Optimizing the Performance of KNN Classifier for Human Activity Recognition

AUTO-HAR: An adaptive human activity recognition framework using an automated CNN architecture design

“… The gradient-based algorithms require the update of a large number of parameters, which will significantly increase the GPU costs. [14] Human Activity Recognition Three optimization search algorithms are used: PSO, Greedy,and Genetic algorithms with KNN classifier. The model achieved excellent classification results based on smartphone embedded sensors to recognize six basic human activities.…”

“…In addition, practitioners of deep learning encounter difficulty when it comes to manually building deep models and determining suitable configurations (e.g., model layers and operation types) through trial and error. Various steps are involved in feeding domain knowledge into DL, including Feature Engineering (FE) [13] , model generation [14] , and model deployment [15] , [16] . Because CNNs are based on layers, they allow the flexibility of adding or removing layers based on the training phase, which is then used in inference (classification) to classify the data.…”

Communication-Efficient Privacy-Preserving Federated Learning via Knowledge Distillation for Human Activity Recognition Systems

“…Machine learning methods were applied to HAR systems [6] by training methods such as Support Vector Machine (SVM) [7], K-Nearest Neighbor (KNN) [8], [9], etc. on supervised datasets to classify activities.…”