Table 6 The most common metric used in studies

Accuracy

Accuracy is a metric that commonly describes how the developed model performs throughout all datasets.

7 (43.75%)

Specificity

Specificity is the extent of true negatives that are accurately anticipated by the developed model.

4 (25%)

Sensitivity

Sensitivity could be a degree of how well machine learning demonstrate can distinguish positive instances.

4 (25%)

F-measure

The F1-score or F-score may be a degree of a model’s precision on a dataset that can be utilized in classification models.

1 (6.25%)

Root Mean Square Error (RMSE)

Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) are measurements utilized to assess a Regression Model.

4 (25%)

R-squared

The R2 score could be a very imperative metric that’s utilized to assess the performance of a regression-based machine learning model. It is known as R squared and is additionally known as the coefficient of assurance.

3 (18.75%)

AUC or ROC curve

ROC curve, moreover known as Receiver Operating Characteristics Curve, could be a metric utilized to degree the execution of a classifier model. The ROC curve represents the rate of true positives about the rate of false positives in the classifier model.

5 (31.25%)

Chi-square or correlation matrix

A chi-square test is utilized to test the independence of two occasions.

3 (18.75%)

Confusion matrix

The confusion matrix is a matrix utilized to show the exact performance of the classification models based on a given set of test data.

4 (25%)