Estimating Local Intrinsic Dimensionality - Archive ouverte HAL Access content directly
Conference Papers Year : 2015

Estimating Local Intrinsic Dimensionality

(1) , (2) , (1) , (3) , (2) , (2) , (4)
1
2
3
4

Abstract

This paper is concerned with the estimation of continuous intrinsic dimension (ID), a measure of intrinsic dimensionality recently proposed by Houle. Continuous ID can be regarded as an extension of Karger and Ruhl’s expansion dimension to a statistical setting in which the distribution of distances to a query point is modeled in terms of a continuous random variable. This form of intrinsic dimensionality can be particularly useful in search, classification, outlier detection, and other contexts in machine learning, databases, and data mining, as it has been shown to be equivalent to a measure of the discriminative power of similarity functions. Several es- timators of continuous ID are proposed and analyzed based on extreme value theory, using maximum likelihood estimation (MLE), the method of moments (MoM), probability weighted moments (PWM), and regularly varying functions (RV). An experimental evaluation is also provided, using both real and artificial data.
Not file

Dates and versions

hal-01159217 , version 1 (02-06-2015)

Identifiers

Cite

Laurent Amsaleg, Oussama Chelly, Teddy Furon, Stéphane Girard, Michael E. Houle, et al.. Estimating Local Intrinsic Dimensionality. 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD'15, ACM, Aug 2015, Sidney, Australia. pp.29-38 ⟨10.1145/2783258.2783405⟩. ⟨hal-01159217⟩
698 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More