Application of Artificial Intelligence Methods for System Design

Major: Computer Science (Design and programming of intelligent systems and devices)
Code of subject: 6.122.12.E.068
Credits: 5.00
Department: Computer-Aided Design
Lecturer: Mariana V. Levkovych, Ph.D., Associate professor of the CAD Department
Semester: 8 семестр
Mode of study: денна
Мета вивчення дисципліни: formation of future specialists of a modern level of information culture in the field of artificial intelligence; familiarization with the main methods and algorithms of artificial intelligence; study by students of methods and means of creating computer systems of artificial intelligence; obtaining information about the conceptual foundations of artificial intelligence, methods of presenting knowledge and knowledge bases; fuzzy logic systems; the structure and possibilities of using expert systems; basic concepts of pattern recognition systems; genetic algorithms; acquainting students with topical issues of using software tools for designing artificial intelligence systems.
Завдання: The study of an academic discipline involves the formation and development of students' competencies: general: - providing knowledge on the application of artificial intelligence methods for designing systems, as well as studying the principles of work in expert systems and the principles of programming in a high-level logical language. professional: - ability to identify statistical regularities of non-deterministics phenomena, application of computational intelligence methods, in particular statistical, neural network and fuzzy data processing, computer methods learning and genetic programming, etc. - ability to think logically, draw logical conclusions, use of formal languages and models of algorithmic calculations, design, development and analysis of algorithms, evaluation of their effectiveness and complexity, solvability and undecidability of algorithmic problems for adequate modeling of subject areas and creation of software and information systems; - ability to intelligently analyze data based on methods of computational intelligence, including large and poorly structured ones data, their operational processing and visualization of analysis results in the process of solving applied problems.
Learning outcomes: As a result of studying the discipline "Application of artificial intelligence methods for system design", students should: - know: basic methods of presenting knowledge, principles of fuzzy logical deduction; structure of expert systems; structure and principles of operation of artificial neural networks; basics of genetic algorithms; basic methods of pattern recognition. - be able to: use methods of practical knowledge acquisition (textological methods, communicative methods, individual methods, expert games); structure and formalize knowledge (dual design strategy, object-structural approach, OSA algorithm or practical methods of structuring); create a knowledge base for an expert system; create knowledge models: production, semantic networks, frames, formal logical models for further use of the model in the expert system; develop the structure of the expert system by analyzing and using the professional knowledge obtained from an expert in the subject field; develop and implement the software of the expert system, knowledge base, subsystems of the logical explanation of the system conclusion, user interface and support the developed system; use fuzzy logical deduction; create, train and use artificial neural networks; solve optimization problems using genetic algorithms; perform pattern recognition.
Required prior and related subjects: Prerequisites: "Discrete mathematics", "Mathematical methods of operations research", "Systems of intellectual analysis and visualization of data", "Technologies of distributed systems and parallel computing", "Application artificial systems intelligence in technological solutions (together with KR)", "Methods of machine learning in design systems".
Summary of the subject: The educational discipline "Application of artificial intelligence methods for system design" is a component of the educational and professional program "Design and programming of intelligent systems and devices" for training specialists at the first (bachelor's) level of higher education. This discipline belongs to the list of disciplines of the student's free choice. It is taught in the 8th semester in the amount of 150 hours (5 ECTS credits), in particular: lectures – 18 hours, laboratory classes – 36 hours, independent work – 96 hours. The discipline ends with an exam. Decision-making under conditions of uncertainty, business analytics, recognition of audio, video and text information, machine learning, computer vision, robotics, game development, creation of intelligent systems in various subject areas, diagnostics and forecasting are only a small part of the areas of artificial intelligence intelligence. Building an information society requires the development and application of intellectual analysis of increasingly large amounts of data. The creation of artifacts capable of replacing a person when making a decision in complex, ambiguous, problematic situations is an integral part of the paradigm of social development. The subject of study of the academic discipline is the theoretical foundations and practical aspects of the areas of artificial intelligence.
Опис: Topic 1. Basic concepts of artificial intelligence. Modern trends and approaches to the creation of artificial intelligence systems (AIS). (Philosophical aspects of artificial intelligence; concepts of artificial intelligence; intellectual system and artificial intelligence system; intellectual task (IT); approaches to building artificial intelligence systems). Topic 2. Representation of knowledge for the design of artificial intelligence systems. (Knowledge and data in AIS; models of knowledge representation in AIS; production models of knowledge representation; management of the search for solutions in production systems; advantages and disadvantages of production systems; examples of production systems). Topic 3. Search for solutions to an intellectual problem (IP) in the space of states. "Blind methods". (Classification of IP search methods in the state space; depth and width search; advantages and disadvantages of "blind" search methods; heuristic search methods). Topic 4. Search for solutions in the space of states. (Algorithm A*; genetic algorithm; methods of finding IS solutions in case of reduction of problems to a set of sub-problems (reduction method)). Topic 5. Expert systems. (Purpose and principles of construction; classes of problems that are solved with the help of expert systems; generalized architecture; stages of development). Topic 6. Knowledge-based problem solvers. (Fuzzy logic; scope and basic concepts). Topic 7. Semantic networks (SN), frames and fuzzy logic. (Basic concepts of SN; types of SN; methods of description of SN; basic concepts; frame structure; frame systems).
Assessment methods and criteria: - current control (45%): written reports on laboratory work, oral survey; - final control (55%, examination control): testing (45%), oral component (10%).
Критерії оцінювання результатів навчання: Current control (laboratory work)-60 points. Current control-40 points. Together for the discipline - 100 points.
Recommended books: 1. Kotsovsky V. M. Methods and systems of artificial intelligence. [Electronic resource]: Synopsis of lectures, V. M. Kotsovsky, Uzhhorod, 2016. – 75 p. 2. Hlybovets M.M., Otetskyi O.V. Artificial Intelligence. Textbook. - K: KM Academy Publishing House, 2002, - 366 p. 3.