An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information

La The Vinh; Nguyen Duc Thang; Young Koo Lee

doi:10.1109/SAINT.2010.50

An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information

La The Vinh
, Nguyen Duc Thang
, Young Koo Lee

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

46 Citations (Scopus)

Abstract

We present in this paper a comprehensive analysis of the mutual information based feature selection algorithms. We point out the limitations of some recent work in this area then propose an improvement to overcome the weak points. The experiment results confirm that we achieve a better feature sets compared with the two recent developed algorithms, which are Maximum Relevance and Minimum Redundancy (mRMR) and Normalized Mutual Information Feature Selection (NMIFS), in terms of the classification accuracy.

Original language	English
Title of host publication	Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010
Pages	395-398
Number of pages	4
DOIs	https://doi.org/10.1109/SAINT.2010.50
Publication status	Published - 2010
Event	2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010 - Seoul, Korea, Republic of Duration: 19 Jul 2010 → 23 Jul 2010

Publication series

Name	Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010

Conference

Conference	2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010
Country/Territory	Korea, Republic of
City	Seoul
Period	19/07/10 → 23/07/10

Access to Document

10.1109/SAINT.2010.50

Cite this

Vinh, L. T., Thang, N. D., & Lee, Y. K. (2010). An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information. In Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010 (pp. 395-398). Article 5598034 (Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010). https://doi.org/10.1109/SAINT.2010.50

Vinh, La The ; Thang, Nguyen Duc ; Lee, Young Koo. / An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information. Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010. 2010. pp. 395-398 (Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010).

@inproceedings{b6d2b25099744cfeb6846e330deff231,

title = "An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information",

abstract = "We present in this paper a comprehensive analysis of the mutual information based feature selection algorithms. We point out the limitations of some recent work in this area then propose an improvement to overcome the weak points. The experiment results confirm that we achieve a better feature sets compared with the two recent developed algorithms, which are Maximum Relevance and Minimum Redundancy (mRMR) and Normalized Mutual Information Feature Selection (NMIFS), in terms of the classification accuracy.",

author = "Vinh, \{La The\} and Thang, \{Nguyen Duc\} and Lee, \{Young Koo\}",

note = "Copyright: Copyright 2010 Elsevier B.V., All rights reserved.; 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010 ; Conference date: 19-07-2010 Through 23-07-2010",

year = "2010",

doi = "10.1109/SAINT.2010.50",

language = "English",

isbn = "9780769541075",

series = "Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010",

pages = "395--398",

booktitle = "Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010",

}

Vinh, LT, Thang, ND & Lee, YK 2010, An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information. in Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010., 5598034, Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010, pp. 395-398, 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010, Seoul, Korea, Republic of, 19/07/10. https://doi.org/10.1109/SAINT.2010.50

An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information. / Vinh, La The; Thang, Nguyen Duc; Lee, Young Koo.
Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010. 2010. p. 395-398 5598034 (Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information

AU - Vinh, La The

AU - Thang, Nguyen Duc

AU - Lee, Young Koo

PY - 2010

Y1 - 2010

N2 - We present in this paper a comprehensive analysis of the mutual information based feature selection algorithms. We point out the limitations of some recent work in this area then propose an improvement to overcome the weak points. The experiment results confirm that we achieve a better feature sets compared with the two recent developed algorithms, which are Maximum Relevance and Minimum Redundancy (mRMR) and Normalized Mutual Information Feature Selection (NMIFS), in terms of the classification accuracy.

AB - We present in this paper a comprehensive analysis of the mutual information based feature selection algorithms. We point out the limitations of some recent work in this area then propose an improvement to overcome the weak points. The experiment results confirm that we achieve a better feature sets compared with the two recent developed algorithms, which are Maximum Relevance and Minimum Redundancy (mRMR) and Normalized Mutual Information Feature Selection (NMIFS), in terms of the classification accuracy.

UR - https://www.scopus.com/pages/publications/78649238560

U2 - 10.1109/SAINT.2010.50

DO - 10.1109/SAINT.2010.50

M3 - Conference contribution

AN - SCOPUS:78649238560

SN - 9780769541075

T3 - Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010

SP - 395

EP - 398

BT - Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010

T2 - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010

Y2 - 19 July 2010 through 23 July 2010

ER -

Vinh LT, Thang ND, Lee YK. An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information. In Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010. 2010. p. 395-398. 5598034. (Proceedings - 2010 10th Annual International Symposium on Applications and the Internet, SAINT 2010). doi: 10.1109/SAINT.2010.50

An improved maximum relevance and minimum redundancy feature selection algorithm based on normalized mutual information

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this