计算机技术与发展2017，Vol.27Issue(9)：7-11,5.DOI:10.3969/j.issn.1673-629X.2017.09.002
基于关联规则挖掘的分类随机游走算法
Random-walk Classification Algorithm with Association Rules Mining
摘要
Abstract
Along with the continuous progress of Internet technology and the rapid development of the Internet,people can easily find all kinds of information on the Internet. Users would spend a lot of time to search for information what they are really interested in,which is inefficient. The phenomenon is called information overload which is solved effectively by recommendation system as an effective method. However,the two most popular recommendation technologies in the current recommendation system are content-based recommendation and collaborative filtering recommendation, which cannot handle the problems of cold start and sparsity well. In order to better solve them,categorical random-walk algorithm based on association rules is proposed,which uses association rules to mine the association be-tween user attributes and items and constructs the initial score vectors for new users. It has made up for the shortage of the classic algo-rithm and better handles the cold start problem. The results of experiments prove its effectiveness and accuracy.关键词
推荐系统/关联规则/分类随机游走算法/信息过载Key words
recommendation system/association rules/categorical random-walk/information overload分类
信息技术与安全科学引用本文复制引用
施海鹰..基于关联规则挖掘的分类随机游走算法[J].计算机技术与发展,2017,27(9):7-11,5.基金项目
上海市科委重点资助项目(91330116) （91330116）
