Abstract:

Aiming at the semi\|structured characteristics of XML document and the traditional tf\|idf method only considering the frequency of keyword in the document, not considering the lack of semantic information of the nodes in the XML document, we designed the relevance ranking strategies that were designed based on XML keyword searching results via the vector space model. To improve the accuracy of the measure of the node weights, correlation calculation fully considers the distinctive degree of the nodes in the XML document, the clear and direct degree of nodes describing the document so as to provide the most relevant information to users. Experimental results show that the proposed method is effective.

Key words: XML query, semantics, correlation, ranking