Semantic relevance ranking for XML keyword search

Abstract

Keyword search is a user-friendly mechanism used to retrieve XML data for web and scientific applications. Unlike text data, XML data contain rich semantics, which are obviously useful for information retrieval. It is observed that most existing approaches for XML keyword search either do not consider relevance ranking or perform relevance ranking using traditional text IR techniques. Based on an in-depth analysis of user information need and XML structural semantics, we propose to rank the relevance between a keyword query and an XML fragment by their semantic similarity. We first present a formula to quantify the concept of semantic similarity and then introduce a novel semantic ranking scheme for XML keyword search. Our extensive experiments demonstrate that the proposed scheme outperforms existing approaches in terms of search quality and achieve high efficiency and scalability.