Cost-Sensitive Test Strategies

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 (attribute cost), and misdiagnoses are like misclassifications which may also incur a cost (misclassification cost). We first propose an improved decision tree learning algorithm that minimizes the sum of attribute costs and misclassification costs. Then we design several novel "test strategies" that may request to obtain values of unknown attributes at cost (similar to doctors' ordering of medical tests at cost) in order to minimize the total cost for test examples (new patients). We empirically evaluate and compare these test strategies, and show that they are effective and that they outperform previous methods. A case study on heart disease is given.

[1]  Marlon Núñez.  The use of background knowledge in decision tree induction , 2004, Machine Learning.

[2]  Maytal Saar-Tsechansky,et al.  Economical active feature-value acquisition through Expected Utility estimation , 2005, UBDM '05.

[3]  Thomas G. Dietterich,et al.  Pruning Improves Heuristic Search for Cost-Sensitive Learning , 2002, ICML.

[4]  Qiang Yang,et al.  Test-cost sensitive naive Bayes classification , 2004, Fourth IEEE International Conference on Data Mining (ICDM'04).

[5]  Dan Roth,et al.  Learning cost-sensitive active classifiers , 2002, Artif. Intell..

[6]  Peter D. Turney.  Types of Cost in Inductive Concept Learning , 2002, ArXiv.

[7]  Ming Tan,et al.  Cost-Sensitive Learning of Classification Knowledge and Its Applications in Robotics , 1993, Machine Learning.

[8]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[9]  Peter D. Turney.  Cost-Sensitive Classification: Empirical Evaluation of a Hybrid Genetic Decision Tree Induction Algorithm , 1994, J. Artif. Intell. Res..

[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]  D. J. Newman,et al.  UCI Repository of Machine Learning Database , 1998 .

[13]  Foster J. Provost,et al.  Active feature-value acquisition for classifier induction , 2004, Fourth IEEE International Conference on Data Mining (ICDM'04).

[14]  G. Gorry,et al.  Experience with a model of sequential diagnosis. , 2011, Computers and biomedical research, an international journal.

[15]  Usama M. Fayyad,et al.  Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning , 1993, IJCAI.

[16]  金田 重郎,et al.  C4.5: Programs for Machine Learning (書評) , 1995 .

[17]  Ming Tan,et al.  Cost-sensitive learning of classification knowledge and its applications in robotics , 2004, Machine Learning.