Sentiment analysis for Arabic social media news polarity

Other Title(s)

تحليل المشاعر و القطبية لمنشورات مواقع التواصل الاجتماعي الإخبارية

Dissertant

Kanan, Umran Hassan

Thesis advisor

Kanan, Ghassan

University

Amman Arab University

Faculty

Collage of Computer Sciences and Informatics

Department

Department of Computer Science

University Country

Jordan

Degree

Master

Degree Date

2019

English Abstract

Recently, social media posts have rapidly grown.

Arabic social media news posts have a great influence on social media users.

Detecting the news posts polarity (positive, negative, neutral) is very important.

Through identifying the polarity, which can help us to understand and study the behavior, reaction, acceptance, and interaction of the social media users.

In our proposed research, we are planning to collect data from Arabic social media news pages.

The news posts will be the main unit in our dataset.

Also, planning to build a corpus of manually judged data to be used for training and testing.

Applying natural language processing on our data is very crucial since Natural Language Processing helps the computer to understand and easily manipulate data.

Thus, plan to apply Stop Word removal, Stemming, and Normalization.

In our research, we plan to build several classifiers like (Support Vector Machine, Naïve Bayes, K-Nearest Neighbor, Random frost, and Decision Tree) using our training data set.

To test our accuracy of classifiers, data testing will be used.

These two steps will be done using the open source WEKA tool.

As a result, we plan to categorize the Arabic social media news posts into three different classes and these classes are positive, negative, and neutral.

This research has fairly concluded that SVM has reached the best level of accuracy among other classifiers with a percentage of 83%.

Main Topic

Media and Communication

Topics

No. of Pages

89

Table of Contents

Table of contents.

Abstract.

Abstract in Arabic.

Chapter One : Introduction.

Chapter Two : Literature review.

Chapter Three : Procedures and methodology.

Chapter Four : Results and analysis.

Chapter Five : Conclusions and future work.

References.

American Psychological Association (APA)

Kanan, Umran Hassan. (2019). Sentiment analysis for Arabic social media news polarity. (Master's theses Theses and Dissertations Master). Amman Arab University, Jordan
https://search.emarefa.net/detail/BIM-932348

Modern Language Association (MLA)

Kanan, Umran Hassan. Sentiment analysis for Arabic social media news polarity. (Master's theses Theses and Dissertations Master). Amman Arab University. (2019).
https://search.emarefa.net/detail/BIM-932348

American Medical Association (AMA)

Kanan, Umran Hassan. (2019). Sentiment analysis for Arabic social media news polarity. (Master's theses Theses and Dissertations Master). Amman Arab University, Jordan
https://search.emarefa.net/detail/BIM-932348

Language

English

Data Type

Arab Theses

Record ID

BIM-932348