Feature ranking methods based on information entropy with Parzen windows

by: J Biesiada, W Duch, A Kachel, K Maczka, S Palucha

International Conference on Research in Electrotechnology and Applied Informatics (REI'05) (2005), pp. 109-119.

View FullText article

CiteSeer (at Penn State), CiteSeer (at MIT), CiteSeer (at Zurich), CiteSeer (at Singapore)

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

sona

Find related articles with these CiteULike tags

artificial-data, feature-ranking, real-data, survey

Abstract

A comparison between several feature ranking methods used on artificial and real dataset is presented. Six ranking methods based on entropy and statistical indices are considered. The Parzen window method for estimation of mutual information and other indices gives similar results as discretization based on the separability index, but results strongly dependent on the # smoothing parameter. The quality of the feature subsets with highest ranks is evaluated by using decision tree, Naive Bayes...

@inproceedings{biesiada05feature, abstract = {A comparison between several feature ranking methods used on artificial and real dataset is presented. Six ranking methods based on entropy and statistical indices are considered. The Parzen window method for estimation of mutual information and other indices gives similar results as discretization based on the separability index, but results strongly dependent on the \# smoothing parameter. The quality of the feature subsets with highest ranks is evaluated by using decision tree, Naive Bayes...}, address = {Katowice-Krak\'{o}w, Poland}, author = {Biesiada, J. and Duch, W. and Kachel, A. and Maczka, K. and Palucha, S. }, booktitle = {Proceedings of the 9th International Conference on Research in Electrotechnology and Applied Informatics (REI'05)}, citeulike-article-id = {2883797}, journal = {International Conference on Research in Electrotechnology and Applied Informatics (REI'05)}, keywords = {artificial-data, feature-ranking, real-data, survey}, pages = {109--119}, posted-at = {2008-06-11 21:48:59}, priority = {3}, title = {Feature ranking methods based on information entropy with Parzen windows}, url = {http://citeseer.ist.psu.edu/742331.html}, year = {2005} }

RIS record

TY - CONF ID - biesiada05feature L3 - citeulike-article-id:2883797 TI - Feature ranking methods based on information entropy with Parzen windows T2 - Proceedings of the 9th International Conference on Research in Electrotechnology and Applied Informatics (REI'05) JF - International Conference on Research in Electrotechnology and Applied Informatics (REI'05) SP - 109 EP - 119 AD - Katowice-Kraków, Poland N2 - A comparison between several feature ranking methods used on artificial and real dataset is presented. Six ranking methods based on entropy and statistical indices are considered. The Parzen window method for estimation of mutual information and other indices gives similar results as discretization based on the separability index, but results strongly dependent on the # smoothing parameter. The quality of the feature subsets with highest ranks is evaluated by using decision tree, Naive Bayes... KW - artificial-data KW - feature-ranking KW - real-data KW - survey AU - Biesiada, J AU - Duch, W AU - Kachel, A AU - Maczka, K AU - Palucha, S PY - 2005/// UR - http://citeseer.ist.psu.edu/742331.html ER -

RIS BibTeX