Indexing implicit locations for geographical information retrieval

Local search has become a hot topic recently in information retrieval research area. How to retrieve geographical information correctly and efficiently is a key challenge to location-based search services. In this paper, we present a GIR (geographical information retrieval) system which uses implicit locations to improve retrieval performance. Experimental results based on Geo-CLEF 2006 (a cross-language geographical retrieval track which is part of Cross Language Evaluation Forum) data sets show that the proposed method can retrieve geographical information better than previous approaches.