With development of internet and database technology, web mining has got more and more attentions from information science domain. This paper proposes a document clustering technique based on term clustering and association rules. In this technique, extract words from document collection firstly, then construct term clustering according to AMI(Average Mutual Information) between terms, document VSM(Vector Space Model) is represented by term clustering, and use association rules to mine document clustering. Experiment results show that performance and clustering quality of this technique are improved than those of traditional methods in the clustering process.