Data Integration Based Human Activity Recognition using Deep Learning Models

Regular monitoring of physical activities such as walking, jogging, sitting, and standing will help reduce the risk many diseases like cardiovascular complications, obesity, diabetes. Recently, much research showed that effective development Human Activity Recognition (HAR) in people aid human healthcare. In this concern, deep learning models with a novel automated hyperparameter generator are proposed implemented to predict walking upstairs, downstairs, more precisely robustly. Conventional HAR systems unable manage real-time changes surrounding infrastructure. Improved approaches overcome constraint by integrating multiple sensing modalities. These sensors can produce accurate information, leading better perception activity recognition. The approach uses sensor-level fusion integrate gyroscope accelerometer sensors. analysis is carried out using widely accepted benchmark UCI-HAR dataset. Based on several performance evaluation experiments, classification accuracy long short-term memory (LSTM), convolutional neural network (CNN), (DNN) classifiers reported be 96%, 92%, 93%, respectively. Compared state-of-the-art models, method gives results.