Rough Set Theory in Intelligent Information Retrieval: A Comprehensive Survey

Today online resources are the largest pool of information in terms of volume, but still users are struggling to get relevant information. Information retrieval systems suffers mainly due to two reasons: first, information overload problem, and second, vagueness and imprecision prevailing in document representations as well as in the information need description by the users. The ability to handle vague, incomplete and imprecise information laid the foundation for applying rough set theory in various domains related to artificial intelligence including information retrieval. In this survey, we have focused on rough set and its generalization models applied in information retrieval. Some existing surveys tried to comprehend rough set based information retrieval models but were restricted in their scope. This study provides systematic and comprehensive elucidation of different rough set based information retrieval models, their basic approaches, key features, strengths and limitations. A comparison of reviewed frameworks is also included for critical analysis.