Improving results for the INEX 2009 thorough and 2010 efficiency tasks.

Abstract

Information retrieval is the process of storing textual data, i.e., documents, and then retrieving information from these stored documents. Web-based retrieval systems that use XML documents can also retrieve specific portions of the documents which are considered to be relevant to a query. This is called element retrieval which is performed using flexible retrieval (Flex) [5] in this thesis. In this thesis we focus on improving the results of the INEX 2009 Thorough Task and the INEX 2010 Efficiency Task. Experiments are performed, methodology is described, and results are reported. Since flexible retrieval (Flex) depends on Lnu-ltu term weighting [13], experiments which generate the best values for the required parameters (i.e., slope and pivot) are also briefly described. Using the best value of slope, the INEX 2009 Thorough Task and the INEX 2010 Efficiency Task are carried out and the results are reported.