The current state of development of information technologies and computer communications has led to the fact that we are surrounded by petabytes of unstructured and poorly structured digital information. Digital infrastructure through modern means of communication: computers, mobile devices, Internet of Things (IoT), provides online and offline access to a wide range of sources, such as indexed data stores created by search engines (Google, Bing, Yahoo!, Baidu , DuckDuckGo); digital encyclopedias (Wikipedia, Encyclopædia Britannica); industry ontologies based on the Semantic Web; open data from government and non-governmental organizations with access through application programming interfaces (APIs). The power of personal computing and communication devices is increasing, theories and methods in the field of knowledge representation are developing, new programming languages and tools in the field of artificial intelligence (AI) are appearing, suitable for solving problems of knowledge representation (KR), logical inference, optimization, search.
The urgent task is to combine access to practically unlimited information with the current level of development of theories, methods, and technologies in the field of AI to solve the current problems of the average user in finding, obtaining, and organizing information by turning it into knowledge. The result of such integration should be an intelligent agent (IA) that will create a knowledge base (KB) in an automated mode according to directions defined by the user, conduct a two-way exchange of requests with the user in a formalized subset of natural language, answer questions and ask them himself. Such an IA as an application on a computer or gadget can become a personal secretary-referent, a consultant or even an expert on user-defined issues.
The paper proposes to investigate approaches to the application of knowledge-based IA (KBIA) in the field of electronic distance education - E-Learning (EL). EL is considered as a personalized and customizable service that allows each user to have easy access through the network to the tools, services and digital tools necessary for learning [1]; as "a combination of educational services and computer technologies to ensure high value of integrated learning: anytime, anywhere" [2]. "It's not about taking a course and placing it on your desktop, it's a new combination of resources, interactivity, performance support and structured learning activities" [2].
The object of the study is KBIA, which operate in a changing virtual environment. The subject of the research is the analysis of agent-oriented design (AOD) approaches and KR methods for their application in EL support systems, such as LMS (Learning Management System), ITS (Intelligent Tutoring System), CBE (Competency-Based Education), CAI (Computer -Assisted Instruction), PLE (Personal Learning Environment).
The purpose of the research is to build a prototype of a software module as a system of interacting KBIA, the implementation of which would allow for the integration of EL support systems with components of structured knowledge and processes that represent such a subject area as higher education with its standards, resources, programs and actors.
Based on the subject of research, a sequence of tasks is defined, the implementation of which allows to achieve the goals and objectives set in the work.
At the analysis stage, determine the possibility and limits of using probabilistic, fuzzy, dynamic, temporal logic, event counting; methods of automatic proof of theorems and resolutions; algorithms of graph theory, dynamic programming, reinforcement learning (RL), Bayesian analysis; KR methods: declarative, procedural, hybrid; compare and choose programming languages and design and implementation tools for use.
At the design stage, develop the overall architecture of the system; define data structures and procedure algorithms.
At the implementation stage, according to the project, create a prototype of the software module as a multi-agent system (MAS) of interacting KBIA.
The main methods of performing the work are the search and analysis of sources relevant to the object, subject, goal and tasks of the research; practical testing of selected approaches, methods and tools on test and real data; comparison and conclusions regarding the possible application of the tested artifacts. Sources include websites, books, scientific articles, discussions, standards, patents, software documentation, program source codes. Practical testing is carried out by means of computer modeling, programming of prototypes, preparation and analysis of data, execution of test examples. Comparing the results of the testing allows you to make a decision on the further use of the researched approaches, methods, structures, tools, languages, packages, and libraries to fulfill the tasks set in the research.
The proposed approaches to the application of KBIA in the field of EL will have a certain innovative value and actual practical value in the case of using the results of this master's thesis to create a real software project and product to expand the existing EL system towards greater interactivity, adaptability and intelligence.