New soft computing techniques for system modeling, pattern classifcation and image processing

Science has advanced significantly in the twentieth century, establishing proper disciplines in fields like physics, computer science, and molecular biology. Alongside this progress, interdisciplinary engineering ideas have emerged, such as artificial intelligence, fuzzy logic, and data mining. To create truly human-friendly technology in the twenty-first century, it is essential to integrate various methods beyond traditional disciplines. Soft computing plays a crucial role in developing such technology for our modern information society. Professor Rutkowski, a pioneer in this field, has dedicated years to publishing a wide range of original work. This volume, primarily based on his research, marks a significant milestone in soft computing by integrating diverse disciplines from information science and engineering. The book is divided into three parts, with the first focusing on probabilistic neural networks. As neural excitation is stochastic, it is logical to explore the Bayesian properties of the connectionist structures developed by Rutkowski. This innovative approach has proven particularly effective for addressing regression and classification challenges in dynamic environments. Throughout the book, key themes are drawn from theoretical subjects closely linked to practical applications.