The widespread use of location-aware services and technologies which retrieve or answer spatial queries has received much interest in today’s society. An increasing number of popular applications, such as digital maps, make use of spatial databases and associated technologies. One of the most important branches of traditional spatial queries is the reverse nearest neighbour (RNN) search. This search retrieves points of interest that consider the query facility as the nearest facility. Most of the existing works on spatial databases only focus on point of interest retrieval. There is barely any work on a region of interest or neighbourhood retrieval. In this paper, we introduce the concept of a group version of reverse nearest neighbour queries called reverse nearest neighbourhood (RNNH) queries. The RNNH query finds all possible reverse nearest neighbourhoods where all the neighbourhoods consider the query facility as the nearest facility. We propose an efficient algorithm for processing snapshot RNNH queries by using R-tree index. The proposed algorithm incrementally retrieves all reverse nearest neighbourhoods of the query facility. We have conducted exhaustive experiments on both real and synthetic datasets to demonstrate the superiority of the proposed algorithm.
|Number of pages||12|
|Journal||Journal of Ambient Intelligence and Humanized Computing|
|Publication status||Accepted/In press - 2019|
- Influence zone
- Query processing algorithm
- Reverse nearest neighbourhood
- Voronoi diagram