摘要
改进了主成分分析法,并将其运用到Web日志的挖掘中,改进了基于主成分分析法的Web页面推荐算法,提出了因子负荷量提取算法,在P(Zk*,X*i)值得确定过程中,采用了实验的方法,通过实验结果的确定当P(Z*k,Xi*)的值为0.6时效果最好。最后,通过具体的实验结果比较了基于主成分分析法的Web页面推荐算法与快速聚类算法的优缺点。
In this paper, the principal component analysis to improve the law and apply it to the web log mining to improve based on principal component analysis of the web page recommended algorithm, load factor extraction algorithm should be determined in the course of using the experiment method for the determination of the experimental results when the value is at best 0.6. Finally, specific comparison between the experimental results of principal component analysis based on the web page recommended algorithm and the advantages and disadvantages of fast clustering algorithm.
出处
《电脑开发与应用》
2009年第7期3-5,共3页
Computer Development & Applications
基金
山西省自然科学基金资助项目(2007011050)
关键词
日志挖掘
Web页面推荐算法
主成分分析法
聚类分析
log mining, Web page recommend algorithm, principal component analysis, clustering analysis
