Receiver Operating Characteristic
Operating characteristic; Relative operating characteristic; ROC
Receiver operating characteristic (ROC) analysis is a graphical approach for analyzing the performance of a classifier. It uses a pair of statistics – true positive rate and false positive rate – to characterize a classifier’s performance. The statistics are plotted on a two-dimensional graph, with false positive rate on the x-axis and true positive rate on the y-axis. The resulting plot can be used to compare the relative performance of different classifiers and to determine whether a classifier performs better than random guessing.
ROC analysis was originally developed in signal detection theory to deal with the problem of discriminating known signals from a random noise background . It was first applied to the radar detection problem to quantify how effective targets such as enemy aircrafts can be identified according to their radar signatures. In the 1960s, ROC analysis...
- 1.Cortes C, Mohri M. Auc optimization vs. error rate minimization. In: Advances in Neural Information Proceedings of the Systems 16, Proceedings of the Neural Information Proceedings of the Systems; 2003.Google Scholar
- 2.Davis J, Goadrich M. The relationship between precision-recall and roc curves. In: Proceedings of the 23rd International Conference on Machine Learning; 2006.Google Scholar
- 3.Drummond C, Holte RC. Explicitly representing expected cost: an alternative to roc representation. In: Proceedings of the 6th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2000. p. 198–207.Google Scholar
- 5.Ferri C, Flach PA, Hernandez-Orallo J. Learning decision trees using the area under the roc curve. In: Proceedings of the 19th International Conference on Machine Learning; 2002.Google Scholar
- 6.Green DM, Swets JA, editors. Swets signal detection theory and psychophysics. New York: Wiley; 1966.Google Scholar
- 8.Joachims T. A support vector method for multivariate performance measures. In: Proceedings of the 22nd International Conference on Machine Learning; 2005.Google Scholar
- 11.Peterson WW, Birdsall TG, Fox WC. The theory of signal detectability. IRE Trans. 1954;PGIT-4(4): 171.Google Scholar
- 12.Provost FJ, Fawcett T. Analysis and visualization of classifier performance: comparison under imprecise class and cost distributions. In: Proceedings of the 3rd International Conference on Knowledge Discovery and Data Mining; 1997. p. 43–8.Google Scholar
- 13.Sackett DL. Clinical diagnosis and the clinical laboratory. Clin Invest Med. 1978;1:37.Google Scholar