Consideration of relevance feedback on keyword space for interactive information retrieval

The relevance feedback based on a keyword map is proposed so that a Web interface can be more interactive. There exists vast amount of information in the Web, from which users usually gather information without definite information needs. The relevance feedback techniques have been studied in the field of document retrieval, aiming to generate appropriate queries for users´ information needs. However, conventional relevance feedback techniques are performed on document space, while the resultant queries should be represented in keyword space. In this paper, it is proposed to perform relevance feedback on keyword space. The relevance feedback is supposed to work with interactive keyword map system, which visualizes the relationship between keywords extracted from retrieved results. As the first stop for realizing relevance feedback based on interactive keyword map, this paper also proposes the algorithm for extracting the pair of keywords that reflects a user´s interest from the keyword map. Experimental results are given for showing how the algorithm works on the keyword map that is modified by the user, and for discussing the difference between the RF based on keyword map and conventional RF methods.