Abstract

Mobile information retrieval throws up exciting opportunities for research due to the complexities in the domain like small screen size, limited processing speed and low power of operation. While some of the above issues can be overcome due to the advancement of technology, issues like information overload due to the volume of the content that can be retrieved by the search engines will still persist. To improve the retrieval efficiency of content retrieved by the search engines, there is a need for methods that are effective and personalized. This paper reports on a study on mobile information retrieval and proposes a term frequency based snippet clustering algorithm. The algorithm combines fuzzy re-ranking methods and the term frequency measures for the pre processing and post processing of snippet content. The work has been validated and shows considerable promise. The proposed algorithm is unique in that it combines a statistical retrieval method with a fuzzy relational system and is based on a personal user's model.

Paper Details