Example-Dependent Cost-Sensitive Classification with Applications in Financial Risk Modeling and Marketing Analytics

Doctoral thesis (2015)

Several real-world binary classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples and not only within classes. However ... [more ▼]

Several real-world binary classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples and not only within classes. However, standard binary classification methods do not take these costs into account, and assume a constant cost of misclassification errors. This approach is not realistic in many real-world applications. For example in credit card fraud detection, failing to detect a fraudulent transaction may have an economical impact from a few to thousands of Euros, depending on the particular transaction and card holder. In churn modeling, a model is used for predicting which customers are more likely to abandon a service provider. In this context, failing to identify a profitable or unprofitable churner has a significant different economic result. Similarly, in direct marketing, wrongly predicting that a customer will not accept an offer when in fact he will, may have different financial impact, as not all customers generate the same profit. Lastly, in credit scoring, accepting loans from bad customers does not have the same economical loss, since customers have different credit lines, therefore, different profit. Accordingly, the goal of this thesis is to provide an in-depth analysis of example-dependent cost-sensitive classification. We analyze four real-world classification problems, namely, credit card fraud detection, credit scoring, churn modeling and direct marketing. For each problem, we propose an example-dependent cost-sensitive evaluation measure. We propose four example-dependent cost-sensitive methods; the first method is a cost-sensitive Bayes minimum risk classifier which consists in quantifying tradeoffs between various decisions using probabilities and the costs that accompany such decisions. Second, we propose a cost-sensitive logistic regression technique. This algorithm is based on a new logistic regression cost function; one that takes into account the real costs due to misclassification and correct classification. Subsequently, we propose a cost-sensitive decision trees algorithm which is based on incorporating the different example-dependent costs into a new cost-based impurity measure and a new cost-based pruning criteria. Lastly, we define an example-dependent cost-sensitive framework for ensembles of decision-trees. It is based on training example-dependent cost-sensitive decision trees using four different random inducer methods and then blending them using three different combination approaches. Moreover, we present the library CostCla developed as part of the thesis. This library is an open-source implementation of all the algorithms covered in this manuscript. Finally, the experimental results show the importance of using the real example-dependent financial costs associated with real-world applications. We found that there are significant differences in the results when evaluating a model using a traditional cost-insensitive measure such as accuracy or F1Score, than when using the financial savings. Moreover, the results show that the proposed algorithms have better results for all databases, in the sense of higher savings. [less ▲]

Example-Dependent Cost-Sensitive Logistic Regression for Credit Scoring

in 2014 13th International Conference on Machine Learning and Applications (2014, December 03)

Several real-world classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples. Credit scoring is a typical example of cost ... [more ▼]

Several real-world classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples. Credit scoring is a typical example of cost-sensitive classification. However, it is usually treated using methods that do not take into account the real financial costs associated with the lending business. In this paper, we propose a new example-dependent cost matrix for credit scoring. Furthermore, we propose an algorithm that introduces the example-dependent costs into a logistic regression. Using two publicly available datasets, we compare our proposed method against state-of-the-art example-dependent cost-sensitive algorithms. The results highlight the importance of using real financial costs. Moreover, by using the proposed cost-sensitive logistic regression, significant improvements are made in the sense of higher savings. [less ▲]

Cost Sensitive Credit Card Fraud Detection using Bayes Minimum Risk

in Proceedings of 12th International Conference on Machine Learning and Applications, ICMLA 2013 (2013), 1

Credit card fraud is a growing problem that affects card holders around the world. Fraud detection has been an interesting topic in machine learning. Nevertheless, current state of the art credit card ... [more ▼]

Credit card fraud is a growing problem that affects card holders around the world. Fraud detection has been an interesting topic in machine learning. Nevertheless, current state of the art credit card fraud detection algorithms miss to include the real costs of credit card fraud as a measure to evaluate algorithms. In this paper a new comparison measure that realistically represents the monetary gains and losses due to fraud detection is proposed. Moreover, using the proposed cost measure a cost sensitive method based on Bayes minimum risk is presented. This method is compared with state of the art algorithms and shows improvements up to 23% measured by cost. The results of this paper are based on real life transactional data provided by a large European card processing company. [less ▲]