Browsing by Author "Supervisor: Lamri, Sayad"

Now showing 1 - 4 of 4
  • No Thumbnail Available
    ItemOpen Access
    Diabetic blood glucose prediction using Machine Learning
    (Mohamed Boudiaf University of M'sila, 2025-06-15) Fadoua, Houichi; Supervisor: Lamri, Sayad
    As we reach the end of this journey, the value of artificial intelligence especially Machine Learning and Deep Learning in addressing chronic medical challenges becomes unmistakably clear. This project was never just about predicting blood glucose levels; it was about empowering diabetic patients with a smarter, more personalized way to manage their condition. By blending scientific rigor with practical implementation and user centric design, we’ve taken a meaningful step toward a future where data driven insights support daily healthcare decisions. While there is still much to explore, this work sets a solid foundation for further innovations in blood glucose prediction and intelligent medical assistance systems.
  • Thumbnail Image
    ItemOpen Access
    Forecasting Volatility of Cryptocurrencies Using Machine Learning and Deep Learning
    (Mohamed Boudiaf University of M'sila, 2025-06-15) Mohammed Elamin, Amari; Supervisor: Lamri, Sayad
    This thesis investigates the application of machine learning and deep learning methods into forecasting cryptocurrencies volatility. Using Bitcoin and Ethereum datasets, the performance of the following models was analyzed: Random Forest, XGBoost, LSTM, and GRU. The results show that deep learning models, specifically LSTM, outperforms machine learning models on every metric. These findings suggest that RNN models are better equipped to handle temporal dependencies in time series data that have sequential nature.
  • No Thumbnail Available
    ItemOpen Access
    Gene-disease association using Machine Learning
    (Mohamed Boudiaf University of M'sila, 2025-06-15) Amina, Boufissiou; Supervisor: Lamri, Sayad
    This thesis examines how machine learning can be used to forecast gene-disease associations, focusing on asthma and cardiomyopathy. Traditional genomic approaches often struggle with complex, high-dimensional data, particularly in non-coding regions. To address this, we developed a predictive framework using models such as XGBoost and MLP, aiming for both accuracy and interpretability. Our results show how machine learning (ML) might enhance early diagnosis and aid in clinical decision-making through a multi-disease prediction model, achieving high accuracy for both conditions (79% for asthma and 96% for cardiomyopathy). Future directions include multi-omics integration, clinical validation, and the application of explainable AI to improve trust and usability
  • No Thumbnail Available
    ItemOpen Access
    Performance evaluation of machine learning models for intrusion detection system
    (Mohamed Boudiaf University of M'sila, 2025-06-15) Mohammed Badr Eddine, Laroussi Graine; Supervisor: Lamri, Sayad
    In order to improve cybersecurity, this final year project investigates the use of machine learning (ML) in intrusion detection systems (IDS). Its main goal is to create an intelligent intrusion detection system (IDS) that can accurately identify and categorize network intrusions. The study demonstrates the efficacy of these methods in enhancing threat detection and security resilience through the application and assessment of multiple machine learning models.