分类的图书自动推荐系统需求和性能需求,将读者的兴趣与图书类别完成聚类分析,并提取最终聚类所获结果匹配图书类别,建立读者兴趣分类图书自动化推荐模型。引入聚类算法、关联规则算法实现读者感兴趣图书规律的统计分析,从而整合读者的图书信息源并充分发现具有较大价值的信息,最终将与相似性需求相符的图书,采用电子邮件或网页方式,自动推荐给读者。该系统设计能够为读者提供可能感兴趣的图书摘要、馆藏类相关信息,且运行性能良好,具有良好的推广应用前景。

An book automatic recommendation system based on readers′interest classification is proposed to meet the individual interest characteristics and application needs of readers.The design theory and technical basis of book automatic recommendation system for readers′interest needs are introduced,including mathematical mining,2K⁃means algorithm and UML language.The requirements and performance requirements of the book automatic recommendation system based on readers′interest classification are analyzed in detail,the clustering analysis of readers′interest and book category is completed,and the results of the obtained final clustering are extracted to match the book category.The book automatic recommendation model of readers′interest classification is established.The clustering algorithm and association rule algorithm are introduced to realize the statistical analysis of the book rule that readers are interested,so as to integrate the book information sources of readers and fully discover the information of greater value.The books that meet the requirements of similarity will be automatically recommended to readers through E⁃mail or webpage.The design of the system can provide readers with the relevant information of book abstracts and collections that may be interested in.It has good operation performance and good application prospect.