ISSN :2582-9793

Extending F1 metric, probabilistic approach

Original Research (Published On: 10-May-2023 )
Extending F1 metric, probabilistic approach
DOI : 10.54364/AAIML.2023.1161

Mikolaj Sitarz

Adv. Artif. Intell. Mach. Learn., 3 (2):1025-1038

Mikolaj Sitarz : refaba.com

Download PDF Here

DOI: 10.54364/AAIML.2023.1161

Article History: Received on: 19-Apr-23, Accepted on: 29-Apr-23, Published on: 10-May-23

Corresponding Author: Mikolaj Sitarz

Email: sitarz@xmmx.eu

Citation: Mikolaj Sitarz (2023). Extending F1 metric, probabilistic approach. Adv. Artif. Intell. Mach. Learn., 3 (2 ):1025-1038

          

Abstract

    

This article explores the extension of well-known F1 score used for assessing the performance of binary classifiers. We propose the new metric using probabilistic interpretation of precision, recall, specificity, and negative predictive value. We describe its properties and compare it to common metrics. Then we demonstrate its behavior in edge cases of the confusion matrix. Finally, the properties of the metric are tested on binary classifier trained on the real dataset.


Statistics

   Article View: 675
   PDF Downloaded: 23