Intelligent Rule Extraction in Complex Event Processing Platform for Health Monitoring Systems

Taking care of people with disabilities is essential in any situation and its cost is increasing every day. Many intelligent remote health monitoring systems have been developed from the past till now. These systems face challenges such as managing large volumes of data, real-time processing, and rule setting human errors. In this research, a complex event processing (CEP)-based platform is proposed for remote health monitoring and user behavior modeling in a health system. This platform can manage massive amounts of data in real-time. In addition, it can solve human errors in rule setting by extracting rules from the previous data using a rule-based learning method. Moreover, rule-based learning is an explainable method to get feedback from the domain experts. Experimental results in the hospital dataset showed that the PART rule-based learning has better accuracy (about 98.61%) in classifying the health conditions of the patients in comparison to other rule-based methods. On the other hand, the JRip rule-based learning has produced 16 fewer rules with similar accuracy (about 97.32%) compared to the PART classification method. Therefore, the JRip classifier is a better option for generating a small set of rules for using in the CEP engine.