![](/images/graphics-bg.png)
Automatic Arabic text categorization using efficient classification techniques
Other Title(s)
التصنيف التلقائي للنصوص العربية باستخدام تقنيات التصنيف ذات الكفاءة
Dissertant
Thesis advisor
Comitee Members
al-Hasanat, Ahmad Bashir
al-Maani, Mudir Musa
al-Hammuri, Awni Mansur
University
Mutah University
Faculty
Information Technology College
University Country
Jordan
Degree
Master
Degree Date
2015
English Abstract
Arabic language is a complex language that needs special treatment.
However, most previous studies were using statistical methods in Arabic texts classification, and these methods neglect meaning of the terms.
Firstly we built an identical Arabic database, so that they are freely available for research purposes in the Arabic language, then designed a framework for preprocessing Arabic text, which consists of multiple steps and modeling techniques, such as stop word removal and a stemmer to improve the results of Arabic texts categorization.
This thesis focuses on the semantics technique, and proposes a hybrid stemmer for Arabic languages.
Varies techniques are used to implement the Arabic text classifications, and to verify our hybrid stemmer.
These techniques include Latent semantic analysis (LSA) and five machine learning approaches.
LSA used to reduce dimensionality in order to improve the accuracy of categorization systems.
The experiment results showed the effectiveness of our Arabic stemmer in terms of classification accuracy and speed.
The best performance was achieved by combining Singular Value Decomposition (SVD) with cosine similarity measure and Manhattan distance.Finally, we Compared experimentally; Hassanat's distance with Euclidean's distance, Manhattan distance and cosine distance, to choose the best way to calculate the similarity between vectors with five text representation .
Main Subjects
Information Technology and Computer Science
No. of Pages
109
Table of Contents
Table of contents.
Abstract.
Abstract in Arabic.
Chapter One : Introduction.
Chapter Two : The background.
Chapter Three : The proposed Arabic stemmer : Arabic text preprocessing.
Chapter Four : Experiments and results.
References.
American Psychological Association (APA)
al-Awadi, Muhammad Mahmud. (2015). Automatic Arabic text categorization using efficient classification techniques. (Master's theses Theses and Dissertations Master). Mutah University, Jordan
https://search.emarefa.net/detail/BIM-729773
Modern Language Association (MLA)
al-Awadi, Muhammad Mahmud. Automatic Arabic text categorization using efficient classification techniques. (Master's theses Theses and Dissertations Master). Mutah University. (2015).
https://search.emarefa.net/detail/BIM-729773
American Medical Association (AMA)
al-Awadi, Muhammad Mahmud. (2015). Automatic Arabic text categorization using efficient classification techniques. (Master's theses Theses and Dissertations Master). Mutah University, Jordan
https://search.emarefa.net/detail/BIM-729773
Language
English
Data Type
Arab Theses
Record ID
BIM-729773