This project is titled “A Time-Frequency Based Suspicious Activity Detection for Anti-Money Laundering.” The rapid growth of digital financial transactions has attracted millions of users worldwide. However, this expansion has also led to more financial crimes, as malicious actors exploit banking systems to launder money by disguising illegal transactions as legitimate ones. To tackle this problem, there is a strong need to implement an automatic real-time detection mechanism for suspicious transactions in financial systems. In response to these challenges, we proposed a new time-frequency based machine learning framework to detect and classify suspicious financial activities. This framework uses statistical feature extraction methods along with Fast Fourier Transform (FFT) to analyze transaction patterns in both time and frequency domains. We then process these features using a Random Forest classifier, which helps learn effective representations of transaction behavior and make accurate classifications.