RunnerDNA: Interpretable indicators and model to characterize human activity pattern and individual difference (2201.07370v1)

Abstract: Human activity analysis based on sensor data plays a significant role in behavior sensing, human-machine interaction, health care, and so on. The current research focused on recognizing human activity and posture at the activity pattern level, neglecting the effective fusion of multi-sensor data and assessing different movement styles at the individual level, thus introducing the challenge to distinguish individuals in the same movement. In this study, the concept of RunnerDNA, consisting of five interpretable indicators, balance, stride, steering, stability, and amplitude, was proposed to describe human activity at the individual level. We collected smartphone multi-sensor data from 33 volunteers who engaged in physical activities such as walking, running, and bicycling and calculated the data into five indicators of RunnerDNA. The indicators were then used to build random forest models and recognize movement activities and the identity of users. The results show that the proposed model has high accuracy in identifying activities (accuracy of 0.679) and is also effective in predicting the identity of running users. Furthermore, the accuracy of the human activity recognition model has significant improved by combing RunnerDNA and two motion feature indicators, velocity, and acceleration. Results demonstrate that RunnerDNA is an effective way to describe an individual's physical activity and helps us understand individual differences in sports style, and the significant differences in balance and amplitude between men and women were found.