Fractal dimension and vector quantization

Authors:
Krishna Kumaraswamy;Vasileios Megalooikonomou;Christos Faloutsos
Affiliations:
School of Computer Science, Carnegie Mellon University, Pittsburgh, PA;Department of Computer and Information Sciences, Temple University, Philadelphia, PA;School of Computer Science, Carnegie Mellon University, Pittsburgh, PA
Venue:
Information Processing Letters
Year:
2004

Citing 2
Cited 1

Vector quantization and signal compression

Vector quantization and signal compression
Searching in metric spaces

ACM Computing Surveys (CSUR)

Using error-correcting dependencies for collaborative filtering

Data & Knowledge Engineering

Quantified Score

Hi-index	0.89

Visualization

Abstract

We show that the performance of a vector quantizer for a self-similar data set is related to the intrinsic ("fractal") dimension of the data set. We derive a formula for predicting the error-rate, given the fractal dimension and discuss how we can use our result for evaluating the performance of vector quantizers quickly.