Study | Pers | AH | Data | CA | LLA | HLA | CI |
|---|
[4] | ✘ | ✘ | SA | CAR | ✘ | 8 | ✘ |
[53] | ✘ | ✘ | SA | CAR | ✘ | 4 | ✘ |
[28] | ✘ | ✘ | SA | CAR | ✘ | 4 | ✘ |
[29] | ✘ | ✘ | SA | Clustering, ARM | ✘ | 8 | Time |
[30] | ✘ | ✘ | SA | ARM | ✘ | 13 | Time |
[43] | ✔ | ✘ | IMU | EM, PL | ✘ | 9 | Time |
[54] | ✘ | ✘ | IMU, SA | MLR, SI | 4 | 6 | Location |
[16] | ✔ | ✘ | IMU, SA | OSVM, PCA | 5 | ✘ | ✘ |
[45] | ✔ | ✘ | IMU | HMM | 2 | 12 | ✘ |
[1] | ✔ | ✘ | IMU | HBN, SVM | 7 | ✘ | ✘ |
[50] | ✔ | ✘ | IMU | RF | 5 | ✘ | ✘ |
[10] | ✘ | ✘ | IMU | KNN, LR, RF, SVM | ✘ | 8 | Location |
[36] | ✘ | ✘ | IMU | XGBoost | 8 | ✘ | Location |
[9] | ✘ | ✘ | IMU | HCA, ELM | 6 | 10 | Location |
[38] | ✘ | ✘ | IMU | CNN | ✘ | 18 | Time |
[40] | ✘ | ✘ | SA | RF | ✘ | 10 | Time |
Proposed | ✔ | ✔ | IMU, SA | CAR, AL, ECOC, RNN | 7 | 8–13 | Multiple contexts |
- Pers, Personalization; AH, Activity Handling; CA, Classification Approach; LLA, Low-Level Actions; HLA, High-Level Activities; CI, Context Information; SA, Sensor activations; IMU, Inertial measurement unit; CAR; Class Association Rules; ARM, Association rule mining; SVM, Support Vector Machines; PCA, Principal Component Analysis; EM, Expectation Minimization; PL, Probabilistic Learning; MLR, Multiple Logistic Regression; SI, Statistical Inferencing; OSVM, Online SVM; HMM, Hidden Markov Models; CNN, Convolutional Neural Networks; HBN, Hybrid Bayesian Networks, KNN, K-Nearest Neighbor; LR, Logistic Regression; RF, Random Forest; XGBoost, Extreme Gradient Boosting; HCA, Hierarchical Classification Approach; ELM, Extreme Learning Machines; CAR, Class association rules; AL, Associative Learning; ECOC, Error-Correcting Output Codes; RNN, Recurrent Neural Networks
