Figure 15
Twenty confusion matrices show the offices models’ performance for balanced and imbalanced training data.The grid of 20 confusion matrices is arranged in five rows (labeled a through e, representing “Random States”) and four columns comparing two machine learning models, “Random Forest (R F)” and “Neural Network (N N),” under two data conditions, “C W B” and “Imbalanced.” Each matrix has three rows and three columns, with rows representing “True” values 0, 1, and 2 (top to bottom) and columns representing “Predicted” values 0, 1, and 2 (left to right). A color bar on the right indicates performance ranging from 0 percent (dark purple) to 100 percent (bright yellow), increasing in 25-percent increments through transitions from dark purple to green to yellow. Random state (a): Random Forest: C W B: R F: The cells are highlighted as follows: Row 1: Column 1: 95 to 10. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 80 to 90. Column 3: 0 to 25. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 75 to 80. Random Forest: Imbalanced: R F: The cells are highlighted as follows: Row 1: Column 1: 50 to 75. Column 2: 50 to 75. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 50 to 75. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 95 to 100. Neural Network: C W B: N N: The cells are highlighted as follows: Row 1: Column 1: 75 to 80. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 75 to 80. Column 3: 0 to 25. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 75 to 80. Neural Network: Imbalanced: N N: The cells are highlighted as follows: Row 1: Column 1: 0 to 25. Column 2: 75 to 80. Column 3: 25 to 50. Row 2: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 95 to 100. Random state (b): Random Forest: C W B: R F: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 50 to 75. Column 3: 25 to 50. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 50 to 75. Random Forest: Imbalanced: R F: The cells are highlighted as follows: Row 1: Column 1: 50 to 75. Column 2: 25 to 50. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 90 to 100. Neural Network: C W B: N N: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 0 to 25. Row 3: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Neural Network: Imbalanced: N N: The cells are highlighted as follows: Row 1: Column 1: 0 to 25. Column 2: 95 to 100. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 50 to 75. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 50 to 75. Random state (c): Random Forest: C W B: R F: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 25 to 50. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 50 to 75. Random Forest: Imbalanced: R F: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 95 to 100. Neural Network: C W B: N N: The cells are highlighted as follows: Row 1: Column 1: 50 to 75. Column 2: 25 to 50. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 0 to 25. Row 3: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Neural Network: Imbalanced: N N: The cells are highlighted as follows: Row 1: Column 1: 0 to 25. Column 2: 95 to 100. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 50 to 75. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 95 to 100. Random state (d): Random Forest: C W B: R F: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 25 to 50. Row 3: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Random Forest: Imbalanced: R F: The cells are highlighted as follows: Row 1: Column 1: 50 to 75. Column 2: 0 to 25. Column 3: 25 to 50. Row 2: Column 1: 0 to 25. Column 2: 50 to 75. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 80 to 90. Neural Network: C W B: N N: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 0 to 25. Row 3: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Neural Network: Imbalanced: N N: The cells are highlighted as follows: Row 1: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Row 2: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 95 to 100. Random state (e): Random Forest: C W B: R F: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 25 to 50. Row 3: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Random Forest: Imbalanced: R F: The cells are highlighted as follows: Row 1: Column 1: 95 to 100. Column 2: 0 to 25. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 50 to 75. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Neural Network: C W B: N N: The cells are highlighted as follows: Row 1: Column 1: 50 to 75. Column 2: 25 to 50. Column 3: 0 to 25. Row 2: Column 1: 25 to 50. Column 2: 50 to 75. Column 3: 0 to 25. Row 3: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Neural Network: Imbalanced: N N: The cells are highlighted as follows: Row 1: Column 1: 0 to 25. Column 2: 95 to 100. Column 3: 0 to 25. Row 2: Column 1: 0 to 25. Column 2: 25 to 50. Column 3: 50 to 75. Row 3: Column 1: 0 to 25. Column 2: 0 to 25. Column 3: 95 to 100. All numerical values in the cells are approximated.

RF and NN Confusion Matrices showing the offices models' performance for balanced and imbalanced training data across 5 random states. Note: 0: Dissatisfied, 1: Neutral, 2: Satisfied. Source: Authors' own work

or Create an Account

Close subscription notice
Close access options