UDC 004.891.2

USE OF EXPERT SYSTEMS IN EDUCATION1

M.S. Chvanova, I.A. Kiseleva, A.A. Molchanov, A.N. Bozyukova

Tambov State University named after G.R. Derzhavin Russia, Tambov. e-mail: [email protected]

The article deals with the problems of application and development of expert systems in education, as well as specific examples of the use of such systems. The authors consider it necessary to use the apparatus of fuzzy logic for the design and development of an intelligent subsystem.

Key words: information technologies, expert system, fuzzy logic, education system.

The study of research on the problem showed that in the early eighties, an independent direction was formed in research on artificial intelligence, called "expert systems" (ES). Researchers in the field of ES often use the term "knowledge engineering" introduced by E. Feigenbaum to name their discipline. Expert systems (ES) are a set of programs that perform the functions of an expert in solving problems from a certain subject area. The name is due to the fact that they seem to imitate people who are experts.

Each expert system consists of three parts: a very large database of modern data, a subsystem for generating questions, and a set of rules that allow drawing conclusions. Some expert systems can talk about the method they use when reaching their conclusion.

In our country, the current state of developments in the field of expert systems can be characterized as a stage of ever-increasing interest among a wide range of economists, financiers, teachers, engineers, doctors, psychologists, programmers, linguists. Unfortunately, this interest has insufficient material support: a clear lack of textbooks and specialized literature, the absence of symbolic processors and artificial intelligence workstations, and limited funding.

1 The topic was supported within the framework of the Program of the Ministry of Education and Science "Conducting scientific research by young scientists - candidates of science" No. 14.В37.21.1141, 20122013.

financing of research in this area, the weak domestic market for software products for the development of expert systems, and the high cost of existing ones makes their application and analysis of the effectiveness of their application practically inaccessible.

It is well known that the process of creating an expert system requires the participation of highly qualified specialists in the field of artificial intelligence, which are still being produced by a small number of higher educational institutions in the country.

An analysis of theoretical research and teaching practice has shown that insufficient attention is paid to the development of expert systems in the system of distance education. Expert systems in the field of education are most often used to build a knowledge base that allows you to reflect the minimum required content of the subject area, taking into account its quantitative and qualitative assessments.

Research in the field of application and development of expert systems in education, as we believe, can be divided into three groups. It seems possible to refer to the first group the authors who study the theoretical and pedagogical aspects of the use of expert systems in education. The second group includes authors who have developed specific expert learning systems together with teachers based on well-known technologies. The third group - authors who explore new approaches to the creation of expert systems in education.

Research in the field of application and development of expert systems in education

Research institutes, as we believe, can be conditionally divided into three groups. It seems possible to refer to the first group the authors who study the theoretical and pedagogical aspects of the use of expert systems in education. The second group includes authors who have developed specific expert learning systems together with teachers based on well-known technologies. The third group - authors who explore new approaches to the creation of expert systems in education.

Let's consider the first group of publications that analyze the theoretical and pedagogical aspects of the application of expert systems.

In the study of N.L. Yugovoy designed the content of specialized training using an expert system. The author considers an expert system for diagnosing the levels of learning and professional preferences of students, which is implemented on the basis of building a frame model of profile educational information, establishing subject-subject relationships of participants in the educational process: student, teacher, teacher-cognitologist.

N.M. Antipina developed a technology for the formation of professional methodological skills in the course of independent work of students of pedagogical universities using an expert system. A specialized training expert system developed by the author is capable of issuing individual tasks of various levels of difficulty in the course of independent work of students at a computer, developing recommendations on how to complete them, providing assistance in the form of consultations, monitoring the knowledge and skills of students at various stages of their implementation of methodological tasks and etc.

N.L. Kiryukhina developed a model of an expert system for diagnosing students' knowledge of psychology. The author considers an expert system for solving the problem of diagnosing the psychological knowledge of students, testing hypotheses about the correctness of the student's answers, the degree of assimilation of the material on various topics of the course. I.V. Grechin implements a new approach to the use of an expert system in learning technology.

