Abstract:
Frauds in Credit Card are an emerging problem with more consequences in the financial sector and even many techniques have been discovered. The huge volumes of complex data analyzed and automate by applying Data mining techniques successfully. Data mining techniques have also played a vital role in the detection of credit card fraud in online transactions. The main aim of the paper is to design and develop a novel fraud detection method for Transaction Data, with an objective, to analyze the past transaction details of the customers and extract the behavioral patterns. The cardholders are categorized into different groups based on their transaction amount. Banks make use of various machine learning methodologies, past data have been collected and new features are been used for enhancing the predictive power for these transactions. In credit card transactions, the performance of fraud detecting is affected greatly by the sampling approach on every data-set, selection of decision variables and detection techniques used. This proposed work investigates and checks the performance of Support Vector Machines, Decision tree, Random Forest and Logistic Regression on highly skewed credit card fraud data. The European credit card fraud dataset containing around 285,000 transactions have been taken and tested. The above mentioned data mining techniques are applied on the raw and preprocessed data and the performance of these techniques are evaluated based on Accuracy, Sensitivity, Specificity and Precision.
Download full Length
Paper......