تجزیه و تحلیل رفتار در خدمات پرستاری و مراقبت با استفاده از مدل های رگرسیون خطی سوئیچ شده

In order to improve quality of nursing and caregiving services, it is important to analyze behaviors of service providers. Then, it is hard to directly analyze time series data. It is appropriate to find changing points and patterns of behaviors. In this paper, we propose a method for behavioral analysis using a switched linear regression model. First, an ICT-based system for nursing and caregiving services, which is developing by the authors, is summarized. Next, in order to analyze actual data obtained from this system, a switched linear regression model is introduced. An identification method of this model using mixed integer programming is proposed. Using this identification method, both the changing points of behaviors and the trend of the behavior between changing points can be obtained. Finally, the proposed method is applied to actual data in nursing and caregiving services.