Understanding Mahalanobis Distance Criterion for Feature Selection

Maz Jamilah, Masnan and Nor Idayu, Mahat and Ali Yeon, Md Shakaff and Abu Hassan, Abdullah and Nur Zawatil, Ishqi Zakaria and Nurlisa, Yusuf and Norazian, Subari and Ammar, Zakaria and Abdul Hallis, Abdul Aziz (2015) Understanding Mahalanobis Distance Criterion for Feature Selection. In: AIP Conf. Proc., 1660, 050075 : International Conference on Mathematics, Engineering and Industrial Applications (ICoMEIA 2014) , 28-30 May 2014 , Penang. pp. 1-6..

[img] PDF
Understanding Mahalanobis Distance Criterion for Feature Selection.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
[img]
Preview
PDF
Understanding Mahalanobis Distance Criterion for Feature Selection.pdf - Published Version

Download (40kB) | Preview

Abstract

Distance criteria are widely applied in cluster analysis and classification techniques. One of the well known and most commonly used distance criteria is the Mahalanobis distance, introduced by P. C. Mahalanobis in 1936. The functions of this distance have been extended to different problems such as detection of multivariate outliers, multivariate statistical testing, and class prediction problems. In the class prediction problems, researcher is usually burdened with problems of excessive features where useful and useless features are all drawn for classification task. Therefore, this paper tries to highlight the procedure of exploiting this criterion in selecting the best features for further classification process. Classification performance for the feature subsets of the ordered features based on the Mahalanobis distance criterion is included.

Item Type: Conference or Workshop Item (Paper)
Additional Information: ISBN: 978-0-7354-1304-7
Uncontrolled Keywords: Mahalanobis distance; classification; feature selection
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division: Faculty of Electrical & Electronic Engineering
Depositing User: Mrs. Neng Sury Sulaiman
Date Deposited: 26 Oct 2015 02:17
Last Modified: 19 Feb 2018 05:13
URI: http://umpir.ump.edu.my/id/eprint/10467
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item