This paper provides information on Credit Card Fraud Analysis with the help of some Machine Learning Techniques. The proposed system uses K-Means, Naïve-Bayes and SVM algorithm to detect fraud in credit card transaction. K-means is an algorithm used to solve clustering problem while Naïve-Bayes and SVM algorithm are used for detecting and classifying credit card transaction into fraudulent and non-fraudulent by matching the pattern of fraud as fraud usually has some past pattern associated with it. Result is compared with each other on the basis of fraudulent and not-fraudulent transaction, based on accuracy and shown in the visualization format.