Cheshire at GeoCLEF 2007: Retesting text retrieval baselines

Abstract

In this paper we will briefly describe the approaches taken by Berkeley for the main GeoCLEF 2007 tasks (Mono and Bilingual retrieval). The approach this year was to use probabilistic text retrieval based on logistic regression and incorporating blind relevance feedback for all of the runs. Our intent was to establish a baseline result without explicit geographic processing for comparision with future geographic processing approaches. All translation for bilingual tasks was performed using the LEC Power Translator machine translation system. © 2008 Springer-Verlag Berlin Heidelberg.