Abstract: In recent years，blog websites as new network media are more and more paid attention.An algorithm on community detection in blog is presented.Because the bloggers with close hobbies and interests will read and comment on the similar posts，which makes it possible to find out the potential communities which are composed of these bloggers who have the similar interests.This algorithm finds the communities in blog hierarchically and generates the hierarchical structure of the communities.The experiments show that this algorithm is feasible and effective.

摘要： 近年来，blog网站作为新型的网络媒体得到越来越多的个人和企业的关注。针对blog数据，提出了一种社区识别的算法。该算法认为具有相近兴趣、爱好的blogger会去阅读和评论类似的post，利用blog网站中各blogger对post的评论信息可以找出这些blogger因共同兴趣而形成的潜在社区。该算法分层识别blog空间中的社区，形成社区的层次结构。实验结果显示该算法是可行且有效的。