New deletion method for dynamic spatial approximation trees

Abstract

The Dynamic Spatial Approximation Tree (DSAT) is a data structure specially designed for searching in metric spaces. It has been shown that it compares favorably against alternative data structures in spaces of high dimension or queries with low selectivity. The DSAT supports insertion and deletions of elements. However, it has been noted that eliminations degrade the structure over time. In [8] is proposed a method to handle deletions over the DSAT, which shown to be superior to the former in the sense that it permits controlling the expected deletion cost as a proportion of the insertion cost. In this paper we propose and study a new deletion method, based on the deletions strategies presented in [8], which has demonstrated to be better. The outcome is a fully dynamic data structure that can be managed through insertions and deletions over arbitrarily long periods of time without any reorganization.