He proposes a system that, using feedback interactively, generates and tracks the sequence of a train of reasoning chains.

ON THE. Baranova considers the issue of using expert systems in continuous pedagogical education. The expert system structures educational information and creates individual curricula for each student with reduced training periods, which increases the efficiency of learning, teaching and self-education processes.

A.B. Andreev, V.B. Moiseev, Yu.E. Usachev use expert systems to analyze students' knowledge in an open education environment. Analysis of the quality of knowledge is carried out with the help of an expert system of knowledge analysis. To implement such a system, the authors consider a structural approach to the creation of intelligent teaching and control computer systems. Thus, this approach makes it possible to develop effective tools for analyzing students' knowledge based on the use of a structural model of educational material. The structural unit of the totality of knowledge in the proposed model is a concept that has content and volume.

E.V. Myagkova considers the possibility of using expert systems as information technologies in the field of higher education. According to the author, expertise lies in the presence in the expert teaching system of knowledge on teaching methods, thanks to which it helps teachers to teach, and students to learn. The main goal of the implementation of the expert training system, according to the author of the article, is the training and assessment of the current level of knowledge of the student in relation to the level of knowledge of the teacher. Thus, a comparison of two grids (the reference one, reflecting the teacher's ideas, and the grid filled in by the student during the dialogue) allows us to evaluate the differences in the teacher's and student's ideas.

B.M. Moskovkin built a simulation expert system for choosing universities for training. The author made a brief review of foreign studies in

the field of modeling decision-making processes on the choice of colleges and universities for further education. At the conceptual level, an appropriate simulation expert system is built.

Let us consider the second group of publications, which deal with expert systems for education developed jointly with teachers based on known technologies.

E.Yu. Levina developed an intra-university diagnostics of the quality of education based on an automated expert system, the application of which, in fact, boils down to diagnosing the quality of the educational process at a university, which allows, on the basis of information tools and mathematical methods, to manage databases for the implementation of research procedures and analysis of statistics on the results of the educational process , development of recommendations for making managerial decisions to ensure the quality of education.

M.A. Smirnova has developed an expert system for assessing the quality of pedagogical training of a future teacher, which boils down to assessing the quality of his training at school, which makes it possible to investigate the level of a teacher's preparedness.

L.S. Bolotova, based on the technology of expert systems of situational management, adaptive distance learning for decision-making is implemented. As instrumental software, experimental samples of instrumental problematic subject-oriented expert systems for situational management of municipalities and small businesses were developed on the basis of the developed situational simulator - simulator.

A computer decision-making system based on the results of expert evaluation in the tasks of assessing the quality of education, developed by O.G. Berestneva and O.V. Marukhina makes it possible to single out the most substantiated statements of specialist experts and use them, ultimately, to prepare various decisions. The universal software product developed by the authors and described in the article makes it possible to most optimally solve the problem of assessing the quality of the educational process based on the results of expert evaluation.

E.F. Snizhko considers the methodology of using expert systems to adjust the learning process and evaluate the effectiveness of pedagogical software. In the course of the study, the author developed an experimental fragment of a pedagogical software tool for learning the Prolog language for students of the 9th grade of secondary school in order to demonstrate the main points of the developed methodology and its experimental verification. The expert system built into the pedagogical software tool was brought to the level of a demonstration prototype.

An analysis of the literature in this area showed that one of the approaches to the creation of expert systems are attempts to propose the use of fuzzy logic methods based on the theory of fuzzy sets.

V.S. Toykin identifies several reasons on the basis of which preference is given to the use of systems with fuzzy logic:

It is conceptually easier to understand;

It is a flexible system and is resistant to inaccurate inputs;

It can model non-linear functions of arbitrary complexity;

It takes into account the experience of expert specialists;

It is based on the natural language of human communication.

I.V. Solodovnikov, O.V. Rogozin, O.V. Shu-ruev consider the general principles of building a software package capable of producing a comprehensive student performance in a semester with the help of an expert system, using elements of the fuzzy logic apparatus.

