Modified Decision Tree Learning for Cost-Sensitive Credit Card Fraud Detection Model

Credit card fraudulent transactions are cost-sensitive in nature, where the cost differs in each misclassification transaction. Generally, the classification methods do not work on the cost factor. It considers a constant cost factor for each misclassification. In this paper, it proposes a modified instance-based cost-sensitive decision tree algorithm which reflects on different cost factor for each misclassified transactions. In the proposed algorithm, it implements different instance-based costs into the cost-based impurity measure as well as cost-based pruning approach. Experimentally, it shows that the proposed algorithm performs better in terms of cost savings as compared against classical decision tree algorithms. Additionally, it observes that the smaller trees are generated in minimum computational time.

[1]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[2]  Bruce A. Draper,et al.  Goal-Directed Classification Using Linear Machine Decision Trees , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Reid A. Johnson,et al.  Calibrating Probability with Undersampling for Unbalanced Classification , 2015, 2015 IEEE Symposium Series on Computational Intelligence.

[4]  Kai Ming Ting,et al.  An Instance-weighting Method to Induce Cost-sensitive Trees , 2001 .

[5]  Taghi M. Khoshgoftaar,et al.  Experimental perspectives on learning from imbalanced data , 2007, ICML '07.

[6]  Kejiang Ye,et al.  FFD: A Federated Learning Based Method for Credit Card Fraud Detection , 2019, BigData.

[7]  Björn E. Ottersten,et al.  Cost Sensitive Credit Card Fraud Detection Using Bayes Minimum Risk , 2013, 2013 12th International Conference on Machine Learning and Applications.

[8]  Björn E. Ottersten,et al.  Example-Dependent Cost-Sensitive Logistic Regression for Credit Scoring , 2014, 2014 13th International Conference on Machine Learning and Applications.

[9]  Gunwoo Kim,et al.  Classification cost: An empirical comparison among traditional classifier, Cost-Sensitive Classifier, and MetaCost , 2012, Expert Syst. Appl..

[10]  Qiang Yang,et al.  Decision trees with minimal costs , 2004, ICML.

[11]  Charles Elkan,et al.  The Foundations of Cost-Sensitive Learning , 2001, IJCAI.

[12]  T Wang,et al.  Efficient techniques for cost-sensitive learning with multiple cost considerations , 2013 .

[13]  Yong Hu,et al.  The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature , 2011, Decis. Support Syst..

[14]  David J. Hand,et al.  Statistical fraud detection: A review , 2002 .

[15]  DJ Hand,et al.  Performance criteria for plastic card fraud detection tools , 2008, J. Oper. Res. Soc..

[16]  Fan Min,et al.  Tri-partition cost-sensitive active learning through kNN , 2017, Soft Computing.

[17]  Nuno Vasconcelos,et al.  Cost-Sensitive Support Vector Machines , 2012, Neurocomputing.

[18]  Björn E. Ottersten,et al.  Example-dependent cost-sensitive decision trees , 2015, Expert Syst. Appl..

[19]  John Langford,et al.  Cost-sensitive learning by cost-proportionate example weighting , 2003, Third IEEE International Conference on Data Mining.

[20]  Sunil Vadera,et al.  CSNL: A cost-sensitive non-linear decision tree algorithm , 2010, TKDD.