GeoSearcher: GeoSpatial Ranking of Search Engine Results

Abstract

Current search engines build indices based on keyword occurrence and frequency and use this information along with link and usage analysis to rank the results of Boolean query negotiation. This approach is fast, robust, and generic but produces the same ranked order of the results no matter the intent of the user. Some queries are related to physical locations and distances. Queries of this type are common including; finding activities, online browsing for shopping, finding schools, planning trips, and, perhaps, finding the closest Italian restaurant. In this paper, we describe a prototype system that provides dynamic ranking of search engine results for queries with a geospatial dimension based on the URL of the host site. We evaluate this approach using both user queries and random web pages. This work makes a contribution to the retrieval experience on the web by providing an alternative ranking order for search engine results. This means that users with queries with a geospatial concern can more readily exploit the results of general search engine results.