Research Article
Wearable Sensor-Based Human Activity Recognition Using Hybrid Deep Learning Techniques
Table 10
Average accuracy of various actions in CNN-LSTM, RF, and KNN models.
| ID | RF (%) | KNN (%) | CNN-LSTM (%) |
| A1 | 99.90 | 88.10 | 99.03 | A2 | 92.50 | 97.80 | 97.78 | A3 | 90.20 | 99.40 | 98.86 | A4 | 91.90 | 83.80 | 90.76 | A5 | 90.80 | 87.50 | 93.09 | A6 | 97.10 | 100 | 99.78 | A7 | 71.30 | 66.70 | 94.44 | A8 | 72.00 | 68.00 | 100 | A9 | 51.30 | 38.60 | 76.00 | A10 | 74.90 | 36.30 | 82.35 | A11 | 59.20 | 33.70 | 82.05 | A12 | 61.10 | 57.90 | 88.89 |
|
|