TY - JOUR
AU - 
AB - (IJACSA) International Journal of Advanced Computer Science and Applications, Anomaly Detection using Unsupervised Methods: Credit Card Fraud Case Study Mahdi Rezapour University of Wyoming United States Abstract—The usage of credit card has increased dramatically Machine learning techniques are primarily methods in due to a rapid development of credit cards. Consequently, credit identifications of frauds. These techniques could be divided card fraud and the loss to the credit card owners and credit cards into two groups: supervised and unsupervised methods. In companies have been increased dramatically. Credit card supervised machine learning techniques, a model would be Supervised learning has been widely used to detect anomaly in trained on a past sample of fraudulent and legitimate credit card transaction records based on the assumption that the transactions in order to classify new transactions as fraudulent pattern of a fraud would depend on the past transaction. or legitimate. In other words, the supervised learning uses the However, unsupervised learning does not ignore the fact that the whole labeled dataset for training. The labels are known since fraudsters could change their approaches based on customers’ card holders did identify the mismatch of a transaction, or an behaviors and patterns. In this study, three 
TI - Anomaly Detection using Unsupervised Methods: Credit Card Fraud Case Study
JF - International Journal of Advanced Computer Science and Applications
DO - 10.14569/ijacsa.2019.0101101
DA - 2019-01-01
UR - https://www.deepdyve.com/lp/unpaywall/anomaly-detection-using-unsupervised-methods-credit-card-fraud-case-GX9ZGsjRc0
DP - DeepDyve
ER -