Research Of Web Mining Based On XML

A web mining system based on XML is proposed based on combination of XML and web mining. This paper discusses types of semi-structured data model, verifies that XML can be expressed by the semi-structured model and investigates the association rule mining problems on the semi-structured data. The frequent item-set mining is the key problem in association rule generating, but traditional methods can not be used on semi-structured data directly. By mining the semi-structure data model(the frequent subtrees ), knowledge discovery of association rules is implemented in this paper. The TreeMiner algorithm is improved by dividing in the process of mining frequent subtrees. The experimentation results show that the dividing is effective and efficient.