Improving E-Government Through Sentiment Analysis in the Context of Developing Countries - A Mauritania Case Study

Aug 29, 2026

Uploaded by Sage

Online platforms such as Facebook and X have become key channels for expressing opinions, making sentiment analysis a practical way to extract public attitudes from user-generated text. This study targets the HASSANIYA dialect, a low-resource Arabic variety, and proposes an NLP-based sentiment classification system to support evidence-based e-government decisions in developing countries. A dataset collected from Facebook is preprocessed with Hassaniya-specific and Arabic stemming, then modeled using n-gram features with TF-IDF weighting. Multiple classifiers, including SVM, Logistic Regression, and Random Forest, are evaluated, and SVM achieves the strongest results, validated with an X (Twitter) case study. A sentiment visualization dashboard prototype is also introduced.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Abstract

Introduction

Sentiment analysis approaches

Relevance of social media for e-government

Methodology

Dataset collection and preprocessing

Feature extraction and weighting

Classification algorithms

Results and Validation

Model performance and real-case validation

Prototype and Application

Sentiment dashboard for e-government

Journal of Theoretical and Applied Information Technology ~~15th January 2026. Vol.104. No.1~~ © Little Lion Scientific ISSN: 1992-8645 [www.jatit.org](www.jatit.org) E-ISSN: 1817-3195 IMPROVING E-GOVERNMENT THROUGH SENTIMENT ANALYSIS IN THE CONTEXT OF DEVELOPING COUNTRIES: A MAURITANIA CASE STUDY MOHAMED EL MOUSTAPHA CHRIF1,2, EL BENANY MOHAMED MAHMOUD1, OMAR EL BEQQALI2, MOHAMEDADE FAROUK NANNE1 1CSIDS, Department of Computer Science, Nouakchott University, Nouakchott, Mauritania 2L3IA Laboratory, Department of Computer Science, University Sidi Mohamed Ben Abdellah, Fes, Morocco [E-mail:](E-mail:1mohamedelmoustapha.chrif@usmba.ac.ma)[1](E-mail:1mohamedelmoustapha.chrif@usmba.ac.ma)[mohamedelmoustapha.chrif@usmba.ac.ma](E-mail:1mohamedelmoustapha.chrif@usmba.ac.ma) ABSTRACT New technologies have given rise to online platforms such as Facebook and X (formerly Twitter), which have become de facto channels of communication, enabling people to express their opinions in writing. Analyzing the user-generated data makes it possible to identify opinions and sentiments on specific topics, thereby providing valuable insights for evidence-based decision-making. Therefore, this study focuses on the HASSANIYA dialect, a low-resource variety of Arabic, and proposes a sentiment classification system based on Natural Language Processing (NLP) to support decision-making. In the first step, A dedicated dataset was collected from Facebook, preprocessed using a combination of a Hassaniya-specific stemmer and an Arabic stemmer. Then, different configurations of feature extraction (n-grams) and weighting (TF-IDF) were applied to construct effective feature representations and determine the best classification model. Support Vector Machines (SVM), Logistic Regression (LR), and Random Forest (RF) were used to classify the HASSANIYA dataset. The proposed approach was designed to analyze comments written in both Modern Standard Arabic (MSA) and the HASSANIYA dialect on Facebook and Twitter. The results show that SVM achieved the best performance and was successfully validated through a real case study involving the classification of comments from an X account (Twitter). Finally, this study introduces a prototype dashboard for visualizing sentiment trends, offering a practical tool to support e-government strategies in Mauritania. Keywords: Sentiment Analysis, NLP, Feature Extraction, Decision-Making, e-Government, Social Media, HASSANIYA Dialect, Machine Learning Algorithms 1. INTRODUCTION Sentiment analysis (SA) is a computational method used to identify the sentiment or opinion expressed in a text and to provide an overview of public attitudes toward specific topics [1] . At the sentence level, the process begins by determining if the sentence is subjective or objective [2] . Two principal approaches to sentiment analysis are commonly used: The first is a lexicon-based approach, which uses a collection of predefined words, each annotated according to whether it represents a positive, negative, or neutral sentiment [3] . This approach is used to extract, interpret, and 1[https://datareportal.com/reports/digital-2024-](https://datareportal.com/reports/digital-2024-)[mauritania?rq=mauritania](mauritania?rq=mauritania) determine opinions from a text [4] . The second is a machine-learning based approach [5], which classifies texts using supervised [6] or unsupervised [7] learning techniques. Recently, due to the technology that has emerged, social media has become widely used by citizens around the world [8]. In Mauritania in particular, the number of social media users has increased to more than 25% of the total population 1; Consequently, a continuous stream of data is being generated, reflecting citizens' information or opinions on published actions, while governors have a crucial need to know citizens' reactions, enabling them to improve their services and review the need for their Journal of Theoretical and Applied

FAQs

What problem does the study address for e-government?

The study addresses how to derive citizens’ opinions from social media text and use sentiment information to support decision-making and improve e-government strategies in developing contexts.

How is the HASSANIYA dataset prepared before classification?

Comments are collected from Facebook, then preprocessed using stop-word removal and stemming with a combination of a Standard Arabic stemmer enhanced with HASSANIYA-adapted rules.

Which machine learning model performs best, and how is it validated?

Support Vector Machines (SVM) achieves the best performance. It is validated through a real case study by classifying comments from an X (Twitter) account.

Aug 29, 2026

Uploaded by Sage