Feelings and thoughts of Arabizi language users in Lebanon territory. An efficient OM model for automatically processing Arabizi language system

Table of Contents

Chapter I

1.1 Introduction

1.2 Problem Statement

1.3 Purpose of the Study

1.4 Research Questions

1.5 Research Hypotheses

1.6 Significance of the Study

1.7 Limitations of the Study

1.8 Challenges of the study

1.9 Research contributions

1.10 Key Terms

1.10.1 Sentiment Analysis

1.10.2 Natural Language Processing (NLP)

1.10.3 Arabizi NLP

1.10.4 Classifier

1.10.5 Big Data

1.10.6 Machine Learning Classifier

1.10.7 Lexicon-based Classifier

1.10.8 Customer Review

1.11 Research Outline

Chapter II

2.1 Literature Review

2.2 Natural Language Processing (NLP)

2.3 Big Data and Sentiment Analysis (SA)

2.4 Approaches to SA

2.4.1 Lexicon-Based Approach

2.4.2 Machine Learning Approach

2.4.3 Hybrid Approach

2.5 Arabizi and the Lebanese Dialect

2.6 Sentiment Analysis and Lebanese Arabizi

Chapter III

3.1 Research Methodology

3.2 Research Design

3.3 Research Sample

3.3.1 The Challenges of Analyzing Arabizi Texts

3.4 Data Preprocessing and Filtering

3.4.1 Removal of reviews with “neutral” sentiment

3.4.2 Ratings’ Encodings

3.4.3 Data splitting for training and testing

3.4.4 Data Cleaning

3.5 Reviews Representation

3.5.1 Selected Features

3.6 Research Tools

3.6.1 Machine Learning Classifier

3.6.2 Lexicon-based Classifier

Chapter IV

3.7 Research Procedure

Chapter IV

4.1 Experiment Preparation

4.2 Data Preprocessing

4.3 Feature Extraction

4.4 Building Classifiers

4.4.1 Machine Learning

4.4.2 Lexicon-based

4.5 Results and Evaluation

Chapter V

5.1 Research Result

5.2 Machine Learning

5.2.1 First phase (Default settings)

5.2.2 Second phase (hyperparameters tuning settings)

5.2.3 Experiment Summary

5.3 Lexicon-based

5.3.1 Experiment Summary

5.4 Discussion

Chapter VI

6.1 Conclusion

6.2 Future Work

Research Objectives and Themes

This study aims to develop and deploy an efficient sentiment analysis model tailored for the Arabizi language system, specifically within the context of customer service reviews in Lebanon. By leveraging both supervised machine learning (using Logistic Regression) and a lexicon-based approach (using the Science of Language and Communication Semantic Analysis System), the research addresses the lack of automated processing tools for this informal, Latin-scripted dialect.

• Sentiment Analysis for Arabizi (informal Arabic in Latin characters).
• Comparison of Machine Learning (Logistic Regression) and Rule-based (Lexicon) classification.
• Processing of user-generated content from social media platforms (Facebook, Google, Zomato).
• Hyperparameter tuning of Logistic Regression models for optimized classification performance.
• Linguistic analysis of common Arabizi challenges such as code-switching, exaggerations, and orthographic variations.

Excerpt from the Book

Summary of Chapters

Chapter I: This chapter provides an introduction to the research, defining the problem of lacking Arabizi sentiment analysis tools and outlining the study's research questions and objectives.

Chapter II: This chapter reviews the literature on Natural Language Processing (NLP), Big Data, and various sentiment analysis approaches, including machine learning and lexicon-based methods.

Chapter III: This chapter details the research methodology, including the design, data collection from social media, and the specific linguistic challenges associated with analyzing Arabizi texts.

Chapter IV: This chapter explains the research procedure, focusing on experiment preparation, data preprocessing, feature extraction, and the construction of both machine learning and lexicon-based classifiers.

Chapter V: This chapter presents the experimental results and evaluations, comparing the performance of different models based on precision, recall, and f1-scores.

Chapter VI: This chapter concludes the research, confirming the applicability of the proposed models and suggesting directions for future research in Arabizi natural language processing.

Keywords

Sentiment Analysis, Natural Language Processing, Arabizi NLP, Classifier, Big Data, Machine Learning, Logistic Regression, Lexicon-based, Customer Review, Computational Linguistics, Opinion Mining, Feature Extraction, Data Preprocessing, Sentiment Classification, Arabic Dialects.

Frequently Asked Questions

What is the core focus of this thesis?

The thesis focuses on building and comparing sentiment analysis tools specifically for the Lebanese Arabizi language system, using customer reviews from public and private service sectors.

What are the primary thematic areas addressed?

The main themes include Natural Language Processing (NLP), Arabizi dialect peculiarities, machine learning classification, lexicon-based rule construction, and big data management within the context of Lebanese customer feedback.

What is the primary research goal?

The goal is to bridge the gap in sentiment analysis for Arabizi by proposing an efficient automated classification system and creating a reliable dataset for future research.

Which scientific methods are utilized?

The study utilizes an experimental, quantitative, and descriptive approach, specifically testing supervised machine learning (Logistic Regression) alongside a rule-based lexicon classifier (SLCSAS).

What is covered in the main section of the paper?

The main part of the paper details the data preprocessing techniques, the construction of BoW and TF*IDF features, the training of classification models, and the comprehensive evaluation of their performance.

Which keywords characterize this research?

Key terms include Sentiment Analysis, Arabizi NLP, Machine Learning, Logistic Regression, Lexicon-based Classification, and Opinion Mining.

What is Arabizi?

Arabizi is an informal slang system used by Arabic speakers to write Arabic using English (Latin) characters, often encountered in computer-mediated communication like social media.

How did the author handle the lack of annotated datasets?

The author performed manual data collection and extraction from Facebook, Google, and Zomato, creating a unique dataset of 2635 reviews which were subsequently preprocessed for training and testing.

How does the Lexicon-based classifier function in this study?

The SLCSAS (Science of Language and Communication Semantic Analysis System) uses manually hand-crafted grammar rules, a dictionary, and a semantic map to categorize sentiment based on specific keywords and linguistic markers found in the text.

What was the outcome of the hyperparameter tuning?

The experiments demonstrated that hyperparameter tuning, particularly for the TF*IDF-based Logistic Regression model, significantly enhanced the classifier's performance, making it the most accurate solution presented in the study.