DOI QR코드

DOI QR Code

ROC and Cost Graphs for General Cost Matrix Where Correct Classifications Incur Non-zero Costs

  • 발행 : 2004.04.01

초록

Often the accuracy is not adequate as a performance measure of classifiers when costs are different for different prediction errors. ROC and cost graphs can be used in such case to compare and identify cost-sensitive classifiers. We extend ROC and cost graphs so that they can be used when more general cost matrix is given, where not only misclassifications but correct classifications also incur penalties.

키워드

참고문헌

  1. UCI Repository of machine iearning databases Blake,C.L.;Merz,C.J.
  2. Classification and Regression Trees Breiman,L.;Friedman,J.H.;Olshen,J.A.;Stone,C.J.
  3. Technical Report, School of Information Technology and Engineering, University of Ottawa Explicitly representing expected cost: An alternative to ROC representation Drummond,C.;Holte,R.
  4. Series in Cognition and Perception Signal Detection Theory and ROC Analysis Egan,J.P.
  5. Proceedings of the Seventeenth International Joint Conference on Artificial Intelligence The fouondations of cost-sensitive learning Elkan,C.
  6. Proceedings of the Third International Conference on Knowledge Discovery and Data Mining Analysis and visualization of classifier performance:Comparison under imprecise class and cost distributions Provost,F.;Fawcett,T.
  7. Technical Report, Mayo Foundation An introduction to recursive partitioning using the RPART routines Therneau,T.M.;Atkinson,E.J.

피인용 문헌

  1. Prediction of Product Life Cycle Using Data Mining Algorithms : A Case Study of Clothing Industry vol.40, pp.3, 2014, https://doi.org/10.7232/JKIIE.2014.40.3.291
  2. Alternative Optimal Threshold Criteria: MFR vol.27, pp.5, 2014, https://doi.org/10.5351/KJAS.2014.27.5.773
  3. Cost Ratios for Cost and ROC Curves vol.17, pp.6, 2010, https://doi.org/10.5351/CKSS.2010.17.6.755