<![CDATA[Because there are so many harmful apps on the market, Android malware detection has emerged as a crucial cybersecurity concern. Using the Drebin dataset, which comprises 5,560 malicious and 9,476 benign applications, this study attempts to assess how well three machine learning algorithms—Neural Network, K-Nearest NeighborI (K-NN), and Decision Tree—detect malware on the Android operating system. The method used involves testing the performance of the three algorithms based on Accuracy, Precision, Recall, and F1-Score. The results show that Neural Network has the highest Accuracy at 98.7%, followed by K-NN with 97.7%, and Decision Tree with 97.5%. Neural Network also excels in Precision, Recall, and F1-Score with values of 98.4%, 98.1%, and 98.2%, respectively. Although the performance differences between the algorithms are not statistically significant, the results indicate that Neural Network offers the best solution for Android malware detection. This study provides guidance in selecting the appropriate algorithm for Android -based security systems.]]>
Copyrights © 2024
