Uncertainty Quantification and Propagation under Traditional AI Based Data Processing (and Related Topics): Legacy of Grigory Tseytin: 660

We need to revisit traditional methods for quantifying and propagating uncertainty, to see how they can help with these new challenges. The book covers uncertainty of measurement results and uncertainty inherent in natural-languages text -- by using both linguistic and traditional AI techniques (e.g., fuzzy). Why revisit uncertainty? Data processing is now often performed by Large Language Models (LLMs) and other AI tools that use natural-language texts. Many LLMs' results are spectacular, but often, there is no good indication of their accuracy. We need to revisit traditional methods for quantifying and propagating uncertainty, to see how they can help with these new challenges. The book covers uncertainty of measurement results and uncertainty inherent in natural-languages text -- by using both linguistic and traditional AI techniques (e.g., fuzzy). It contains both general results -- e.g., what can be computed -- and applications to engineering, physics, chemistry, and education. It also analyzes the effect of emerging computing paradigms -- such as quantum computing -- on uncertainty-related computations. This book can be recommended to everyone -- from students to researchers -- who is eager to learn, apply, and improve the uncertainty-related techniques.