Comparative Analysis of Continuous Entropy Estimation with Different Unsupervised Discretization Methods

DOI

10.2991/iccsee.2013.94How to use a DOI?

Keywords

continuous entropy estimation, probability density distribution, unsupervised discretization

Abstract

In this paper, we compare and analyze the performances of nine unsupervised discretization methods, i.e., equal width discretization (EWD), equal frequency discretization (EFD), k-means clustering discretization (KMCD), ordinal discretization (OD), fixed frequency discretization (FFD), nondisjoint discretization (NDD), proportional discretization (PD), weight proportional discretization (WPD), mean value and standard deviation discretization (MVSDD), based on the application of continues entropy estimation. Firstly, we give the detailed description about the concept of continuous entropy estimation. Then, we introduce the nine different unsupervised discretization methods. Finally, we conduct the estimation of continuous entropy based on 15 probability density distributions, i.e., Beta, Cauchy, Central Chi-Squared, Exponential, F, Gamma, Laplace, Logistic, Lognormal, Normal, Rayleigh, Student’s-t, Triangular, Uniform, and Weibull distributions. The experimental results show that in comparison with the sophisticated discretization methods-OD, FFD, NDD, PD, and WPD, EWD and EFD can also the considerable estimation performances. Moreover, we also illustrate the relationship between the size of training dataset and the estimation performance.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jian Fang
AU  - Li-Na Sui
AU  - Hong-Yi Jian
PY  - 2013/03
DA  - 2013/03
TI  - Comparative Analysis of Continuous Entropy Estimation with Different Unsupervised Discretization Methods
BT  - Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013)
PB  - Atlantis Press
SP  - 367
EP  - 370
SN  - 1951-6851
UR  - https://doi.org/10.2991/iccsee.2013.94
DO  - 10.2991/iccsee.2013.94
ID  - Fang2013/03
ER  -