2026-10-15T11:11:44Z https://data.mendeley.com/oai

oai:data.mendeley.com/2g36672s5f.4 2020-09-14T21:46:20Z

Canbek, Gürol Binary-Classification Performance-Metrics Benchmarking Data Mendeley Data This data provides the detailed test results of the benchmarking for the binary-classification performance metrics. The benchmark comprising three stages was applied on 13 metrics namely True Positive Rate, True Negative Rate, Positive Predictive Value, Negative Predictive Value, Accuracy, Informedness, Markedness, Balanced Accuracy, G, Normalized Mutual Information, F1, Cohen’s Kappa, and Mathew’s Correlation Coefficient (MCC). The new benchmarking method is described in Gürol Canbek, Tugba Taskaya Temizel, and Seref Sagiroglu, "BenchMetrics: A Systematic Benchmarking Method for Binary-Classification Performance Metrics", Neural Computing and Applications, 2020 (Submitted). The related benchmarking API and an interactive ready-to-run experimentation platform is available online (see the references) Machine Learning Machine Learning Theory Contingency Table Analysis Performance Measurement Accuracy Analysis Classification (Machine Learning) Classifier Evaluation Taskaya Temizel, Tugba Sagiroglu, Seref Dataset doi:10.17632/2g36672s5f.4 oai:data.mendeley.com/2g36672s5f.4 Creative Commons Attribution 4.0 International http://creativecommons.org/licenses/by/4.0 https://data.mendeley.com/datasets/2g36672s5f 2020-09-14T21:46:20Z