Table 2 Resubstitution error and confusion matrix for the random forest (RF) 50 model
(a) Resubstitution error matrix for 2014 | ||||
|---|---|---|---|---|
Land-use type | Agriculture land | Built-up land | Wasteland | Waterbody |
Agriculture land | 55 | 1 | 4 | 0 |
Built-up land | 0 | 78 | 0 | 0 |
Wasteland | 0 | 1 | 295 | 1 |
Waterbody | 1 | 0 | 1 | 62 |
Training Overall accuracy = 0.98 | ||||
(b) Confusion matrix for 2014 | ||||
|---|---|---|---|---|
Land-use type | Agriculture land | Built-up land | Wasteland | Waterbody |
Agriculture land | 11 | 0 | 13 | 3 |
Built-up land | 1 | 17 | 7 | 3 |
Wasteland | 3 | 3 | 109 | 5 |
Waterbody | 1 | 2 | 4 | 19 |
Testing (Validation) accuracy = 0.78 | ||||
(c) Resubstitution error matrix for 2020 | ||||
|---|---|---|---|---|
Land-use type | Agriculture land | Built-up land | Wasteland | Waterbody |
Agriculture land | 50 | 0 | 2 | 0 |
Built-up land | 0 | 76 | 2 | 0 |
Wasteland | 0 | 0 | 309 | 0 |
Waterbody | 0 | 0 | 3 | 60 |
Training Overall accuracy = 0.98 | ||||
(d) Confusion matrix for 2020 | ||||
|---|---|---|---|---|
Land-use type | Agriculture land | Built-up land | Wasteland | Waterbody |
Agriculture land | 16 | 4 | 14 | 1 |
Built-up land | 1 | 27 | 0 | 0 |
Wasteland | 2 | 1 | 101 | 4 |
Waterbody | 0 | 1 | 8 | 18 |
Testing (Validation) accuracy = 0.82 | ||||