Table 1.
Confusion matrix of HAR based on handcrafted feature extraction.
| Target Class | ||||||||
|---|---|---|---|---|---|---|---|---|
| Walking | Upstairs | Downstairs | Sitting | Standing | Lying | Precision | ||
|
Predicted
Class |
Walking | 492 16.69% |
22 0.75% |
4 0.14% |
0 0.00% |
0 0.00% |
0 0.00% |
94.98% 5.02% |
| Upstairs | 1 0.03% |
448 15.20% |
17 0.58% |
0 0.00% |
0 0.00% |
0 0.00% |
96.14% 3.86% |
|
| Downstairs | 3 0.10% |
1 0.03% |
399 13.54% |
0 0.00% |
0 0.00% |
0 0.00% |
99.01% 0.99% |
|
| Sitting | 0 0.00% |
0 0.00% |
0 0.00% |
464 15.74% |
17 0.58% |
0 0.00% |
96.47 3.53% |
|
| Standing | 0 0.00% |
0 0.00% |
0 0.00% |
27 0.92% |
515 17.48% |
0 0.00% |
95.02% 4.98% |
|
| Lying | 0 0.00% |
0 0.00% |
0 0.00% |
0 0.00% |
0 0.00% |
537 18.22% |
100.00% 0.00% |
|
| Recall | 99.19% 0.81% |
95.12% 4.88% |
95.00% 5.00% |
94.50% 5.50% |
96.80% 3.20% |
100.00% 0.00% |
96.88% 3.12% |
|