Towards Explainable Fuzzy AI: Concepts, Paradigms, Tools, and Techniques

Modern AI techniques -- especially deep learning -- provide, inmany cases, very good recommendations: where a self-driving carshould go, whether to give a company a loan, etc. The problem isthat not all these recommendations are good -- and since deeplearning provides no explanations, we cannot tell whichrecommendations are good.It is therefore desirable to provide natural-language explanationof the numerical AI recommendations. The need to connect naturallanguage rules and numerical decisions is known since 1960s, whenthe need emerged to incorporate expert knowledge -- described byimprecise words like "small" -- into control and decision making.For this incorporation, a special "fuzzy" technique was invented,that led to many successful applications. This book described howthis technique can help to make AI more explainable.The book can be recommended for students, researchers, andpractitioners interested in explainable AI.