Optimizing the decision-making process in fuzzy logic medical expert systems using historical data

Two of the most popular approaches to building decision support systems are fuzzy inference systems and supervised machine learning systems. However, fuzzy inference systems are based solely on expert decision-making process formalization and don’t take historical data into account, while machine learning systems infer certain statistical dependencies based solely on historical data, and those dependencies are very hard to formally analyze reason about from an expert point of view. The possible ways of combining these two approaches — building and optimizing medical fuzzy inference systems based on historical data are considered.