Enhancing Activity Recognition using CPD-based Activity Segmentation

Segmenting behavior-based sensor data and recognizing the activity that the data represents are vital steps in all applications of human activity learning such as health monitoring, security, and intervention. In this paper, we enhance activity recognition by identifying activity borders. To accomplish this goal, we introduce a change point detection-based activity segmentation model which segments behavior-driven sensor data in real time, which in turn increases the performance of activity recognition. We evaluate our proposed method on data collected from 29 smart homes. Results of this analysis indicate that the method not only provides useful information about their activity boundaries and transitions between activities but also increase the average accuracy of recognizing individuals' activities while they are performing their daily routines more than 1%.