A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets
Ant-tree-miner (ATM) has an advantage over the conventional decision tree algorithm in terms of feature selection. However, real world applications commonly involved imbalanced class problem where the classes have different importance. This condition impeded the entropy-based heuristic of existing A...
Saved in:
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Institute of Advanced Engineering and Science
2021
|
Subjects: | |
Online Access: | http://repo.uum.edu.my/27849/1/IJEECS%2020%201%202021%20412%20419.pdf http://repo.uum.edu.my/27849/ http://ijeecs.iaescore.com/index.php/IJEECS/article/view/22253 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.uum.repo.27849 |
---|---|
record_format |
eprints |
spelling |
my.uum.repo.278492020-11-09T00:22:34Z http://repo.uum.edu.my/27849/ A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets Mohd Razali, Muhamad Hasbullah Saian, Rizauddin Yap, Bee Wah Ku-Mahamud, Ku Ruhana QA75 Electronic computers. Computer science Ant-tree-miner (ATM) has an advantage over the conventional decision tree algorithm in terms of feature selection. However, real world applications commonly involved imbalanced class problem where the classes have different importance. This condition impeded the entropy-based heuristic of existing ATM algorithm to develop effective decision boundaries due to its biasness towards the dominant class. Consequently, the induced decision trees are dominated by the majority class which lack in predictive ability on the rare class. This study proposed an enhanced algorithm called hellingerant-tree-miner (HATM) which is inspired by ant colony optimization (ACO) metaheuristic for imbalanced learning using decision tree classification algorithm. The proposed algorithm was compared to the existing algorithm, ATM in nine (9) publicly available imbalanced data sets. Simulation study reveals the superiority of HATM when the sample size increases with skewed class (Imbalanced Ratio < 50%). Experimental results demonstrate the performance of the existing algorithm measured by BACC has been improved due to the class skew in sensitiveness of hellinger distance. The statistical significance test shows that HATM has higher mean BACC scorethan ATM. Institute of Advanced Engineering and Science 2021 Article PeerReviewed application/pdf en http://repo.uum.edu.my/27849/1/IJEECS%2020%201%202021%20412%20419.pdf Mohd Razali, Muhamad Hasbullah and Saian, Rizauddin and Yap, Bee Wah and Ku-Mahamud, Ku Ruhana (2021) A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets. Indonesian Journal of Electrical Engineering and Computer Science, 21 (1). pp. 412-419. ISSN 2502-4752 http://ijeecs.iaescore.com/index.php/IJEECS/article/view/22253 |
institution |
Universiti Utara Malaysia |
building |
UUM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Utara Malaysia |
content_source |
UUM Institutional Repository |
url_provider |
http://repo.uum.edu.my/ |
language |
English |
topic |
QA75 Electronic computers. Computer science |
spellingShingle |
QA75 Electronic computers. Computer science Mohd Razali, Muhamad Hasbullah Saian, Rizauddin Yap, Bee Wah Ku-Mahamud, Ku Ruhana A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets |
description |
Ant-tree-miner (ATM) has an advantage over the conventional decision tree algorithm in terms of feature selection. However, real world applications commonly involved imbalanced class problem where the classes have different importance. This condition impeded the entropy-based heuristic of existing ATM algorithm to develop effective decision boundaries due to its
biasness towards the dominant class. Consequently, the induced decision trees are dominated by the majority class which lack in predictive ability on
the rare class. This study proposed an enhanced algorithm called hellingerant-tree-miner (HATM) which is inspired by ant colony optimization (ACO)
metaheuristic for imbalanced learning using decision tree classification algorithm. The proposed algorithm was compared to the existing algorithm, ATM in nine (9) publicly available imbalanced data sets. Simulation study reveals the superiority of HATM when the sample size increases with skewed class (Imbalanced Ratio < 50%). Experimental results demonstrate the performance of the existing algorithm measured by BACC has been improved due to the class skew in sensitiveness of hellinger distance. The statistical significance test shows that HATM has higher mean BACC scorethan ATM. |
format |
Article |
author |
Mohd Razali, Muhamad Hasbullah Saian, Rizauddin Yap, Bee Wah Ku-Mahamud, Ku Ruhana |
author_facet |
Mohd Razali, Muhamad Hasbullah Saian, Rizauddin Yap, Bee Wah Ku-Mahamud, Ku Ruhana |
author_sort |
Mohd Razali, Muhamad Hasbullah |
title |
A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets |
title_short |
A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets |
title_full |
A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets |
title_fullStr |
A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets |
title_full_unstemmed |
A class skew-insensitive ACO-based decision tree algorithm for imbalanced data sets |
title_sort |
class skew-insensitive aco-based decision tree algorithm for imbalanced data sets |
publisher |
Institute of Advanced Engineering and Science |
publishDate |
2021 |
url |
http://repo.uum.edu.my/27849/1/IJEECS%2020%201%202021%20412%20419.pdf http://repo.uum.edu.my/27849/ http://ijeecs.iaescore.com/index.php/IJEECS/article/view/22253 |
_version_ |
1684655808270303232 |
score |
13.149126 |