Impacts of Unbalanced Test Data on the Evaluation of Classification Methods
Manh Hung Nguyen · International Journal of Advanced Computer Science and Applications · 2019
The performance of a classifier in a supervised machine learning problem is popularly evaluated by using the accuracy, precision, recall, and F1-score. These parameters could evaluate very well classifiers in the case that the number of positive label sample and the number of negative label sample in the testing set are balanced or nearly balanced. However, these parameters may miss-evaluate the classifiers in some case where the positive and negative samples in the testing set is unbalanced. This paper proposes some update in these parameters by taking into account the unbalanced factor which represents the unbal-ance ratio of positive and negative samples in the testing set. The new updated parameters are then experimentally evaluated to compare to the traditional parameters.