一种基于WEB站点的多媒体数据挖掘方法  

An excavating method of multi-media datum based on Web sites

作　　者：涂晓红[1] 林江莉[2] 

机构地区：[1]重庆工学院商贸信息学院,重庆400050 [2]四川大学生物医学工程系,成都610065

出　　处：《西南民族大学学报（自然科学版）》2005年第4期623-628,共6页Journal of Southwest Minzu University(Natural Science Edition)

摘　　要：随着Web技术及Web数据挖掘技术的成熟和发展,对Web中多媒体的挖掘已经开始成为人们研究的一个热点.基于国内现阶段对Web数据挖掘的研究大多停留在对文本的挖掘,而对多媒体挖掘的研究还很少的状况,采用了基于内容的多媒体相联规则方法——最大出现次数算法,对多媒体数据进行挖掘.先建立了挖掘的基本框架,然后具体介绍这种方法,并详细描述了挖掘的基本流程,最后展望了该领域将要作进一步研究的内容.With the development of Web technologies and excavating technologies of Web datum, the excavating of multi-media on Web sites has become a heated field. Because the internal researches of Web datum excavating are still on the phase of text excavating, just a little research of multi-media datum has been carried on. So this paper introduces a new method to excavate multi-media datum by the multi-media relating rules based on the content-the arithmetic of ultimate presenting times. In this paper, the basic frame of excavating is set up first, and then the method and its basic flow is introduced in details. In the end, what needs further research is proposed.

关 键 词：WEB 多媒体 数据挖掘 相联规则 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]