Investigation of Deep Learning-Based Models for Identifying Human Behavior in Internet of Things Applications
Tarunima Chatterjee, Pinaki Pratim Acharjya · 2025
With the rapid development of the Internet of Things (IoT), intelligent healthcare applications and systems are increasingly integrated with wearable sensors and mobile devices. These sensors serve not only to gather data but also, more crucially, to aid users in monitoring and managing their daily activities. Various approaches to human activity recognition (HAR) enhance this tracking capability. However, many existing HAR techniques rely on exploratory case-based shallow feature learning architectures, which struggle with accurate activity recognition in real-world scenarios. To address this challenge, some investigation of deep learning-based models for identifying human behavior in Internet of Things applications has been studied and discussed in this chapter. Also in this study enhanced activity recognition and accuracy by integrating attention into multi-head convolutional neural networks to improve feature extraction and selection is also have been highlighted.