Figure 9
The top row displays two confusion matrices. The matrix on the left, labeled “(a) Confusion matrix for predicting A w e B D model,” is predominantly blue. Both matrices have “Predicted label” on the x-axis and “True label” on the y-axis, with values 0 and 1. The matrix indicates that the model correctly predicted 86 true negatives (top-left) and 17 true positives (bottom-right). It incorrectly predicted 19 false positives (top-right) and 15 false negatives (bottom-left). The matrix on the right, labeled “(b) Confusion matrix for predicting I n t B D Model,” is predominantly orange. It shows the number of correct and incorrect predictions for the “I n t B D” model, with 27 true negatives, 78 true positives, 15 false positives, and 17 false negatives. The bottom row displays two performance evaluation curves. The plot on the left, labeled “(c) Performance evaluation curves for predicting A w e B D model,” shows an ROC Curve (blue solid line) with an A U C of 0.73 and a Precision-Recall Curve (orange dashed line) with an A P of 0.46. A dotted black diagonal line represents the “Random Chance (R O C).” The plot on the right, labeled “(d) Performance evaluation curves for predicting I n t B D Model,” shows an R O C Curve with an A U C of 0.77 and a Precision-Recall Curve with an A P of 0.85. A dotted black diagonal line also represents the “Random Chance (R O C).” Both plots have “False Positive Rate or Recall” on the horizontal axis and “True Positive Rate or Precision” on the vertical axis, with a scale from 0.0 to 1.0 in increments of 0.2. The blue curve in both graphs exhibits a concave-down increasing trend, while the orange curve is overall decreasing. The black line extends from the bottom left to the top right corners in both graphs.RF model performance confusion matrices and evaluation curves for predicting awareness and adoption of biophilic development. Source: Authors’ own creation