Lecture attendance. The attendance score was calculated by the arithmetic mean of all available scores;

Seminar work. Evaluation of performance was carried out in a similar way;

Performance of control works. Evaluation of the performance of control work was carried out taking into account the coefficient of complexity;

Doing homework. Performance evaluation was carried out in a similar way.

To assess academic performance, the authors used linguistic variables: “attended lectures”, “worked at a seminar”, “performed tests”, “did homework”. The characteristics of these variables were the concepts of "activity", "efficiency", "assessment". This approach makes it possible to analyze the work of the student and, on the basis of the formulated criteria, evaluate the effectiveness of the quality of the student's knowledge.

Based on fuzzy logic models I.V. Samoilo, D.O. Zhukov consider the problem of creating expert systems that make it possible to give recommendations on professional orientation to a specific applicant.

Group of variables (O) - estimates. In the general case, for a group of variables O, one can write O = (O1, O2, O3, ..., Op).

Group of variables (C) - psychological tests aimed at identifying abilities related to learning and intelligence.

Group of variables (C) - characteristics of the student's personality.

The group of variables (M) is the results of diagnosing the student's sphere of interest: M = (t1, t2, ..., tk).

Thus, the prototype of such a system made it possible to form a mechanism for managing the cathedral choice:

The applicant enters the start page of the system, enters school grades and (or) enters the results of the unified state exam, the results of current academic performance, the system evaluates the reliability of the result using fuzzy logic;

The user is tested for the psychological characteristics of the personality and the ability to learn, areas of interest with

evaluating the reliability of the result using fuzzy logic;

The automated expert system (AES) checks whether the applicant meets the requirements of the department (educational institution). If “yes”, then with the help of the managing educational environment, the user’s knowledge is corrected, optimal conditions for overcoming the departmental “barrier” are created, in addition, the user has the opportunity to refuse to fight for the department that interests him and continue his education at the department where his achievements allow;

Subsequent tests are held every six months. The test results help to track the dynamics of the student's development, to choose the optimal strategy for the formation of a future professional.

O.A. Melikhov considers the issue of the possibility of implementing an expert system for monitoring the educational process of a higher education institution based on a fuzzy approach to modeling intelligent systems. This approach uses "linguistic" variables, the relationships between which are described using fuzzy statements and fuzzy algorithms.

Building a system for monitoring the educational process includes the following steps:

Formulation of learning objectives, determination of the level of requirements of each teacher (higher, middle, lower);

Building a monitoring system, determining the degree of training in each discipline. Indicators: discrimination, memorization, understanding, elementary skills, knowledge transfer;

Determination of the actual effectiveness of the teacher's activities based on indicators of the degree of students' learning. The main indicators of the effectiveness of the teacher's activity are the strength, depth and awareness of the knowledge of the trainees. These same indicators determine the quality of education.

DI. Popov in his work considers the intellectual distance learning system (ISDO) "KnowledgeCT" based on Internet technologies, which is planned to be used for educational purposes by the Center for Distance Education. It allows

not only evaluate knowledge, but also collect data about students, which is necessary to create mathematical models of the student, collect statistics.

Knowledge is assessed using an adaptive testing system based on fuzzy logic methods and algorithms: for each level of complexity, a discipline expert (teacher) needs to develop an appropriate set of questions. Such a system makes it possible to make the learning process more flexible, take into account the individual characteristics of the student and improve the accuracy of assessing the student's knowledge.

V.M. Kureichik, V.V. Markov, Yu.A. Kravchenko in their work explore an approach to designing intelligent distance learning systems based on rules and precedent-based inference technologies.

Expert systems model the decision-making process of an expert as a deductive process using rule-based inference. A set of rules is laid into the system, according to which, based on the input data, a conclusion is generated on the adequacy of the proposed model. There is a drawback: the deductive model emulates one of the rarer approaches that an expert takes when solving a problem.

