Table 9.
Transfer Learning Techniques results.
| Dataset | Technique | Metrics | References | |||
|---|---|---|---|---|---|---|
| Accuracy | Precision | Recall | F-Measure | |||
| CSI | KNN | 98.3 | - | - | - | [179] |
| SVM | 98.3 | - | - | - | ||
| CNN | 99.2 | - | - | - | ||
| Opportunity | KNN+PCA | 60 | - | - | - | [180] |
| GFK | 59 | - | - | - | ||
| STL | 65 | - | - | - | ||
| SA-GAN | 73 | - | - | - | ||
| USC-HAD | MMD | 80 | - | - | - | [181] |
| DANN | 77 | - | - | - | ||
| WD | 72 | - | - | - | ||
| Proposal | KNN-OS | 79.84 | 85.84 | 91.88 | 88.61 | [182] |
| KNN-SS | 89.64 | 94.41 | 94.76 | 94.52 | ||
| SVM-OS | 77.14 | 97.04 | 79.23 | 87.09 | ||
| SVM-SS | 87.5 | 94.39 | 92.61 | 93.27 | ||
| DT-OS | 87.5 | 94.61 | 92.16 | 93.14 | ||
| DT-SS | 91.79 | 95.19 | 96.26 | 95.71 | ||
| JDA | 86.79 | 92.71 | 93.07 | 92.89 | ||
| BDA | 91.43 | 95.9 | 95.18 | 95.51 | ||
| IPL-JPDA | 93.21 | 97.04 | 95.97 | 96.48 | ||
| KNN-OS | 79.84 | 85.84 | 91.88 | 88.61 | ||
| Wiezmann Dataset | VGG-16 MODEL | 96.95 | 97.00 | 97.00 | 97.00 | [183] |
| VGG-19 MODEL | 96.54 | 97.00 | 97.00 | 96.00 | ||
| Inception-v3 Model | 95.63 | 96.00 | 96.00 | 96.00 | ||
| PAMAP2 | DeepConvLSTM | - | - | - | 93.2 | [184] |
| Skoda Mini Checkpoint | - | - | - | 93 | ||
| Opportunity | PCA | 66.78 | - | - | - | [185] |
| TCA | 68.43 | - | - | - | ||
| GFK | 70.87 | - | - | - | ||
| TKL | 70.21 | - | - | - | ||
| STL | 73.22 | - | - | - | ||
| TNNAR | 78.4 | - | - | - | ||
| PAMAP2 | PCA | 42.87 | - | - | - | |
| TCA | 47.21 | - | - | - | ||
| GFK | 48.09 | - | - | - | ||
| TKL | 43.32 | - | - | - | ||
| STL | 51.22 | - | - | - | ||
| TNNAR | 55.48 | - | - | - | ||
| UCI DSADS | PCA | 71.24 | - | - | - | |
| TCA | 73.47 | - | - | - | ||
| GFK | 81.23 | - | - | - | ||
| TKL | 74.26 | - | - | - | ||
| STL | 83.76 | - | - | - | ||
| TNNAR | 87.41 | - | - | - | ||
| UCI HAR | CNN-LSTM | 90.8 | - | - | - | [186] |
| DT | 76.73 | . | - | - | [187] | |
| RF | 71.96 | - | - | - | ||
| TB | 75.65 | - | - | - | ||
| TransAct | 86.49 | - | - | - | ||
| Mhealth | DT | 48.02 | - | - | - | |
| RF | 62.25 | - | - | - | ||
| TB | 66.48 | - | - | - | ||
| TransAct | 77.43 | - | - | - | ||
| Daily Sport | DT | 66.67 | . | . | . | |
| RF | 70.38 | . | . | . | ||
| TB | 72.86 | . | - | - | ||
| TransAct | 80.83 | - | - | - | ||
| Proposal | Without SVD (Singular Value Decomposition) | 63.13% | - | - | - | [188] |
| With SVD (Singular Value Decomposition) | 43.13% | - | - | - | ||
| Transfer Accuracy | 97.5% | - | - | - | ||
| PAMAP2 | CNN | 84.89 | - | - | - | [189] |
| UCI HAR | 83.16 | - | - | - | ||
| UCI HAR | kNN | 77.28 | - | - | - | [190] |
| DT | 72.16 | - | - | - | ||
| DA | 77.46 | - | - | - | ||
| NB | 69.93 | - | - | - | ||
| Transfer Accuracy | 83.7 | - | - | - | ||
| UCF Sports Action dataset | VGGNet-19 | 97.13 | - | - | - | [191] |
| AMASS | DeepConvLSTM | 87.46 | - | - | - | [192] |
| DIP | 89.08 | - | - | - | ||
| DAR Dataset | Base CNN | 85.38 | - | - | - | [193] |
| AugToAc | 91.38 | - | - | - | ||
| HDCNN | 86.85 | - | - | - | ||
| DDC | 86.67 | - | - | - | ||
| UCI HAR | CNN_LSTM | 92.13 | - | - | - | [194] |
| CNN_LSTM_SENSE | 91.55 | - | - | - | ||
| LSTM | 91.28 | - | - | - | ||
| LSTM_DENSE | 91.40 | - | - | - | ||
| ISPL | CNN_LSTM | 99.06 | - | - | - | |
| CNN_LSTM_SENSE | 98.43 | - | - | - | ||
| LSTM | 96.23 | - | - | - | ||
| LSTM_DENSE | 98.11 | - | - | - | ||