A Comparison of MCC and CEN Error Measures in Multi-Class Prediction
Open Access
- 8 August 2012
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 7 (8), e41882
- https://doi.org/10.1371/journal.pone.0041882
Abstract
We show that the Confusion Entropy, a measure of performance in multiclass problems has a strong (monotone) relation with the multiclass generalization of a classical metric, the Matthews Correlation Coefficient. Analytical results are provided for the limit cases of general no-information (n-face dice rolling) of the binary classification. Computational evidence supports the claim in the general case.Keywords
This publication has 18 references indexed in Scilit:
- The MicroArray Quality Control (MAQC)-II study of common practices for the development and validation of microarray-based predictive modelsNature Biotechnology, 2010
- A novel measure for evaluating classifiersExpert Systems with Applications, 2010
- Evaluating diagnostic tests: The area under the ROC curve and the balance of errorsStatistics in Medicine, 2010
- A fast and efficient gene-network reconstruction method from multiple over-expression experimentsBMC Bioinformatics, 2009
- An experimental comparison of performance measures for classificationPattern Recognition Letters, 2009
- Visualization and analysis of classifiers performance in multi-class medical dataExpert Systems with Applications, 2008
- Multi-class ROC analysis from a multi-objective optimisation perspectivePattern Recognition Letters, 2006
- Comparing two K-category assignments by a K-category correlation coefficientComputational Biology and Chemistry, 2004
- Repeated observation of breast tumor subtypes in independent gene expression data setsProceedings of the National Academy of Sciences of the United States of America, 2003
- The use of the area under the ROC curve in the evaluation of machine learning algorithmsPattern Recognition, 1997