Abstract
Spatial-textual data is ubiquitous on websites such as OpenStreetMap or Wikipedia and is published progressively by Open Government Data Initiatives. Those data sources provide the base for novel mashup applications for Location-Based Services. We present a mobile prototype for the San Francisco area based on Preference SQL that allows an intuitive expression of spatial keyword queries. Search can further be extended towards temporal, categorical, and numerical attributes. Our demo provides a fully flexible toolbox for generating extended spatial-textual queries in non-metric spaces which are evaluated using a Best-Matches-Only query model. Spatial relevance is defined by an asymmetric routing distance supporting complex geometries. Textual relevance is determined using the Apache Lucene library.
Authors
Florian Wenzel, Dominik Köppl, Werner Kießling
Publication
Advances in Spatial and Temporal Databases, 13th International Symposium, SSTD 2013, Munich, Germany, August 21-23, 2013, Proceedings, pp. 462-466.