Aggregated search in XML documents: What to retrieve?

Abstract

In this paper, we are interested in aggregated search in XML documents. Our goal is to retrieve the best set of XML elements to be returned. We present a structured information retrieval model based on the Bayesian networks theory. The networks structure provides a natural representation of links between a document, its elements, and its contents. In this model, the user's query starts a propagation process to recover the XML elements. Thus, instead of retrieving a whole document or a list of disjoint elements that are likely to answer partially the query, we attempt to build a virtual document that aggregates a set of elements that are relevant and complementary. We evaluated our approach using the INEX 2009 collection and presented some empirical results for evaluating the impact of the aggregation approach.