3P2-X06 An Anomaly Detection Method for an Elderly Monitoring System Using Financial Theories(Nursing and Mechatronics)
Kazuyuki Komichi, Hiroshi NOGUCHI, Hiromi Sanada, Taketoshi Mori · The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec) · 2014
As the number of the elderly living alone is on the rise, the need for home monitoring systems for the elderly is growing. In this paper, we propose an anomaly detection method in a home monitoring system for the elderly that takes a time-series of subject's activity level data captured by simplistic motion sensors as its input. First, the activity level data were classified into four types based on the trend and volatility of the data. Second, financial models suitable for the characteristics of the four types were applied and thus constituted an anomaly detection method. For example, the underlying asset price model was applied to a type of activity level data exhibiting the characteristics of trend-presence and constant-volatility. Here an anomaly was defined as an outlier from the 95% confidence interval calculated by the models.