Significant Factors Extraction: A Combined Logistic Regression and Apriori Association Rule Mining Approach

The global COVID-19 pandemic has become a phenomenon that has severely disrupted human life. It is widely recognized that taking faster, evidence-based measurements based on disease parameters is crucial for monitoring and preventing the further spread of COVID-19. One of the essential tasks in data mining is mining rules because rules provide concise statements of potentially important information that end users can easily understand. Therefore, attaining significant information in rules is the key to containing COVID-19 outbreaks. Our objective is to discover hidden but critical knowledge in the form of rules based on the risk factor dataset of COVID-19 patients. In this paper, we use association rule mining to extract information from rules in COVID-19 patients’ risk factor data that could be used to initiate prevention strategies. We discovered the rules of dead and recovered or hospitalized patients to understand and compare their characteristics. This approach can assist clinicians in effectively managing and treating diseases by providing valuable insight.