Basket based sorting method for activity recognition in smart environments

Activity recognition in smart environments is an important technology for assisted living and e-health. Recently there are growing interests in applying machine learning algorithms to activity recognition tasks. One of the main problems with previous work is that concurrent activities of multiple targets may fail the sensor event based prediction if no proper preprocessing method is used. To address this problem, this paper proposes a new basket based sorting method for multiple target classification in a sensor based smart environment, which can significantly improve activity recognition accuracy in real-time monitoring. The proposed structure and method can be plugged into different machine learning models to achieve good activity recognition performance.