An enhanced feature selection technique for classification of group-based holy quran verses

Oyekunle, Adeleke Abdullahi (2018) An enhanced feature selection technique for classification of group-based holy quran verses. Masters thesis, Universiti Tun Hussein Onn Malaysia.

[img]
Preview
Text
24p ADELEKE ABDULLAHI OYEKUNLE.pdf

Download (25MB) | Preview
[img] Text (Copyright Declaration)
ADELEKE ABDULLAHI OYEKUNLE COPYRIGHT DECLARATION.pdf
Restricted to Repository staff only

Download (978kB) | Request a copy
[img] Text (Full Text)
ADELEKE ABDULLAHI OYEKUNLE WATERMARK.pdf
Restricted to Registered users only

Download (101MB) | Request a copy

Abstract

This thesis is about proposing an enhanced feature selection technique for text classification applications. Text classification problem is primarily applied in document labeling. However, the major setbacks with the existing feature selection techniques are high computational runtime associated with wrapper-based FS techniques and low classification accuracy performance associated with filter-based FS techniques. Therefore, in this study, a hybrid feature selection technique is proposed. The proposed FS technique is a combination of JUter-based information gain (JG) and wrapper-based CFS algorithms. The purpose of combining these two FS algorithms is to achieve both high classification accuracy perfonnance (wrapper) at lower computational runtime (filter). The study also developed a group-based Quran dataset to improve on the understanding and analysis of the textual data (Quranic verses). The group-based dataset is a combination of Holy Quran translation and commentary (tafsir). The Quranic verses were selected from two chapters, Surah Al­Baqarah and Surah Al-Anaam. The verses are classified into three categories: Faith, Worship, and Etiquette. In the experiment, six feature selection algorithms were applied: In.formation Gain (JG), Chi-square (CH), Pearson Correlation Coefficient (PCC), RelieJF, Correlation-based (CFS), and the proposed JG-CFS algorithms. The textual data (Quranic verses) were preprocessed using StringtoWordVector with weighted Term Frequency-Inverse Document Frequency (IF-IDF). Meanwhile, the classification phase has involved four algorithms: Nai've Bayes (NB), k-Nearest Neighbor (k-NN), Support Vector Machine (LibSVM), and Decision Trees (148). The experiment results were evaluated based on two established perfonnance metrics in text classification: Accuracy and Area under Receiver Operating Characteristics (ROC) curve (A UC). The proposed hybrid feature selection technique has shown promising results in tenns of Accuracy and Area under Receiver Operating Characteristics (ROC) curve (A UC) by achieving at a lower computational runtime (3.89secs) Accuracy of94.5% and AUC of0.944 with the group-based Quran dataset.

Item Type: Thesis (Masters)
Subjects: H Social Sciences > HB Economic Theory
H Social Sciences > HB Economic Theory > HB131-147 Methodology > HB135-147 Mathematical economics. Quantitative methods. Including econometrics, input-output analysis, game theory
Divisions: Faculty of Computer Science and Information Technology > Department of Web Technology
Depositing User: Mrs. Sabarina Che Mat
Date Deposited: 21 Aug 2022 01:44
Last Modified: 21 Aug 2022 01:44
URI: http://eprints.uthm.edu.my/id/eprint/7549

Actions (login required)

View Item View Item