Case based inference draws conclusions from the results of searching for analogies stored in the case database. This method is effective in situations where the main source of knowledge about a problem or situation is experience, not theory; solutions are not unique to a particular situation and can be used in others to solve similar problems; the purpose of the inference is not a guaranteed correct solution, but the best possible one. The implementation of this inference technology can be carried out using neural network algorithms.

An analysis of the literature on the problem of using expert systems in the distance learning system showed that this area has been little studied and is only being developed, as evidenced by the small number of publications of research teachers working in this problem field. Publications in this area are mainly predictive in nature.

There is an interest in distributed intelligent systems in the distance learning system, however, it is not entirely clear how the educational process can be effectively organized so that it leads to the desired quality of education. Apparently, we should talk, first of all, about the construction of pedagogical educational models in the system of open education.

In our opinion, the problem is due to the fact that a significant part of researchers in the field of distance learning technologies transfer methods and techniques known in practice, filling distance learning with them. At the same time, it is quite obvious that new technologies in education should be based on the principle of "new tasks". Advanced technologies carry a new solution, new methods, new approaches, new opportunities that are not yet known to the education system. Now it has become obvious that the "traditional lecture" and "traditional textbook" are ineffective in distance learning. We need organized and directed access to dynamic systems of up-to-date information, we need “automated consultations” available at any time, we need new ways and methods of organizing joint project activities, and much more.

To date, certain experience has been accumulated in the transfer of part of the intellectual functions for organizing and conducting the educational process in the system of open education to informatization tools.

So, G.A. Samigulina gives an example of an intelligent expert system of distance learning based on artificial immune systems, which allows, depending on the student's belonging to a certain group, to assess his intellectual potential and, in accordance with it, promptly provide an individual training program. The output is a comprehensive assessment of knowledge, differentiation of students and a forecast of the quality of the education received. Groups are determined by experts and correspond to certain knowledge, practical skills, creativity, logical thinking, etc. The developed expert system implies the implementation of subsystems:

- "Information subsystem" - development of methods and means of information storage, development of databases, knowledge bases. Includes electronic textbooks, references, catalogs, libraries, etc.;

- "Intellectual subsystem" - training of the immune network, processing of multidimensional data in real time. The use of an algorithm for estimating binding energies based on the properties of homologous peptides makes it possible to reduce errors in predicting an intelligent system, which makes it possible to train students in accordance with their individual characteristics;

- "Training subsystem" develops methods, means and forms of presentation of training information adapted to a specific user, taking into account his individual characteristics. A schedule is drawn up for the scope of the required work and the timing of implementation;

- "Controlling subsystem" is designed for a comprehensive assessment of the student's knowledge in order to promptly adjust the program and the learning process.

Thus, as a result of the operational analysis of the knowledge of a huge number of students, it is possible to quickly correct the learning process, since the expert system offers an individual training program.

An analysis of research on expert systems in the field of distance education has shown that this is a new and relevant area in science that has been little studied. Often, educators understand the expert system as testing students in a particular distance education system and examining their knowledge.

So, A.V. Zubov and T.S. Denisova developed complex expert Internet systems for distance learning based on the Finport Training System distance learning system. The system has the ability to develop training courses, conduct training and certification, and at the same time analyze the results and effectiveness of training based on tests developed by highly qualified specialists.

V.G. Nikitaev and E.Yu. Berdnikovi-than developed multimedia cur-

distance learning courses for doctors in histological and cytological diagnostics using expert systems based on the Moodle content management system. The system allows you to add courses to the content and, based on testing, check the level of assimilation of the material depending on the students' response.

Thus, in distance learning systems it is possible to make an expert assessment of knowledge based on test tasks developed by specialists.

At the same time, in our opinion, distance learning technologies require the use of many subsystems to relieve the routine burden on organizers and tutors. This load increases due to the fact that a person chooses for himself his own rhythm, pace and time of learning. Individualization requires a developed automated system of "intelligent" prompts, assistance, consultations throughout the entire period of distance learning and when using various educational methods and techniques: lectures, practices, project activities, conferences, etc. Only unique questions are addressed to the expert teacher. Based on the analysis of publications and personal practice of organizing distance learning, we came to the conclusion that the above intellectual subsystems can be organized on a different theoretical and program basis in the form of separate modules connected to the system. This is due to the fact that subsystems carry different intellectual “loads”: somewhere it is enough to use traditional logic when designing a specific subsystem, and in another case it is convenient to create a subsystem using the fuzzy logic apparatus.

