CREDIT CARD FRAUD DETECTION USING DEEP LEARNING TECHNIQUES

CREDIT CARD FRAUD DETECTION USING DEEP LEARNING TECHNIQUES

Asst. Prof. Satyanarayana, M. Prasanth Paul, B. Sai Neha, B. Keerthana, B. Apoorva Chowdary

CSE AI&ML, Malla Reddy University, Hyderabad

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract -In order to effectively detect credit card fraud, new methods must be developed given its rising prevalence in today's digital environment. We provide a novel deep learning approach application in this work to deal with this urgent problem. Our suggested approach aims to achieve greater accuracy in recognising fraudulent transactions while minimising false positives by leveraging the capability of convolutional neural networks (CNNs). To overcome the challenge of imbalanced data, we carefully preprocess and balance the dataset. Evaluation metrics such as accuracy precision, recall, and F1 score have been employed to assess the model efficiency incorporating convolutional neural network layers, our model surpasses traditional methods by capturing complex dependencies and uncovering subtle patterns within the data. This advanced level of analysis enables more precise and reliable identification of fraudulent transactions.

Key Words: Convolutional Neural Network, Accuracy Precision, Recall, F1 Score