Then to verify if the accuracy of the Minimum Adequate Model is comparable to the other model we can run a confusion matrix to calculate the classification accuracy. Figure 8 below shows the confusion matrix for this model. We can use the same formula we used for figure 5 (diagonally 90+21=111, misclassified 25+12=37, total instances 111+37=148) and determine that the classification accuracy of the data is 111/148=0.75 or 75%.Figure 8Comparing the results, the standard model on the training data had a classification accuracy of 73% and the Minimum Adequate Model was 75%, only a 2% difference. This meansthe Minimal Adequate Model is sufficient and cleaner. Now that we have chosen the Minimal Adequate Model we can run a residuals plot, figure 9 below to double check that our model is correct. The residuals are plotted against the predictions and we can see the lowess smooth lines that approximates a line having zero slope and intercept. From this information, we can concludethat our model does not have any significant inadequacies. Therefore, we can use the Minimum Adequate Model