Bibliography

1. Andreichikov A.V., Andreichikova O.N. Intelligent information systems. M.: Finance and statistics, 2006.

2. Yugova N.L. Designing the content of specialized education using an expert system: Ph.D. dis. ... cand. ped. Sciences. Izhevsk, 2006.

3. Antipina N.M. Technology for the formation of professional methodological skills in the course of independent work of students of pedagogical

of higher education institutions with the use of an expert system: Ph.D. ... cand. ped. Sciences. M., 2000.

4. Kiryukhina N.L. Model of an expert system for diagnosing students' knowledge in psychology: dis. ... cand. psychol. Sciences. M., 1998.

5. Grechin I.V. A new approach to the expert system in teaching technology // Izvestiya TSURE. Thematic issue "Intellectual CAD". Taganrog: TRTU, 2001. No. 4.

6. Baranova N.A. On the question of the application of expert systems in continuous pedagogical education // Education and Science. 2008. No. 4. S. 24-28.

7. Moiseev V.B., Andreev A.B. Intrauniversity system for ensuring the quality of training of specialists // Engineering education.

2005. No. 3. S. 62-74.

8. Myagkova E.V. The role and possibility of application

of expert systems as information technologies in the field of higher education // Information technologies in design and production: scientific and technical

magazine. 2008. No. 1. S. 13-15.

9. Moskovkin V.M. Simulation expert system for choosing universities for training // NTI. Series 2. 2009. No. 10. S. 19-21.

10. Levina E.Yu. Intrauniversity diagnostics of the quality of education based on an automated expert system: Ph.D. dis. ... cand. ped. Sciences. Kazan, 2008.

11. Smirnova M.A. Application of an expert system for assessing the quality of pedagogical training of a future teacher: dis. ... cand. ped. Sciences. Tula, 1997.

12. Bolotova L.S. [et al.] Adaptive distance learning for decision-making based on the technology of expert systems of situational systems of situational management of municipalities and small businesses // Scientific research. Issue. 5. Annual report on the main results of scientific research, 2003. M., 2004.

13. Berestneva O.G., Marukhina O.V. Computer decision-making system based on the results of expert evaluation in the tasks of assessing the quality of education // Proceedings of the regional scientific and methodological conference "Modern education: systems and practice of quality assurance", Tomsk, January 29-30. 2002. Tomsk, 2002. S. 29-30.

14. Snizhko E.A. Methodology for the use of expert systems for adjusting the learning process and evaluating the effectiveness of teaching staff: dis. ... cand. ped. Sciences. SPb., 1997.

15. Toiskin V.S. Intelligent information systems: a tutorial. Stavropol: Publishing house of SGPI, 2010. Part 2.

16. Expert system for evaluating the effectiveness of training based on the mathematical apparatus of fuzzy logic / I.V. Solodovnikov [and others] // Quality. Innovation. Education. 2006. No. 1. S. 19-22.

17. Samoilo I.V., Zhukov D.O. Information technologies in providing a new quality of higher education // Collection of scientific articles. Book. 2. Proceedings of the All-Russian scientific-practical conference with international participation "Information technologies in providing a new quality of higher education (April 14-15, 2010, Moscow, NUST MISiS)". M.: Research Center for Quality Problems in Training Specialists at NUST MISIS, 2010, pp. 89-95.

18. Melikhova O.A., Melikhova Z.A. The use of fuzzy mathematics in modeling artificial intelligence systems // Thematic issue "Intelligent CAD": in 2 volumes. Taganrog: Publishing House of TRTU, 2007. P. 113-119.

19. Popov D.I. Designing intelligent systems for distance education // Proceedings of the Southern Federal University. Series: Engineering sciences. 2001. V. 22. No. 4. S. 325-332.

