ARCHIVES

Abstract : Over few years, Credit card fraud has become one of the most common types of fraud prevalent issues. The major purpose of this project is to detection of fraudulent activity in the actual world. The recent increase in transactions has resulted with such a huge increase in illegal transactions. The purpose is to obtain things without having to pay for them or to remove money from one account without being authorized. All credit card providing institutions must implement robust detecting fraud systems in order to minimize their losses. The fact that neither cards nor card users must be present in the transaction is among the most difficult parts of running a business. As a result, the retailer has no way of determining whether or not the customer is purchasing the actual cardholder. The accuracy of the proposed scheme is achieved by utilizing the random forest. A random forest algorithm involves analyzing the dataset and the user's current dataset. Finally, improve the correctness of the outcome data and then process some of the offered qualities to identify fraud detection. A thorough review of existing and planned fraud detection methods and a comparison of these strategies, have been conducted. As a result, classification models based on the Random forest technique are applied to the data, and the model's performance is measured using graphical representations of precision, classification accuracy and f1-score.