RANKING METHODS FOR GEOGRAPHIC INFORMATION RETRIEVAL

Patricia Frontiera
College of Environmental Design
University of California, Berkeley
Berkeley, California, USA, 94720-1839
pattyf@regis.berkeley.edu

Ray R. Larson
School of Information Management and Systems
University of California, Berkeley
Berkeley, California, USA, 94720-4600
ray@sherlock.berkeley.edu

This presentation will describe and presents results from an evaluation of ranking algorithms for Geographic Information Retrieval (GIR) applications. We will demonstrate a probabilistic algorithm for GIR ranking based on logistic regression from samples of the test collection. We also show the effects of different representations of the geographic regions being searched, including minimum bounding rectangles, convex hulls, and complex polygons.