20. Astanin S.V. [et al.] Intellectual educational environment of distance learning // News of artificial intelligence. 2003. No. 1.

21. Samigulina G.A. Intelligent expert system of distance learning based on artificial immune systems // Information technologies of modeling and control. 2007. Issue. 9 (43). pp. 1019-1024.

22. Zubov A.V., Denisova T.S. Creation of complex expert Internet systems for distance learning // Informatization of education and science. Moscow: State Scientific Research Institute of Information Technologies and Telecommunications, 2010.

23. V. G. Nikitaev and E. Yu. Development of multimedia distance learning courses for doctors on histological and cytological diagnostics using expert systems. Fundamental research: a scientific journal. 2007. No. 12. S. 334-334.

1. Andrejchikov A.V., Andrejchikova O.N. Intel-lektual "nye informacionnye sistemy. M .: Finansy i statistika, 2006.

2. Jugova N.L. Konstruirovanie soderzhanija pro-fil "nogo obuchenija s primeneniem ekspertnoj sis-temy: avtoref. dis. ... kand. ped. nauk. Izhevsk,

3. Antipina N.M. Tehnologija formirovanija profes-sional "nyh metodicheskih umenij v hode samosto-jatel" noj raboty studentov pedagogicheskih vuzov

s primeneniem ekspertnoj system: dis. ... cand. ped. science. M., 2000.

4. Kirjuhina N.L. Model "ekspertnoj sistemy diag-nostiki znanij studentsov po psihologii: dis. ... kand. psihol. nauk. M., 1998.

5. Grechin I.V. Novyj podhod k jekspertnoj sisteme v tehnologii obuchenija // Izvestija TRTU. Tema-ticheskij vypusk "Intellektual" nye SAPR ". Taganrog: TRTU, 2001. No. 4. S. 343-344.

6. Baranova N.A. K voprosu o primenenii ekspertnyh sistem v nepreryvnom pedagogicheskom obrazo-vanii // Obrazovanie i nauka. 2008. No. 4. S. 24-28.

7. Moiseev V.B., Andreev A.B. Vnutrivuzovskaja sistema obespechenija kachestva podgotovki specialistov // Inzhenernoe obrazovanie. 2005. No. 3. S. 62-74.

8. Mjagkova E.V. Rol "i vozmozhnost" primenenija

ekspertnyh sistem kak informacionnyh tehnologij v sfere vysshego obrazovanija // Informacion-nye tehnologii v proektirovanii i proizvodstve: nauchno-tehnicheskij zhurnal. 2008. No. 1.

9. Moskovkin V.M. Imitacionnaja ekspertnaja sistema vybora universitetov dlja obuchenija // NTI. Serija 2. 2009. No. 10. S. 19-21.

10. Levina E.Ju. Vnutrivuzovskaja diagnostika kachestva obuchenija na osnove avtomatizirovan-noj ekspertnoj sistemy: avtoref. dis. ... cand. ped. science. Kazan", 2008.

11. Smirnova M.A. Primenenie ekspertnoj sistemy dlja ocenki kachestva pedagogicheskoj podgotovki budushhego uchitelja: dis. ... cand. ped. science. Tula, 1997.

12. Bolotova L.S. . Adaptivnoe distancionnoe obuchenie prinjatiju reshenij na osnove tehnologii ekspertnyh sistem situacionnogo sistem situacion-nogo upravlenija municipal "nymi obrazovanijami i malym biznesom // Nauchnye issledovanija. Vyp.

5. Ezhegodnyj otchet ob osnovnyh rezul "tatah nauchnoissledovatel" skih rabot, 2003. M., 2004.

13. Berestneva O.G., Maruhina O.V. Komp "juternaja sistema prinjatija reshenij po rezul" tatam jeks-pertnogo ocenivanija v zadachah ocenki kachest-va obrazovanija // Materialy regional "noj nauch-no-metodicheskoj konferencii "Sovremennoe ob-razovanie: sistemy i praktika obespechenija ka-chestva", Tomsk, January 29-30, 2002. Tomsk, 2002. S. 29-30.

