Abstract :
Fraud cases through voice calls are on the rise, causing losses to the public. In an effort to address this issue, the application of machine learning using the Support Vector Machine (SVM) algorithm and artificial intelligence (AI) is implemented through an Android application. This application is capable of detecting sounds and sentences identified as potential fraud, providing warnings to users through notifications in the form of sound and vibration. Through this approach, it is expected to offer a proactive solution in protecting users from fraud threats through voice calls. The implemented SVM algorithm achieves an accuracy level of 86%, demonstrating the effectiveness of the proposed solution.