A hybrid approach using Bidirectional Neural Networks for Human Activity Recognition
S. Sowmiya, D. Menaka · 2022 Third International Conference on Intelligent Computing Instrumentation and Control Technologies (ICICICT) · 2022
Human Activity Recognition has a vital role for its immense applications in human life. With the advances in sensor technologies and utilizing the IoT, HAR has a wide area of research with the help of deep learning algorithms. The advanced deep learning paradigm provides end-to-end learning models from unstructured, complex data. IoT wearables and smartphones are now widely used embedded with mobile apps for telemedicine, e-health monitoring, AAL, sports monitoring, smart homes etc. This paper presents a novel approach using hybrid neural networks implemented with Bidirectional GRU, Bidirectional LSTM and CNN. The algorithm was tested using two activity recognition datasets WISDM and MHEALTH. It provides improved accuracy over the other activity recognition techniques.