In medical diagnosis doctors must often determine what medical tests (e.g., X-ray, blood tests) should be ordered for a patient to minimize the total cost of medical tests and misdiagnosis. In this paper, we design cost-sensitive machine learning algorithms to model this learning and diagnosis process. Medical tests are like attributes in machine learning whose values may be obtained at cost (a...

Many algorithms in decision tree learning are not designed to handle numeric valued attributes very well. Therefore, discretization of the continuous feature space has to be carried out. In this article we introduce the concept of cost sensitive discretization as a preprocessing step to induction of a classifier and as an elaboration of the error-based discretization method to obtain an optimal...

This paper explores two techniques for boosting cost-sensitive trees. The two techniques diier in whether the misclassiication cost information is utilized during training. We demonstrate that each of these techniques is good at diierent aspects of cost-sensitive classiications. We also show that both techniques provide a means to overcome the weaknesses of their base cost-sensitive tree induct...

We present an efficient and effective method which extends the Boosting family of classifiers to allow the weighted classes. Typically classifiers do not treat individual classes separately. For most real world applications, this is not the case, not all classes have the same importance. The accuracy of a particular class can be more critical than others. In this paper we extend the mathematica...

We describe a formal framework for diagnosis and repair problems that shares elements of the well known partially observable MDP and cost-sensitive classification models. Our cost-sensitive fault remediation model is amenable to implementation as a reinforcementlearning system, and we describe an instance-based state representation that is compatible with learning and planning in this framework...

We study the costs and benefits of plasticity by evolving agents in environments with different rates of environmental change. Evolution allows both hard-coded strategies and learned strategies, with learning rates varying throughout life. We observe a range of change rates where the balance of costs and benefits are just right for evolving learning. Inside this range, we see two separate strat...

In many classification settings, mistakes incur different application-dependent penalties based on the predicted and actual class labels. Costsensitive classifiers minimizing these penalties are needed. We propose a robust minimax approach for producing classifiers that directly minimize the cost of mistakes as a convex optimization problem. This is in contrast to previous methods that minimize...

While POMDPs provide a general platform for conditional planning under a wide range of quality metrics they have limited scalability. On the other hand, uniform probability conditional planners scale very well, but many lack the ability to optimize plan quality metrics. We present an innovation to planning graph based heuristics that helps uniform probability conditional planners both scale and...

Concept drift is a phenomenon typically experienced when data distributions change continuously over a period of time. In this paper we propose a cost-sensitive boosting approach for learning under concept drift. The proposed methodology estimates relevance costs of ‘old’ data samples w.r.t. to ‘newer’ samples and integrates it into the boosting process. We experiment this methodology on usenet...

The evaluation of classifier performance in a cost-sensitive setting is straightforward if the operating conditions (misclassification costs and class distributions) are fixed and known. When this is not the case, evaluation requires a method of visualizing classifier performance across the full range of possible operating conditions. This talk outlines the most important requirements for cost-...