Yazar "Al-obaidi, Adham Madrooj Khaleefah" seçeneğine göre listele

Listeleniyor 1 - 1 / 1
  • Küçük Resim
    Yayın
    A new data management system in IOT system
    (İstanbul Gedik Üniversitesi, 2024) Al-obaidi, Adham Madrooj Khaleefah; Gümüş, Tuğbay Burçin
    The proliferation of Internet of Things (IOT) devices has led to an unprecedented growth in data, necessitating effective management and classification strategies. This thesis presents a novel approach to IoT data management by utilizing machine learning techniques, specifically focusing on the classification of IoT devices. We developed a system that employs a Random Forest classifier, renowned for its accuracy and efficiency in handling large datasets with multiple features. To address the challenge of imbalanced datasets, which is common in IoT environments, we integrated Synthetic Minority Over-sampling Technique (SMOTE) with the Random Forest algorithm. This integration enhances the classifier's ability to accurately identify and categorize various types of IoT devices, even when some device types are underrepresented in the dataset. Our methodology involves a thorough analysis of IoT data, preprocessing steps, and the application of SMOTE for data balancing, followed by device classification using the Random Forest classifier. The results demonstrate significant improvements in classification accuracy and provide a scalable solution for managing the diversity and volume of data generated by IoT devices. This study not only contributes to the field of IoT data management but also provides a framework for applying machine learning techniques in similar contexts.