Access Restriction

Author Stolfo, Salvatore J. ♦ Fan, David W. ♦ Lee, Wenke ♦ Prodromidis, Andreas L. ♦ Chan, Philip K.
Source CiteSeerX
Content type Text
File Format PDF
Language English
Subject Domain (in DDC) Computer science, information & general works ♦ Data processing & computer science
Subject Keyword Current Meta-learning Strategy ♦ Initial Experiment ♦ Fraud Detection Domain ♦ Real-world Credit Card Transaction Data ♦ Credit Card Fraud Detection ♦ False Positive Rate ♦ First Step ♦ Skewed Distribution ♦ False Alarm Rate ♦ Fraudulent Transaction ♦ Legitimate One ♦ Balanced Training Data ♦ Fraudulent Credit Card Transaction ♦ Overall Accuracy ♦ Fraud Classifier ♦ Different Classifier ♦ Initial Result ♦ True Positive Rate ♦ Original Data ♦ Meta-learning Technique
Description Of AAAI Workshop on AI Approaches to Fraud Detection and Risk Management
In this paper we describe initial experiments using meta-learning techniques to learn models of fraudulent credit card transactions. Our collaborators, some of the nation's largest banks, have provided us with real-world credit card transaction data from which models may be computed to distinguish fraudulent transactions from legitimate ones, a problem growing in importance. Our experiments reported here are the first step towards a better understanding of the advantages and limitations of current meta-learning strategies. We argue that, for the fraud detection domain, fraud catching rate (True Positive rate) and false alarm rate (False Positive rate) are better metrics than the overall accuracy when evaluating the learned fraud classifiers. We show that given a skewed distribution in the original data, artificially more balanced training data leads to better classifiers. We demonstrate how meta-learning can be used to combine different classifiers (from different learning algorithms) ...
Educational Role Student ♦ Teacher
Age Range above 22 year
Educational Use Research
Education Level UG and PG ♦ Career/Technical Study
Learning Resource Type Article
Publisher Date 1997-01-01