Table 6.
Deep Learning Techniques results.
| Dataset | Technique | Metrics | References | |||
|---|---|---|---|---|---|---|
| Accuracy | Precision | Recall | F-Measure | |||
| Uci Har | CNN | 92.71 | 93.21 | 92.82 | 92.93 | [154] |
| LSTM | 89.01 | 89.14 | 88.99 | 88.99 | ||
| BLSTM | 89.4 | 89.41 | 89.36 | 89.35 | ||
| MLP | 86.83 | 86.83 | 86.58 | 86.61 | ||
| SVM | 89.85 | 90.5 | 89.86 | 89.85 | ||
| PAMAP2 | CNN | 91.00 | 91.66 | 90.86 | 91.16 | |
| LSTM | 85.86 | 86.51 | 84.67 | 85.34 | ||
| BLSTM | 89.52 | 90.19 | 89.02 | 89.4 | ||
| MLP | 82.07 | 83.35 | 82.17 | 82.46 | ||
| SVM | 84.07 | 84.71 | 84.23 | 83.76 | ||
| Propio Infrared Images | LBP-Naive Bayes | 42.1 | - | - | - | [155] |
| HOG-Naive Bayes | 77.01 | - | - | - | ||
| LBP-KNN | 53.261 | - | - | - | ||
| HOG-KNN | 83.541 | - | - | - | ||
| LBP-SVM | 62.34 | - | - | - | ||
| HOF-SVM | 85.92 | - | - | - | ||
| Uci Har | DeepConvLSTM | 94.77 | - | - | - | [156] |
| CNN | 92.76 | - | - | - | ||
| Weakly Dataset | DeepConvLSTM | 92.31 | - | - | - | |
| CNN | 85.17 | - | - | - | ||
| Opportunity | HC | 85.69 | - | - | - | [157] |
| CBH | 84.66 | - | - | - | ||
| CBS | 85.39 | - | - | - | ||
| AE | 83.39 | - | - | - | ||
| MLP | 86.65 | - | - | - | ||
| CNN | 87.62 | - | - | - | ||
| LSTM | 86.21 | - | - | - | ||
| Hybrid | 87.67 | - | - | - | ||
| ResNet | 87.67 | - | - | - | ||
| ARN | 90.29 | - | - | - | ||
| UniMiB-SAHR | HC | 21.96 | - | - | - | |
| CBH | 64.36 | - | - | - | ||
| CBS | 67.36 | - | - | - | ||
| AE | 68.39 | - | - | - | ||
| MLP | 74.82 | - | - | - | ||
| CNN | 73.36 | - | - | - | ||
| LSTM | 68.81 | - | - | - | ||
| Hybrid | 72.26 | - | - | - | ||
| ResNet | 75.26 | - | - | - | ||
| ARN | 76.39 | - | - | - | ||
| Uci Har | KNN | 90.74 | 91.15 | 90.28 | 90.48 | [158] |
| SVM | 96.27 | 96.43 | 96.14 | 96.23 | ||
| HMM+SVM | 96.57 | 96.74 | 06.49 | 96.56 | ||
| SVM+KNN | 96.71 | 96.75 | 96.69 | 96.71 | ||
| Naive Bayes | 77.03 | 79.25 | 76.91 | 76.72 | ||
| Logistic Regression | 95.93 | 96.13 | 95.84 | 95.92 | ||
| Decision Tree | 87.34 | 87.39 | 86.95 | 86.99 | ||
| Random Forest | 92.30 | 92.4 | 92.03 | 92.14 | ||
| MLP | 95.25 | 95.49 | 95.13 | 95.25 | ||
| DNN | 96.81 | 96.95 | 96.77 | 96.83 | ||
| LSTM | 91.08 | 91.38 | 91.24 | 91.13 | ||
| CNN+LSTM | 93.08 | 93.17 | 93.10 | 93.07 | ||
| CNN+BiLSTM | 95.42 | 95.58 | 95.26 | 95.36 | ||
| Inception+ResNet | 95.76 | 96.06 | 95.63 | 95.75 | ||
| Utwente Dataset | Naive Bayes | - | - | - | 94.7 | [159] |
| SVM | - | - | - | 91.6 | ||
| Deep Stacked Autoencoder | - | - | - | 97.6 | ||
| CNN-BiGRu | - | - | - | 97.8 | ||
| PAMAP2 | DeepCOnvTCN | - | - | - | 81.8 | |
| InceptionTime | - | - | - | 81.1 | ||
| CNN-BiGRu | - | - | - | 85.5 | ||
| FrailSafe dataset | CNN | 91.84 | - | - | - | [160] |
| CASAS Milan | LSTM | 76.65 | - | - | - | [135] |
| Bi-LSTM | 77.44 | - | - | - | ||
| Casc-LSTM | 61.01 | - | - | - | ||
| ENs2-LSTM | 93.42 | - | - | - | ||
| CASAS Cairo | LSTM | 82.79 | - | - | - | |
| Bi-LSTM | 82.41 | - | - | - | ||
| Casc-LSTM | 68.07 | - | - | - | ||
| ENs2-LSTM | 83.75 | - | - | - | ||
| CASAS Kyoto 2 | LSTM | 63.98 | - | - | - | |
| Bi-LSTM | 65.79 | - | - | - | ||
| Casc-LSTM | 66.20 | - | - | - | ||
| ENs2-LSTM | 69.76 | - | - | - | ||
| CASAS Kyoto 3 | LSTM | 77.5 | - | - | - | |
| Bi-LSTM | 81.67 | - | - | - | ||
| Casc-LSTM | 87.33 | - | - | - | ||
| ENs2-LSTM | 88.71 | - | - | - | ||
| Proposal | ANN | 89.06 | - | - | - | [160] |
| SVM | 94.12 | - | - | - | ||
| DBN | 95.85 | - | - | - | ||