Validating AHP, TOPSIS, and ELECTRE with AI for Real-Time Financial Fraud Detection and Prevention

Abstract

In the modern digital financial system, financial transactions are carried out on an unprecedented scale, making it prone to various types of fraud. It is essential that the real-time fraud detection systems must be accurate and interpretable. Three of the most commonly used Multi-Criteria Decision-Making (MCDM) methods, namely the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), Analytic Hierarchy Process (AHP), and Elimination and Choice Expressing Reality (ELECTRE), are integrated in this paper to form a hybrid model for fraud detection and prevention. For better judgment, the system employs an ensemble logic to detect fraudulent activities. In addition, it suggests specific measures for prevention depending on the identified category of hazard. This system works better than single-method models in terms of accuracy and transparency, as it utilizes a Streamlit interface for implementation and synthetic test cases for validation.