Figure 4
The illustration displays a grid of four boxplots, each showing the distribution of different performance metrics across a set of machine learning models. In each plot, the horizontal axis represents the machine learning models, which are labeled from left to right as “L R,” “S V M,” “D T,” “G B,” “R F,” “S T,” “C T 1 V,” and “C T 2 V.” These models are evaluated across four different performance metrics: “Accuracy,“ “F 1 score,” “True Positive Rate,” and “True Negative Rate.” The vertical axis represents the value of the corresponding metric for each model in that plot. In the top-left boxplot, the vertical axis is labeled “Accuracy,” which ranges from 0.8386 to 0.9186 with an interval of 0.02. Each model is represented by a vertical box, with a horizontal line inside the box indicating the median accuracy for that model. The whiskers extend from the top and bottom boxes, representing the range of accuracy values for each model. Outliers, represented by diamond-shaped markers, are located outside the whiskers. The median of the boxes ranges between 0.856 and 0.908. The outliers are present above and below the whiskers for the model S V M, an outlier is present below the bottom whisker for the model R F, and an outlier is present below the bottom whisker for the model C T 1 V. In the top-right boxplot, the vertical axis is labeled “F 1,” which ranges from 0.8359 to 0.9159 with an interval of 0.02. Each model is represented by a vertical box, with a horizontal line inside the box indicating the median accuracy for that model. The whiskers extend from the top and bottom boxes, representing the range of accuracy values for each model. Outliers, represented by diamond-shaped markers, are located outside the whiskers. The median of the boxes ranges between 0.85 and 0.904. Two outliers are present below the bottom whisker for the model L R, an outlier is present above the top whisker for the model S V M, and an outlier is present below the bottom whisker for the model R F. In the bottom-left boxplot, the vertical axis is labeled “True Positive Rate,” which ranges from 0.8496 to 0.9496 with an interval of 0.02. Each model is represented by a vertical box, with a horizontal line inside the box indicating the median accuracy for that model. The whiskers extend from the top and bottom boxes, representing the range of accuracy values for each model. Outliers, represented by diamond-shaped markers, are located outside the whiskers. The median of the boxes ranges between 0.882 and 0.937. An outlier is present above the top whisker for the model D T. In the bottom-right boxplot, the vertical axis is labeled “True Negative Rate,” which ranges from 0.7879 to 0.9279 with an interval of 0.02. Each model is represented by a vertical box, with a horizontal line inside the box indicating the median accuracy for that model. The whiskers extend from the top and bottom boxes, representing the range of accuracy values for each model. Outliers, represented by diamond-shaped markers, are located outside the whiskers. The median of the boxes ranges between 0.831 and 0.909. An outlier is present below the bottom whisker for the model D T, and the outliers are present above and below the whiskers for the model C T 2 V. Note: All numerical data values are approximated.Accuracy, F1, True Positive Rate and True Negative Rate scores. Source: Authors’ own creation