Human Activity Recognition Using Feedforward Backpropagation Neural Network
Fadi Al Debs, Makram El Jurdy, Elias Fares, Gaby Abou Haidar, Michel J. Owayjan, Roger Achkar · 2024
This paper presents a feedforward backpropagation neural network designed to recognize human activities using loT sensor data. Human activity recognition based on mobile phones and wearable sensor data has garnered significant attention due to its wide range of applications in healthcare., smart environments., and beyond. Previous studies in this field have utilized various sensors., such as accelerometers., gyroscopes., and orientation sensors., among others., to categorize human activities using modern wearable technologies. The proposed system employs a Human Activity Recognition framework., which collects real-time sensor data and analyzes human movements using deep learning techniques. The model achieves an accuracy of approximately 96% in identifying human motions based on the sensor data.