An Extra Tree Ensemble Optimization-Based Deep Learning Framework for Human Activity Recognition

Human activity recognition (HAR) is a burgeoning field of study due to its real-life applications in the medical field, the e-health system, and elder care or care of physically impaired people in a smart healthcare environment. Using sensors built into wearable devices, such as smartphones, HAR provides an opportunity to identify human behavior and better understand an individual’s health. In the past years, conventional machine learning techniques have made significant progress in HAR. However, these methods significantly rely on conventional feature extraction, which may hinder the effectiveness of the generalization model. The other challenges these methods face for HAR are performance degradation of many models with the increase in the number of activities, lack of capability of models to capture different layers of activities, high dimensionality, overfitting, and optimization problems. These challenges attracted numerous people to explore this field. The development in deep learning techniques has addressed most of these problems by automatically extracting discriminative features from raw input sequences obtained by multimodal sensing devices to acknowledge human activities accurately. This paper proposes an extra tree ensemble optimization-based deep learning framework (DELETO) for HAR. The model shows better results than recent and conventional techniques, with an accuracy of 98.7% and a computation time of 2.533 seconds.