Aim: The main aim of the research is to detect Credit Card Fraud using Novel Support Vector Machine (SVM) compared with Random Forest (RF). Materials and Methods: When implementing an accurate prediction model it might not be sufficient to just consider one or two parameters. This analysis will be fed to the prediction model. Following the novel SVM algorithm, Random Forest algorithm based on the previously collected datasets, credit card fraud with calculations can be predicted. Result: Comparison is done by using SPSS Software. The Support vector Machine algorithm produces 95.87% whereas Random Forest algorithm produces 94.89% accuracy while detecting credit card fraud on a data set (p > 0.05). Hence Support Vector Machine is better than Random forest. Conclusion: After using iterations to get that by using novel SVM algorithm gets 95.87% (0.95) and Random Forest algorithm gets 94.89% (0.94). So it can be stated that by using novel SVM Algorithm performs with more accuracy than RF Algorithm.