Reliability of medical production rules obtained by means of aggregate data mining

In the paper a method for designing production rules with uncertainty from medical aggregate data is proposed. Our main goal is to define the parameters that have an influence on the reliability of obtained rules. We distinguish two factors of reliability: global and internal ones. They determine a rule's importance in comparison to other obtained rules. Those rules compose the knowledge base of a medical Rule-Based System (RBS) aiding medical diagnosis and treatment.