Customizing Local Context Analysis for Farsi Information Retrieval by Using a New Concept Weighting Algorithm

A lot of digital Farsi content has been produced recently in middle-east. Local context analysis (LCA) is an automated query expansion method that adds concepts to the original query based on the initial retrieval using the original query. In our previous works we attempted to tune this method for Farsi language by manipulating three parameters which are number of concepts used for query expansion, number of initially retrieved documents for local feedback and number of passages for concept discovery and weighting. In this paper we seek to further customize this method for Farsi information retrieval. To compare our work to the previous attempts we have used Hamshahri collection and 60. We have experimented with different number of concepts and have also changed the concept weighting algorithm to improve retrieval performance.