Abstract

Abstract. We propose a new data structure to search in metric spaces. A metric space is formed by a collection of objects and a distance function defined among them which satisfies the triangle inequality. The goal is, given a set of objects and a query, retrieve those objects close enough to the query. The complexity measure is the number of distances computed to achieve this goal. Our data structure, called sa-tree (“spatial approximation tree”), is based on approaching the searched objects spatially, that is, getting closer and closer to them, rather than the classic divide-and-conquer approach of other data structures. We analyze our method and show that the number of distance evaluations to search among n objects is sublinear. We show experimentally that the sa-tree is the best existing technique when the metric space is hard to search or the query has low selectivity. These are the most important unsolved cases in real applications. As a practical advantage, our data structure is one of the few that does not need to tune parameters, which makes it appealing for use by non-experts.

Citations

1969	R-trees: a dynamic index structure for spatial searching - Guttman - 1984
1927	Modern Information Retrieval - Baeza-Yates, Ribeiro-Neto - 1999
911	Multidimensional binary search trees used for associative searching - Bentley - 1975
472	Voronoi diagrams: A survey of a fundamental geometric data structure - Aurenhammer - 1991
447	M-tree: an efficient access method for similarity search in metric spaces - Ciaccia, Patella, et al. - 1997
306	A guided tour to approximate string matching - Navarro
284	Searching in metric spaces - Chávez, Navarro, et al. - 2001
240	Distance Browsing in Spatial Databases - Hjaltason, Samet - 1999
225	Data structures and algorithms for nearest neighbor search in general metric spaces - Yianilos - 1993
186	Overview of the Third Text REtrieval Conference (TREC-3 - Harman - 1995
159	Near neighbor search in large metric spaces - Brin - 1995
142	Satisfying general proximity/similarity queries with metric trees - Uhlmann - 1991
117	Some approaches to best-match file searching - Burkhard, Keller - 1973
111	A simple algorithm for nearest neighbor search in high dimensions - Nene, Nayar - 1997
110	Distance-based indexing for high-dimensional metric spaces - Bozkaya, Özgsoyoglu - 1997
58	Proximity matching using fixed-queries trees - Baeza-Yates, Cunto, et al.
55	An algorithm for finding nearest neighbors in (approximately) constant average time - Vidal - 1986
42	A new version of the nearestneighbor approximating and eliminating search algorithm (AESA) with linear preprocessing time and memory requirements - Mico, Oncina, et al. - 1994
35	The choice of reference points in best-match file searching - Shapiro - 1977
26	Fixed queries array: A fast and economical data structure for proximity searching - Chávez, Marroquín, et al.
24	Locally lifting the curse of dimensionality for nearest neighbor search (extended abstract - Yianilos - 2000
24	neighbor search in large metric spaces - Near - 1995
22	Dynamic spatial approximation trees - Navarro, Reyes
18	Spaghettis: an array based algorithm for similarity queries in metric spaces - Chávez, Marroquín, et al. - 1999
18	An effective clustering algorithm to index high dimensional metric spaces - Chávez, Navarro
18	A probabilistic spell for the curse of dimensionality - CHÁVEZ, G
17	Voronoi trees and clustering problems - Dehne, Noltemeier - 1987
16	M-tree: An ecient access method for similarity search in metric spaces - Ciaccia, Partella, et al. - 1997
14	A fast branch and bound nearest neighbor classifier in metric spaces - Mic'o, Oncina, et al. - 1996
10	Proximity queries in metric spaces - Ch'avez, Marroqu'in - 1997
10	Monotonous Bisector Trees -- a tool for efficient partitioning of complex schenes of geometric objects - Nolteimer, Verbarg, et al. - 1992
10	Implementing metric trees to satisfy general proximity/similarity queries - Uhlmann - 1991
6	Some approaches to best-match le searching - Burkhard, Keller - 1973
5	An algorithm for nearest neighbors in (approximately) constant average time - Vidal - 1986
3	Voronoi trees and applications - Nolteimer - 1989
3	Dynamic data structures for searching metric spaces - Reyes - 2001
2	Spaghettis: an array-based algorithm for similarity queries in metric spaces - CHÁVEZ, MARROQUíN, et al. - 1999
1	G (2001) A probabilisticspell for the curse of dimensionality - Chávez, Navarro
1	A new version of the nearest-neighbor approximating and eliminating search (aesa) with linear preprocessing-time and memory requirements - Letters - 1996
1	Searching in metric spaces by spatial approximation - Letters - 1994
1	The choice of reference points in best-match le searching - Shapiro - 1977