Performance Optimization of Naïve Bayes Algorithm for Malware Detection on Android Operating Systems with Particle Swarm Optimization

The most popular mobile device operating system today is the Android operating system. With so many users of the Android operating system, there are also more crimes, one of which is malware. Malware can infiltrate or damage the system which is certainly very detrimental. During 2021, there were 1,652,521,839 traffic anomaly cases in Indonesia where 62% of these cases were malware infections. Even though it is pretty serious and pervasive, many smartphone users are ignorant and are not aware that their devices have been infected with malware. One way to detect it is by using a machine learning algorithm, namely Naïve Bayes. However, naïve bayes has a weakness, namely, there is no relationship between attributes so it does not improve the performance of the algorithm. Therefore, it is necessary to optimize the problem of selecting and optimizing the features that will be used using the particle swarm optimization algorithm. Based on the research results, the particle swarm optimization algorithm has proven to be quite good at performing feature selection on datasets for the naïve bayes algorithm by achieving 95% accuracy, 95% precision, and 80% recall.