DEEP LEARNING. AN AUTOMATED IMBALANCED CLOUD BASED FRAUD DETECTION MODEL

      Аннотация: Research focuses on designing and implementing a cloud-based security system utilizing deep and machine learning optimization techniques. The system employs multiple real-time monitoring parameters and achieves high precision, a critical requirement in cloud computing design. Fraudsters often target cloud-based e-commerce and trade websites, making it essential to develop an accurate fraud detection system. Introduces deep learning mechanisms such as Fully Convolutional Neural Networks (FCNN), Convolutional Neural Networks (CNN), and machine learning methods like Support Vector Machine (SVM), Fuzzy logic, and Logistic Regression (LR). The advanced FCNN-GBML (Global Binary Multiclass Learning) methods effectively address the limitations of existing approaches, improving accuracy, decision rates, and reducing false alarm rates in the detection of cloud-based frauds.

Ключевые слова: Cloud Computing, Deep Learning, Software Design Approach, Fully Convolution Neural Networks, Gradient Boosting Machine Learning, Privacy