14. Snizhko E.A. Metodika primenenija jekspertnyh sistem dlja korrektirovki processa obuchenija i ocenki jeffektivnosti PPS: dis. ... cand. ped. science. SPb., 1997.

15. Toiskin V.S. Intellectual "nye informacionnye sistemy: uchebnoe posobie. Stavropol": Izd-vo SGPI, 2010. Ch. 2.

16. Solodovnikov I.V. Ekspertnaja sistema ocenki jeffektivnosti obuchenija na osnove ma-

thematicheskogo apparata nechetkoj logiciki // Kachestvo. Innovacii. Obrazovanie, 2006. No. 1.

17. Samojlo I.V., Zhukov D.O. Informacionnye tehnologii v obespechenii novogo kachestva vys-shego obrazovanija // Sbornik nauchnyh statej. Kn. 2. Trudy Vserossijskoj nauchno-praktiche-skoj konferencii s mezhdunarodnym uchastiem "Informacionnye tehnologii v obespechenii no-vogo kachestva vysshego obrazovanija (Apr. 14-15, 2010, Moscow, NITU "MISiS")". M.: Issledo-vatel "skij centr problem kachestva podgotovki specialistov NITU "MISiS", 2010. S. 89-95.

18. Melihova O.A., Melihova Z.A. Ispol "zovanie nechetkoj matematiki pri modelirovanii sistem iskusstvennogo intellekta // Tematicheskij vypusk "Intellektual" nye SAPR": v 2 t. Taganrog: Izd-vo TRTU, 2007. S. 113-119.

19. Popov D.I. Proektirovanie intellektual "nyh sistem distancionnogo obrazovanija // Izvestija Juzhno-go federal" nogo universiteta. Serija: Technical sciences. 2001. T. 22. No. 4. S. 325-332.

20. Astanin S.V. Intellektual "naja obrazova-tel" naja sreda distancionnogo obuchenija // No-vosti iskusstvennogo intellekta. 2003. No. 1.

21. Samigulina G.A. Intellektual "naja jekspertnaja sistema distancionnogo obuchenija na osnove iskusstvennyh immunnyh sistem // Informacion-nye tehnologii modelirovanija i upravlenija.

2007. Vyp. 9 (43). S. 1019-1024.

22. Zubov A.V., Denisova T.S. Sozdanie kompleks-nyh ekspertnyh Internet-sistem dlja distancion-nogo obuchenija // Informatizacija obrazovanija i nauki. M .: Gosudarstvennyj nauchnoissledova-tel "skij institut informacionnyh tehnologij i tele-kommunikacij, 2010.

23. Nikitaev V.G., Berdnikovich E.Ju. Razrabotka mul "timedijnyh kursov distancionnogo obuchenija vrachej po gistologicheskoj i citologicheskoj diagnostike s primeneniem ekspertnyh sistem // Fundamental" nye issledovanija: nauchnyj zhur-nal. 2007. No. 12. S. 334-334.

USE OF EXPERT SYSTEMS IN EDUCATION

M.S. Chvanova, I.A. Kiseleva, A.A. Molchanov, A.N. Bozyukova Tambov State University named after G.R. Derzhavin Tambov, Russia. e-mail: [email protected]

The article considers the problems of use and development of expert systems in education, as well as actual examples of use of such systems. The authors consider it necessary to use fuzzy logic to design and develop an intelligent subsystem.

Key words: information technologies, expert system, fuzzy logic, system of education.

Topic1. EOS as a component of intensive training of specialists.

Lecture 8. Expert-training systems.

Spheres of application of expert systems in management.

Cost of expert systems.

Development of expert systems.

Over the past twenty years, experts in the field of intelligent systems have been actively researching in the field of creating and using expert systems designed for the field of education. A new class of expert systems has appeared - expert learning systems - the most promising direction for improving software pedagogical tools in the direction of procedural knowledge.

An expert system is a set of computer software that helps a person make informed decisions. Expert systems use information received in advance from experts - people who are the best specialists in any field.

