摘要
Intenet的快速增长导致了个性化服务的需求急剧增加。基于页面结构的信息提取与推荐是Web数据挖掘中三大研究领域之一。该研究的关键技术是识别Web页面的组织形式,从中挖掘所需要的个性化页面信息。基于Web数据挖掘的个性化信息推荐系统可以满足互联网未来发展趋势的需要。与传统的以页面为单位的Web信息提取相比,基于页面结构分区的信息推荐更符合实际情况,粒度优势明显。以一组数据为实例阐述了基于Web挖掘的协同过滤推荐算法是如何进行数据表示、近邻查询以及产生推荐页面分区信息的。
With development of Internet,personalized service demand rapidly increased.Information extraction and recommendation based web page structure is one of three web data mining's research fields.Key technology of the research is how to recognize web page's organization form and mine the needed information.Personalized recommender system based web data mining can meet the need of the Internet's future development trends.Compared with based on web page,the based web block is more accordant to the fact and the advantage of granularity is evident.In this paper,with one set of data as an example collaborative filtering recommendation algorithm was elaborated based on web data mining how to work in the progress of data expression,neighbors queries and recommend generation.
出处
《计算机技术与发展》
2010年第6期67-69,73,共4页
Computer Technology and Development
基金
广西自然科学基金(桂科自0640069)
关键词
WEB数据挖掘
推荐系统
协同过滤
页面分区
个性化信息
web data mining
recommendation system
collaborative filterring
web block
personalized information