University of Bordj Bou Arreridj

Welcome to the Digital Repository of University Mohamed El Bachir El Ibrahimi - Bordj Bou Arreridj










 

Recent Submissions

Thumbnail Image
Item
Diagnostic de la maladie de Parkinson basé sur l’analyse de la voix et l’application de l’apprentissage profond
(Faculté des sciences et de la technologie, 2025-07-01) Brahimi Romayssa; Bechami NourELHouda
Le diagnostic automatique de la maladie de Parkinson à partir de la voix, en s’appuyant sur l’apprentissage profond, est une méthode moderne et prometteuse pour la détection précoce de cette maladie neurologique. Cette approche se caractérise par son caractère non invasif et son faible coût. Plus particulièrement, un système de diagnostic de MP par la voix consiste à classifier des signaux vocaux en classes N (Normale) et AN (Anormale : Parkinsonien), en utilisant des algorithmes de Machine Learning (ML). Dans ce projet, on propose d’appliquer des algorithmes d’apprentissage profond sur les paramètres acoustiques pour cette tâche de diagnostic. La conception d’un tel système se base sur une phase d’apprentissage permettant la modélisation des différentes classes et une phase de reconnaissance permettant la classification du signal d’entrée en une classe N ou AN. Les résultats d’une étude comparative entre les performances du système proposé et celles d’un système basé sur des algorithmes d’apprentissage automatique tels KNN et SVM, nous ont montré l’efficacité du classificateur SVM en modes indépendant et dépendant du locuteur. Cependant, l’algorithme d’apprentissage profond montre une faible précision, causée par le nombre limité des données de la base d’apprentissage
Thumbnail Image
Item
Artificial intelligence application for diabetes prediction
(Faculté des sciences et de la technologie, 2025-07) CHERFAOUI Mehdi; CHOUITER Fouad
This study investigates multiple approaches for the prediction of type 2 diabetes Based on Biometric signs. Three supervised machine learning models (Logistic Regression, Random Forest, and XGBoost) were developed and evaluated based on their predictive accuracy, feature interpretability, and computational performance. Additionally a fuzzy logic system and a rule-based expert system approaches were implemented to simulate human reasoning and clinical decision-making. The models were applied to the Pima Indians Diabetes dataset and tested using a combination of statistical metrics and visual diagnostics. Results show that while machine learning algorithms outperform in terms of raw accuracy, fuzzy and expert systems offer greater transparency and explainability. This work highlights the complementary strengths of data-driven and rule-based systems in the design of intelligent diagnostic tools for healthcare.
Thumbnail Image
Item
Study of image compression techniques based on deep learning.
(Faculté des sciences et de la technologie, 2025-07-01) Saoudi Khadidja; .Boutahar Amel
This thesis explores various image compression techniques, with particular focus on approaches grounded in deep learning. It presents detailed analysis and comparison between traditional compression algorithms such as JPEG, JPEG 2000, and BPG and modern deep learning methods, including factorized and hyperprior models. While conventional techniques have been widely used for their balance of image quality and bitrate, recent advances in deep learning offer more effective and efficient alternatives. Neural network-based models, in particular, have demonstrated superior capabilities in achieving higher compression rates while maintaining perceptual image quality. This study underscores the strengths of these advanced techniques, establishing deep learning as a promising and powerful direction for the future of image compression.
Thumbnail Image
Item
شعر عبد المالك بومنجل دراسة اسلوبية
(جامعة محمد البشير الابراهمي برج يوعريريج, 2025) سيسون سعاد
الملخص: تسعى هذه الدراسة إلى قراءة النصوص الشعرية التي حفلت بها تجربة الشاعر عبد الملك بومنجل، والتعرف على مضامينها وفهم علاقاتها السياقية وخصائص أساليبها وتحولها من ديوان لآخر بعد عرض أهم المستويات للبنية اللغوية للنصوص الشعرية، واقتضت طبيعة الموضوع أن تتخذ المنهج الأسلوبي التحليلي الإحصائي لتحديد مختلف الظواهر الأسلوبية على مستوى الصوت والتركيب والدلالة التي تمت دراستها لفحص القيمة الفنية والقيمة الجمالية والدلالية. الكلمات المفتاحية: النصوص الشعرية، البنية اللغوية، القيم الجمالية، السياق، الدلالة. Abstract: This study seeks to read the poetic texts that were filled with the experience of the poet Abdelmalek Boumendjel, and to identify their contents and understand their contextual relationships and the characteristics of their styles and their transformation from one collection to another after presenting the most important levels of the linguistic structure of the poetic texts. The nature of the subject required adopting the analytical statistical stylistic approach to determine the various stylistic phenomena at the level of sound, structure and meaning that were studied to examine the artistic value and the aesthetic and semantic value. Keywords: Poetic texts, linguistic structure, aesthetic values, context, meaning.
Thumbnail Image
Item
Implementation of 3D Scanner for Small Businesses
(Faculté des sciences et de la technologie, 2025-06-29) BOUCHERK Oussama
This thesis presents the design and development of a low-cost high-performance 3D scanner based on ultrasonic Time-of-Flight (ToF) technology. The system is composed of a microcontroller-controlled platform equipped with stepper motors and a VL53L0X distance sensor. The scanner captures layer-by-layer measurements of an object by rotating it and lifting the sensor along the vertical axis. The collected data is stored as a point cloud on an SD card and processed using a Python-based software pipeline. The reconstruction phase uses Poisson Surface Reconstruction (PSR) via the Open3D library, with alternatives such as the Ball Pivoting Algorithm (BPA). The complete system integrates mechanical, electronic, and software elements. Tests were carried out on real objects, and the results demonstrated the scanner’s ability to produce usable 3D models for prototyping and educational purposes. Future improvements are proposed to enhance precision, automation, and reconstruction quality.