A COMPREHENSIVE ANALYSIS ON DETECTING FRAUD IN FINANCIALTRANSACTIONS USING MACHINE LEARNING APPROACHES

Detecting and preventing fraudulent conduct in financial transactions involves keeping a watch on consumer behavior and transactions.For some businesses, it's also a need for Anti-Money Laundering (AML) compliance and an integral aspect of their loss prevention plan.We used 8 articles for this survey analysis.Artificial Bee Colony (ABC), Principal Component Analysis (PCA), K-Nearest Neighbors (KNN), Logistic Regression, Decision Tree, and Random Forest are some of the Machine Learning approaches included in this survey, which aims to identify instances of financial transaction fraud.In order to improve the detection of relevant patterns for fraud, ABC optimization is used to boost feature selection by imitating the behavior of bees.In order to reduce the number of dimensions, PCA is used.This method anonymise sensitive data while keeping important characteristics.Apply Logistic Regression for likelihood-based transaction classification and KNN for similarity metrics.Random Forest is an ensemble technique that combines several decision trees to enhance prediction accuracy and reduce over-fitting, in contrast to Decision Trees that partition data into branches based on the importance of attributes.It's a basic yet effective model.The assessment highlights the algorithms' strengths and shortcomings in identifying fraudulent financial transactions by comparing their performance and efficacy.