Sentiment Analysis of Libyan Tweets Using Machine Learning Algorithms - Decision Tree vs SVM vs Naive Bayes

Aug 3, 2026

Upload by Aurelia

Sentiment analysis, also known as opinion mining, extracts positive or negative judgments from textual data and supports decision-making using social media as a rich information source. This study builds on the gap in Libyan dialect research by training three machine-learning algorithms on a dataset of Libyan tweets. Model performance is assessed for accuracy, precision, and recall, while annotation quality is evaluated using Cohen’s Kappa reliability with 89.1% observed agreement. Results show the decision tree performs best (72%) compared with SVM (69%) and Naive Bayes (65%).

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Social media and NLP for sentiment analysis

Approaches to sentiment analysis

Study objective and paper structure

Previous Work

Challenges in Arabic dialect sentiment analysis

Machine learning and transformer-based methods

Methodology

Data collection and dataset description

Annotation quality and reliability measurement

Proposed machine learning models and evaluation

MJIT 2024 Malaysian Journal of Industrial Technology SENTIMENT ANALYSIS OF LIBYAN TWEETS USING MACHINE LEARNING ALGORITHMS Ramadan Alsayed Alfared. University of Zawia, Zawia, Libya. [Ramadan.Alfared@zu.edu.ly](Ramadan.Alfared@zu.edu.ly) [Hanan Mohamed Aljarm.](Hanan Mohamed Aljarm.) University of Zawia, Zawia, Libya. [Pg01401068@zu.edu.ly](Pg01401068@zu.edu.ly) *Corresponding author’s [email: Ramadan.Alfared@zu.edu.ly](email: Ramadan.Alfared@zu.edu.ly) | ARTICLE INFO | ABSTRACT | | --- | --- | | Handling Editor: Rahimah Mahat
Article History:
Received 25 December 2023 Received in revised form 12 January 2024
Accepted 5 February 2024 Available online 15 March 2024
Keywords:
Sentiment Analysis; Text classification; machine learning; Twitter; social media; Libyan dialect | Sentiment analysis is a highly active field of study in natural language processing, also known as opinion mining. Social media is a communication tool between internet users. Thus, these platforms become valuable data resources that can be exploited and used efficiently to support decision-making. Many researchers are still working on improving the processing of sentiment analysis in textual data (positive or negative comments) . Although there are several studies in Arabic dialects using the machine learning approach, no prior work has been conducted on the Libyan dialect.
In this paper, a training dataset of Libyan tweets for sentiment analysis is used to train three machine-learning algorithms. The aim is to determine which algorithm has the best accuracy for our dataset. We use Cohen’s Kappa measure to evaluate the quality and measure the reliability of the sentiment annotations which observed agreement is 89.1% . The experiments of the three algorithms showed that the decision tree algorithm achieved better results compared to the other algorithms in terms of accuracy. The results are (72%, 69% and 65%) in (Decision Tree, Support Vector Machine and Naive Bayes) respectively. | 1.0 Introduction Natural language processing (NLP) is a very active domain in Artificial Intelligence (AI) [1] that exploits the most advanced algorithms to allow understanding of human language. NLP is a branch of AI and has several applications including sentiment Analysis [2], Machine Translation [3] and Information Retrieval [4] . In recent years, many users have been interested in social networking platforms like Twitter, Facebook, and Instagram. Social media is a communication tool between internet users, it is considered a good free resource for any NLP applications. The majority use social sites to express their emotions, beliefs or opinions about products, some issues, services, and life, to gain valuable information for decision-making. A machine learning algorithm (MLA) is used to analyse the huge collection of data. There are several works on Arabic sentiment analysis (Dialects) which have gained considerable interest in the research community such as Jordanin [5], Morocan [6], Saudia [7] and Tunisian [8] . Recently Libyan dialect had the one of the attention dialects that is building a dataset for use in many NLP applications such as [9] . In the NLP field, there are three principal approaches such as machine-learning [10], Lexicon-based [11] and hybrid [12] . In this paper, we apply three ML models for the Libyan dialect sentiment analysis (SA) including Naive Bayes, SVM and Decision Tree to evaluate each model and compare between them in terms of accuracy of recall and precision. This comparison determines which model is appropriate for the Libyan dialect. The study includes a dataset of 2296 Libyan dialect tweets randomly retrieved over the period from the 20th of March to the 23rd of July 2023. The remainder of this paper is structured as follows: Section 2 surveys the literature on sentiment analysis. In Section 3, we explore the different types of machine learning algorithmsand sentiment analysis. We present t

FAQs

What is the main goal of the study on Libyan tweets?

To determine which of three machine-learning algorithms achieves the best accuracy for sentiment analysis of Libyan dialect tweets.

Which algorithms are compared in the experiments?

Naive Bayes, Support Vector Machine (SVM), and Decision Tree are compared using accuracy, precision, and recall.

How is the quality and reliability of sentiment annotations evaluated?

Cohen’s Kappa is used to measure reliability, with observed agreement reported as 89.1%.

Aug 3, 2026

Upload by Aurelia