A Small Analysis on Learning to Rank for Information Retrieval

Abstract: Learning to rank for information retrieval has gained a lot of interest in the recent years because, ranking is the central problem in many information retrieval applications, such as document retrieval, collaborative filtering, question answering, multimedia retrieval, text summarization, and online advertising machine translation etc. The extremely large size of the Web documents makes it generally impossible for the common users to find their desired information by surfing the Web. As a consequence, effective and efficient information retrieval has become more important and also search engine (information retrieval system) has become an essential tool for people to locate their needed information. So, we propose novel active learning algorithm that is two stages Expected loss optimization (ELO), which minimizes the expected loss of information and rank the document which is more relevant to the query and gives the user the most informative document instead of displaying all the related documents which is not useful for the user.

Keywords: Ranking, document, query, Expected loss Optimization, Learning to rank.