|
Original research DEFENCE-AGAINST RANSOMWARE: SMART TECHNIQUE TO DETECT AND MITIGATE ATTACKSPages 231-240
Abstract
Innovative AI-powered solutions have emerged in the never-ending fight against ransomware, driven by the need for more precise detection approaches. Here, we present a Convolutional Neural Network (CNN) design that has been fine-tuned for the purpose of more accurately identifying ransomware attacks. We test the suggested model against state-of-the-art machine learning techniques and assess its performance through extensive experimentation and comparison analysis. In addition to our suggested CNN-based model, our evaluation covers four well-known machine learning algorithms: ANN, Random Forest, Decision Tree, and SVM [1]. To evaluate the effectiveness of each method, performance measures including as accuracy, precision, recall, and F1-score are carefully examined. Our comparison analysis shows encouraging outcomes. While traditional ML algorithms do decent work—Random Forest in particular stands out—our suggested CNN-based model outperforms them all across the board in terms of recall, accuracy, precision, and F1-score. With an outstanding accuracy of 0.90, precision of 0.88, recall of 0.89, and F1-score of 0.89, our Optimized CNN demonstrates its competence in reliably and accurately detecting ransomware attacks. The revolutionary power of using convolutional neural network (CNN) models to improve ransomware detection skills is shown by these results. Organizations can strengthen their cybersecurity defenses against ransomware by utilizing optimization techniques and deep learning. This will protect important data assets and ensure operational continuity even when faced with challenges.
Keywords: Ransomware Detection, Convolutional Neural Network (CNN), Machine Learning, Optimization, Performance Evaluation, Cybersecurity.
|
Rahul Papalkar, 
PDF 