gR*-tree: An Index for Querying Approximate Keywords in Geographic Information System
Zhijun Wang, Ming Du, Jiajin Le · 2009
Many hybrid index structures have been proposed to answer Spatial-Keyword (SK) queries in geographic information system (GIS). To the best of our knowledge, however, few works focus on querying approximate keyword in GIS efficiently. In this paper we propose gR*-tree for Approximate SK queries- ASK queries, whose objective is to find the spatial objects with their textual attributes similar to the user-specified keyword and their locations satisfied with the location requirement. The gR*tree combines R*-tree and inverted lists with the ^-grains of the keywords of the objects. It partitions the objects as well as local q-grams, which are the ^-grains contained in the node of the gR*tree. In addition, we give an algorithm using gR*-tree to prune the search space in terms of both spatial and textual information. Experiments show our approach is efficient because of the reduction of search space.