This book is about techniques that can make AI more explainable.
While AI techniques have been spectacularly successful, they are not perfect: their answers are often wrong. To distinguish between great and flawed recommendations, it is desirable to make AI systems explain their recommendations.
This means translating numerical computationsthat AI systems performinto human-understandable natural-language form. It thus makes sense to use existing techniquese.g., fuzzythat connect numerical computations with natural language.
The use of fuzzy techniques to make AI explainable is the book's focus. This book contains both theoretical results and applicationsto aerospace engineering, agriculture, biology, digital twins, education, law enforcement, medicine, etc. It can be recommended to students and practitioners interested in the state-of-the-art fuzzy-related explainable AI, and to researchers focused on remaining challenges.

Covers all aspects of theory and applications in Fuzzy Systems, Soft Computing, and Explainable AI Contains papers presented at the 2024 Annual Conference of the North American Fuzzy Information Processing Society Written by experts in the field

Inhalt

Modified Hargreaves Equation Using Interval Type 2 Fuzzy Logic System for Predication of Reference Evapotranspiration Case Study for Arid.- From Fuzzy to Clear Visualizing System through Heatmaps Derived from Approximate Inverse Model Explanations and Membership Function Activity.- Evolutionary Optimization of 1D CNN for Non contact Respiration Pattern Classification.- On the fuzzy Ishigami function.