Deal with multiplicity and diversity of relevance factors in XML retrieval
In this paper, we are interested in content oriented XML information retrieval whose aim is to retrieve not a list of relevant documents, but only fragments of document (XML element) relevant to the user information need. Retrieved XML elements must not only contain relevant information but also be at good level of granularity. The coexistence of content and structural information in XML documents makes that multiple and diverse relevance sources condition the selection and the rank of relevant XML element. It is admitted that the consideration of several relevance sources will surely improve quality of results. However, actual XML retrieval approaches consider just some relevance sources and neglect others due to relevance sources heterogeneity. In this paper, we try to identify and classify theses relevance sources and to propose a way to exploit multiple and diverse relevance sources in retrieval process.
XML Retrieval, Relevance criteria, Partial aggregation, Outranking
- Research Reports