Cover Image

A partition-based feature selection method for mixed data: A filter approach

Feature selection is fundamentally an optimization problem for selecting relevant features from several alternatives in clustering problems. Though several algorithms have been suggested, however till this day, there has not been any one of those that has been dubbed as the best for every problem sc...

Full description

Saved in:
Bibliographic Details
Main Authors: Dutt, Ashish, Ismail, Maizatul Akmar
Format: Article
Published: Univ Malaya, Fac of Computer Science and Information Technology 2020
Subjects:
QA75 Electronic computers. Computer science
Online Access:http://eprints.um.edu.my/38175/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.um.eprints.38175
record_format eprints
spelling my.um.eprints.381752022-12-01T07:29:56Z http://eprints.um.edu.my/38175/ A partition-based feature selection method for mixed data: A filter approach Dutt, Ashish Ismail, Maizatul Akmar QA75 Electronic computers. Computer science Feature selection is fundamentally an optimization problem for selecting relevant features from several alternatives in clustering problems. Though several algorithms have been suggested, however till this day, there has not been any one of those that has been dubbed as the best for every problem scenario. Therefore, researchers continue to strive in developing superior algorithms. Even though clustering process is considered a pre-processing task but what it really does is just dividing the data into groups. In this paper we have attempted an improved distance function to cluster mixed data. A similarity measure for mixed data is Gower distance is adopted and modified to define the similarity between object pairs. A partitional algorithm for mixed data is employed to group similar objects in clusters. The performance of the proposed method has been evaluated on similar mixed and real educational dataset in terms of the silhouette coefficient. Results reveal the effectiveness of this algorithm in unsupervised discovery problems. The proposed algorithm performed better than other clustering algorithms for various datasets. Univ Malaya, Fac of Computer Science and Information Technology 2020 Article PeerReviewed Dutt, Ashish and Ismail, Maizatul Akmar (2020) A partition-based feature selection method for mixed data: A filter approach. Malaysian Journal of Computer Science, 33 (2). pp. 152-169. ISSN 0127-9084, DOI https://doi.org/10.22452/mjcs.vol33no2.5 <https://doi.org/10.22452/mjcs.vol33no2.5>. 10.22452/mjcs.vol33no2.5
institution Universiti Malaya
building UM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaya
content_source UM Research Repository
url_provider http://eprints.um.edu.my/
topic QA75 Electronic computers. Computer science
spellingShingle QA75 Electronic computers. Computer science
Dutt, Ashish
Ismail, Maizatul Akmar
A partition-based feature selection method for mixed data: A filter approach
description Feature selection is fundamentally an optimization problem for selecting relevant features from several alternatives in clustering problems. Though several algorithms have been suggested, however till this day, there has not been any one of those that has been dubbed as the best for every problem scenario. Therefore, researchers continue to strive in developing superior algorithms. Even though clustering process is considered a pre-processing task but what it really does is just dividing the data into groups. In this paper we have attempted an improved distance function to cluster mixed data. A similarity measure for mixed data is Gower distance is adopted and modified to define the similarity between object pairs. A partitional algorithm for mixed data is employed to group similar objects in clusters. The performance of the proposed method has been evaluated on similar mixed and real educational dataset in terms of the silhouette coefficient. Results reveal the effectiveness of this algorithm in unsupervised discovery problems. The proposed algorithm performed better than other clustering algorithms for various datasets.
format Article
author Dutt, Ashish
Ismail, Maizatul Akmar
author_facet Dutt, Ashish
Ismail, Maizatul Akmar
author_sort Dutt, Ashish
title A partition-based feature selection method for mixed data: A filter approach
title_short A partition-based feature selection method for mixed data: A filter approach
title_full A partition-based feature selection method for mixed data: A filter approach
title_fullStr A partition-based feature selection method for mixed data: A filter approach
title_full_unstemmed A partition-based feature selection method for mixed data: A filter approach
title_sort partition-based feature selection method for mixed data: a filter approach
publisher Univ Malaya, Fac of Computer Science and Information Technology
publishDate 2020
url http://eprints.um.edu.my/38175/
_version_ 1751536766460887040
score 13.211869

Similar Items