AI-BASED ANAMOLY DETECTION IN FINANCIAL AND TRANSACTION RECORDS

Abstract

This project presents an AIBased Anomaly Detection in Financial Transactions and Records system that uses Machine Learning to detect fraudulent financial transactions in real time. The proposed system improves transaction security by analyzing multiple transaction parameters and identifying suspicious activities with high accuracy. It ensures secure and intelligent decision-making through Artificial Intelligence and Machine Learning techniques. The system uses a Flask-based web application and an SQLite database to process transactions, classify them as Normal, Suspicious, or High Risk, and automatically perform appropriate security actions such as fraud alerts, user verification, and temporary card freezing. Traditional fraud detection methods mainly rely on predefined rules and manual verification, resulting in delayed detection and reduced accuracy. Therefore, there is a need for an intelligent, secure, and costeffective financial fraud detection system that can protect digital transactions and reduce financial risks in modern banking environments.