The rise of e-commerce and digital payment systems has been accompanied by an increase in financial fraud, especially involving credit cards. Ensuring the detection of fraudulent activities is crucial to protecting users' financial assets and preserving trust in online transactions. This study introduces a novel method for detecting credit card fraud by integrating machine learning (ML) techniques with a genetic algorithm (GA) for feature selection. Feature selection plays a vital role in improving fraud detection models by pinpointing the most relevant features linked to fraudulent transactions. The effectiveness of the proposed method is assessed using a dataset from European cardholders, a widely used benchmark in this research area. In terms of performance, the Isolation Forest Algorithm exhibits slightly higher accuracy compared to the Local Outlier Factor (LOF) Algorithm. Consequently, based on accuracy alone, the Isolation Forest Algorithm is deemed more effective for this specific dataset. Keywords: Credit card fraud detection, Machine learning (ML), E-commerce.