Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification

Classification has become an important task for automatically classifying the documents to their respective categories. For text classification, feature selection techniques are normally used to identify important features and to remove irrelevant, and noisy features for minimizing the dimensionalit...

Full description

Saved in:

Bibliographic Details
Main Author:	Sharif, Wareesa
Format:	Thesis
Language:	English English English
Published:	2019
Subjects:	QA76 Computer software
Online Access:	http://eprints.uthm.edu.my/135/1/24p%20WAREESA%20SHARIF.pdf http://eprints.uthm.edu.my/135/2/WAREESA%20SHARIF%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/135/3/WAREESA%20SHARIF%20WATERMARK.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-uthm-ep.135
record_format	uketd_dc
spelling	my-uthm-ep.1352021-07-05T02:26:42Z Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification 2019-12 Sharif, Wareesa QA76 Computer software Classification has become an important task for automatically classifying the documents to their respective categories. For text classification, feature selection techniques are normally used to identify important features and to remove irrelevant, and noisy features for minimizing the dimensionality of feature space. These techniques are expected particularly to improve efficiency, accuracy, and comprehensibility of the classification models in text labeling problems. Most of the feature selection techniques utilize document and term frequencies to rank a term. Existing feature selection techniques (e.g. RDC, NRDC) consider frequently occurring terms and ignore rarely occurring terms count in a class. However, this study proposes the Improved Relative Discriminative Criterion (IRDC) technique which considers rarely occurring terms count. It is argued that rarely occurring terms count are also meaningful and important as frequently occurring terms in a class. The proposed IRDC is compared to the most recent feature selection techniques RDC and NRDC. The results reveal significant improvement by the proposed IRDC technique for feature selection in terms of precision 27%, recall 30%, macro-average 35% and micro- average 30%. Additionally, this study also proposes a hybrid algorithm named: Ringed Seal Search-Support Vector Machine (RSS-SVM) to improve the generalization and learning capability of the SVM. The proposed RSS-SVM optimizes kernel and penalty parameter with the help of RSS algorithm. The proposed RSS-SVM is compared to the most recent techniques GA-SVM and CS-SVM. The results show significant improvement by the proposed RSS-SVM for classification in terms of accuracy 18.8%, recall 15.68%, precision 15.62% and specificity 13.69%. In conclusion, the proposed IRDC has shown better performance as compare to existing techniques because its capability in considering rare and informative terms. Additionally, the proposed RSS- SVM has shown better performance as compare to existing techniques because it has capability to improve balance between exploration and exploitation. 2019-12 Thesis http://eprints.uthm.edu.my/135/ http://eprints.uthm.edu.my/135/1/24p%20WAREESA%20SHARIF.pdf text en public http://eprints.uthm.edu.my/135/2/WAREESA%20SHARIF%20COPYRIGHT%20DECLARATION.pdf text en staffonly http://eprints.uthm.edu.my/135/3/WAREESA%20SHARIF%20WATERMARK.pdf text en validuser phd doctoral Universiti Tun Hussein Onn Malaysia Fakulti Sains Komputer dan Teknologi Maklumat
institution	Universiti Tun Hussein Onn Malaysia
collection	UTHM Institutional Repository
language	English English English
topic	QA76 Computer software
spellingShingle	QA76 Computer software Sharif, Wareesa Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification
description	Classification has become an important task for automatically classifying the documents to their respective categories. For text classification, feature selection techniques are normally used to identify important features and to remove irrelevant, and noisy features for minimizing the dimensionality of feature space. These techniques are expected particularly to improve efficiency, accuracy, and comprehensibility of the classification models in text labeling problems. Most of the feature selection techniques utilize document and term frequencies to rank a term. Existing feature selection techniques (e.g. RDC, NRDC) consider frequently occurring terms and ignore rarely occurring terms count in a class. However, this study proposes the Improved Relative Discriminative Criterion (IRDC) technique which considers rarely occurring terms count. It is argued that rarely occurring terms count are also meaningful and important as frequently occurring terms in a class. The proposed IRDC is compared to the most recent feature selection techniques RDC and NRDC. The results reveal significant improvement by the proposed IRDC technique for feature selection in terms of precision 27%, recall 30%, macro-average 35% and micro- average 30%. Additionally, this study also proposes a hybrid algorithm named: Ringed Seal Search-Support Vector Machine (RSS-SVM) to improve the generalization and learning capability of the SVM. The proposed RSS-SVM optimizes kernel and penalty parameter with the help of RSS algorithm. The proposed RSS-SVM is compared to the most recent techniques GA-SVM and CS-SVM. The results show significant improvement by the proposed RSS-SVM for classification in terms of accuracy 18.8%, recall 15.68%, precision 15.62% and specificity 13.69%. In conclusion, the proposed IRDC has shown better performance as compare to existing techniques because its capability in considering rare and informative terms. Additionally, the proposed RSS- SVM has shown better performance as compare to existing techniques because it has capability to improve balance between exploration and exploitation.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Sharif, Wareesa
author_facet	Sharif, Wareesa
author_sort	Sharif, Wareesa
title	Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification
title_short	Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification
title_full	Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification
title_fullStr	Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification
title_full_unstemmed	Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification
title_sort	improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification
granting_institution	Universiti Tun Hussein Onn Malaysia
granting_department	Fakulti Sains Komputer dan Teknologi Maklumat
publishDate	2019
url	http://eprints.uthm.edu.my/135/1/24p%20WAREESA%20SHARIF.pdf http://eprints.uthm.edu.my/135/2/WAREESA%20SHARIF%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/135/3/WAREESA%20SHARIF%20WATERMARK.pdf
_version_	1747830537626583040

Improved relative discriminative criterion using rare and informative terms and ringed seal search-support vector machine techniques for text classification

Similar Items