Browsing by Author "Rashmi, R"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Credit Card Fraud Detection Using Big Data
    (2018-09-01T12:16:10Z) Rashmi, R; Sampurna, J; Shilpa Shree, N; Swati, Maurya
    Credit card fraud detection is a wide-ranging term for theft and fraud committed using or involving a payment card, such as credit card or debit card, as a fraudulent source of funds in a transaction. Due to the fast growth of E-commerce, use of credit card for online purchases has dramatically increased and it caused and increase in the credit card fraud. It is relevant problem that draws the attention of Machine Learning. In the fraud detection task, there are some peculiarities present, such as the unavoidable condition of a strong class imbalance, the existence of unlabelled transactions, and the large number of records that must be processed. The present paper aims to propose a methodology for automatic detection of fraudulent transactions that tackle all these problems. The methodology is based on a Balanced Random Forest that can be used in supervised and semi-supervised scenarios through a co-training approach, which allows to compensate the class imbalance problem. In FDS, it has described the alert-feedback interaction, which is mechanism of providing recent supervised samples to train/update the classifiers. The main objective of FDS is to identify the fraud as soon as possible in order to take the necessary actions to revert it. Feedbacks play a central role in the proposed learning strategy.