A Quantitative Analysis and Performance Study for Similarity-Search Methods in High-Dimensional Spaces

Explore this paper's citation graph

Summary

It is shown formally that partitioning and clustering techniques for similarity search in HDVSs exhibit linear complexity at high dimensionality, and that existing methods are outperformed on average by a simple sequential scan if the number of dimensions exceeds around 10.

Type
article
Published
1998-08-24
Cited by
1,859
References
40

Keywords

Curse of dimensionality, Computer science, Cluster analysis, Similarity (geometry), Simple (philosophy)

References

Cited by

Related papers