TUNING SEARCH ENGINE TO FIT XML RETRIEVAL SCENARIO

XML usage is growing to describe documents. Consequently, systems to search in XML collections are necessary. Various proposals of systems for XML retrieval intend to provide solutions to handle XML documents. This paper describes an XML approach based on direct contribution of the components constituting an information need. The search engine is largely configurable in order to be adapted to different context of search. Beyond being globally adapted to a collection of documents an important objective is to define a search engine that can be adapted to different retrieval scenarios and to identify how to adapt it. This paper presents first experiments on INEX testbeds that show how the engine can be adapted to better respond to different retrieval scenarios.