A machine Learning Technique for detecting Online Fraud Payment

Abstract

Online fraud payment detection is a critical challenge in the e-commerce industry. As online payment methods gain more popularity, the instances of fraudulent activities also increase. To tackle this problem, machine learning algorithms have been employed to develop models that can detect fraudulent transactions. This study explores the use of Python machine learning tools to detect fraudulent online payment transactions. The dataset used in this study is collected from a large e-commerce firm with considerable online transactions. Feature engineering and preprocessing techniques such as one-hot encoding, feature scaling, data cleaning, and data normalization are used to prepare the data for model training. Three machine learning algorithms, namely Random Forest, Support Vector Machine, and Logistic Regression, are utilized to develop fraud detection models. These algorithms are trained using the prepared dataset, and their performance is evaluated using metrics such as precision, recall and F1-score. The results show that the Random Forest algorithm outperforms the other two models, achieving an F1-score of 97.5% in detecting fraudulent transactions. This study demonstrates the effectiveness of using Python machine learning tools to detect online fraud payments and highlights the importance of continuous improvement of fraud detection methodologies for the e-commerce industry.