Prof. R.A. ALIEV

Director of Joint MBA Program, Georgia State University, USA and Azerbaijan State Oil and Industry University,
Prof. of the Near East University
Department of Artificial Intelligence, Near East University, Lefkosa, North Cyprus

Rafik A. Aliev received the Ph.D. and Doctorate degrees from the Institute of Control Problems, Moscow, Russia, in 1967 and 1975, respectively. His major fields of study are decision theory with imperfect information, fuzzy logic, soft computing and control theory. He is a Professor and the Head of the Department of the joint MBA Program between Georgia State University (Atlanta, GA, USA) and the Azerbaijan State Oil Academy. His current research is focused on the generalized theory of stability, recurrent fuzzy neural networks, fuzzy type-2 systems, evolutionary computation, decision theory with imperfect information, calculus with Z-numbers, and fuzzy economics. He has over 350 scientific publications, including 55 books published by well-known international publishers Springer, World Scientific, Elsevier, etc.,  15 edited volumes, and 280 research papers. Dr. Aliev is a regular Chairman of the International Conferences on Applications of Fuzzy Systems and Soft Computing and International Conferences on Soft Computing and Computing with Words. He is an Editor of the Journal of Advanced Computational Intelligence and Intelligent Informatics (Japan), associate editor of the Information Sciences journal, a member of the Editorial Boards of the International Journal of Information Technology and Decision Making, International Journal of Web-based Communities (The Netherlands), Iranian Journal of Fuzzy Systems (Iran), International Journal of Advances in Fuzzy Mathematics (Italy), and International Journal “Intelligent Automation and Soft Computing.” He is a series editor of “Advances in Uncertain Computation” and “World Scientific”. He was awarded the USSR State Prize in the field of Science (1983), the USA Fulbright Award (1997), and the Lifetime Achievement Award in Science (2014).

 He supervised more than 150 Ph.D. students and over 30 Doctor of Sciences.

Keynote Speech:

Z-Neural Networks and Inference System


Z-number based inference system, being built on the grounds of fuzzy inference systems, but further extending their potential to better express the inherent uncertainty and randomness of real-world systems and human-like decision-making habits, should suggest higher computational power, and allow more reliability and trustworthiness in regard to decisions made on its basis as compared to ordinary fuzzy and crisp systems.

We propose Z-number-based Neural Networks Inference System (ZNIS) as a prototype of an adaptive Z-Number based inference system capable of accepting the rules from a human user, generating the rules from a clustering technique, and learning/adapting the rule parameters from raw data using the DEC algorithm.

The main advantages of ZNIS over other inference systems are better semantic expressing power, higher degree of perception and interpretability of the linguistic rules by humans, and a higher confidence in the reliability of achieved decision due to the transparency of the underlying decision-making mechanism. 

Experiments on Parkinson disease, and non-linear system identification have shown that ZNIS performance is better than FLS Type 2 and far superior to FLS Type 1, showing on average 2-3 times lower MSE.