Data structures for accelerating Tanimoto queries on real valued vectors

Authors:
Thomas G. Kristensen;Christian N. S. Pedersen
Affiliations:
Bioinformatics Research Center, Aarhus University, Aarhus, Denmark;Bioinformatics Research Center, Aarhus University, Aarhus, Denmark
Venue:
WABI'10 Proceedings of the 10th international conference on Algorithms in bioinformatics
Year:
2010

Citing 6
Cited 0

Data structures and algorithms for nearest neighbor search in general metric spaces

SODA '93 Proceedings of the fourth annual ACM-SIAM Symposium on Discrete algorithms
Multidimensional binary search trees used for associative searching

Communications of the ACM
M-tree: An Efficient Access Method for Similarity Search in Metric Spaces

VLDB '97 Proceedings of the 23rd International Conference on Very Large Data Bases
A Quantitative Analysis and Performance Study for Similarity-Search Methods in High-Dimensional Spaces

VLDB '98 Proceedings of the 24rd International Conference on Very Large Data Bases
Near Neighbor Search in Large Metric Spaces

VLDB '95 Proceedings of the 21th International Conference on Very Large Data Bases
An Introduction to Chemoinformatics

An Introduction to Chemoinformatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Previous methods for accelerating Tanimoto queries have been based on using bit strings for representing molecules. No work has gone into examining accelerating Tanimoto queries on real valued descriptors, even though these offer a much more fine grained measure of similarity between molecules. This study utilises a recently discovered reduction from Tanimoto queries to distance queries in Euclidean space to accelerate Tanimoto queries using standard metric data structures. The presented experiments show that it is possible to gain a significant speedup and that general metric data structures are better suited than a data structure tailored for Euclidean space on vectors generated from molecular data.