Credit card financial fraudster activity is a critical issue in today's technological era for financial institutions and their customers. In this study, we investigate the use of multiple machine learning models such as support vector classifier, logistic regression, K-Nearest Neighbor (KNN) classifier, random forest classifier, decision tree classifier, a machine learning algorithm, to detect fraudulent credit card transactions in a dataset where only a tiny fraction of transactions is fraudulent and need to investigate the detection with at most accuracy. To address the issue of imbalanced data, we employ under-sampling of the majority class and oversampling of the minority class. Our results show that the logistic regression, random forest classifier, support vector classifier, decision tree classifier, and KNN classifier model can achieve 99% accuracy in detecting fraudulent transactions. Additionally, we perform feature importance analysis to identify the most significant variables that contribute to fraud detection. The approach used in this study can contribute to the development of more sophisticated fraud detection models for financial institutions. It also provided the pattern of fraud detection daily on the week chart. Overall, our findings have practical implications for financial institutions and researchers seeking to improve their fraud detection capabilities, which can lead to a more secure and trustworthy financial system for all stakeholders involved.