网页的内容信息对于提高聚类质量来说并不完全够用,针对网络社区网页之间存在的天然链接关系,本文提出了一种挖掘用户标签的增强型社区网页聚类算法.本文采用多种距离度量方法,并挖掘网页链接关系,然后将网页的内容信息相似度和链接关系结合起来进行聚类.实验表明,提出的算法是有效的.
With the development of Internet, textual content is not enough for web clustering sometimes. This paper proposes mining information for social web clustering algorithm. User information of pages is mined, including the link information, tag information. Experimental results show that the proposed social web clustering algorithm using mining information is effective.