This study aims to investigate the effectiveness of machine learning methods for malware detection. Nowadays, cyber-attacks are becoming increasingly sophisticated, rendering traditional signature-based detection methods inadequate. Therefore, the use of machine learning algorithms has grown significantly in efforts to detect malicious software. These methods analyze the behaviors, structures, and characteristics of malware files to identify malicious activities. Models were created using machine learning algorithms including OneR, Naive Bayes, SVM, SMO, and Simple Logistic, and their performances were compared. We have evaluated the effectiveness of the methods using various metrics, including precision, accuracy, specificity, and F1 score. The modeling efforts conducted on the TUANDROMD dataset yielded an impressive accuracy rate of 98.364% for both SMO and Simple Logistic in detecting Android malware, demonstrating high performance in identifying malicious software. However, it is important to consider the advantages and limitations of each method. In conclusion, this research enhances our understanding of the effectiveness of machine learning methods for malware detection and serves as a guide for future studies.