Expert systems should:

• store knowledge about a particular subject area (facts, descriptions of events and patterns);
• be able to communicate with the user in limited natural language (i.e. ask questions and understand answers);
• have a set of logical tools for deriving new knowledge, identifying patterns, detecting contradictions;
• set a task on request, clarify its formulation and find a solution;
• explain to the user how the solution was obtained.

It is also desirable that the expert system could:

• communicate such information that increases the user's confidence in the expert system;
• "tell" about yourself, about your own structure

An expert learning system (ETS) is a program that implements a particular pedagogical goal based on the knowledge of an expert in a certain subject area, diagnosing learning and learning management, and also demonstrating the behavior of experts (subject specialists, methodologists, psychologists). ETS expertise lies in its knowledge of teaching methods, through which it helps teachers to teach and students to learn.

The architecture of an expert learning system includes two main components: a knowledge base (repository of knowledge units) and a software tool for accessing and processing knowledge, consisting of mechanisms for deriving conclusions (solutions), acquiring knowledge, explaining the results and an intelligent interface.

The exchange of data between the student and the EOS is performed by an intelligent interface program that perceives the student's messages and converts them into the form of a knowledge base representation and, conversely, translates the internal representation of the processing result into the student's format and outputs the message to the required media. The most important requirement for the organization of the student's dialogue with the EOS is naturalness, which does not mean literally formulating the student's needs with natural language sentences. It is important that the sequence of solving the problem is flexible, consistent with the ideas of the student and conducted in professional terms.

The presence of a developed system of explanations (SE) is extremely important for ETS working in the field of education. In the learning process, such an ETS will play not only the active role of a “teacher”, but also the role of a reference book that helps the student to study the internal processes occurring in the system using application domain modeling. The developed SS consists of two components: active, which includes a set of information messages issued to the student in the process of work, depending on the specific way of solving the problem, completely determined by the system; passive (the main component of the CO), focused on the initializing actions of the student.

The active component of the CO is a detailed commentary that accompanies the actions and results obtained by the system. The passive component of SR is a qualitatively new type of information support inherent only in knowledge-based systems. This component, in addition to the developed system of HELPs called by the trainee, has a system of explanations for the progress of solving the problem. The system of explanations in the existing EOS is implemented in various ways. It can be: a set of information about the state of the system; full or partial description of the path traversed by the system along the decision tree; a list of hypotheses to be tested (the grounds for their formation and the results of their verification); a list of goals that govern the operation of the system, and ways to achieve them.

An important feature of the developed SS is the use of a natural language of communication with the student in it. The widespread use of "menu" systems allows not only to differentiate information, but also in developed EOS to judge the level of preparedness of the student, forming his psychological portrait.

However, the trainee may not always be interested in the complete derivation of the solution, which contains many unnecessary details. In this case, the system should be able to select only key points from the chain, taking into account their importance and the level of knowledge of the student. To do this, it is necessary to maintain a model of knowledge and intentions of the student in the knowledge base. If the student continues not to understand the received answer, then the system should teach him certain fragments of knowledge in a dialogue based on the supported model of problematic knowledge, i.e. disclose individual concepts and dependencies in more detail, even if these details were not used directly in the output.

• Specialty HAC RF13.00.02
• Number of pages 192

INTRODUCTION

CHAPTER 1. COMPUTER TRAINING SYSTEMS IN

PROCESS OF EDUCATION

1.1. A brief overview of the introduction of computer technology in education.

1.2. Expert systems: their fundamental properties and applications.

1.3. The use of expert systems in the learning process. Expert-training systems.

1.4. Carrying out and analysis of the main results of the ascertaining experiment.

1.5. Prospects for the use of expert systems in the educational process.

CONCLUSIONS ON THE FIRST CHAPTER

CHAPTER 2. THEORETICAL QUESTIONS OF CONSTRUCTION

EXPERT TRAINING SYSTEMS

2.1. EOS architecture.

2.2. Representation of knowledge in EOS.

2.3. The learner model.

2.4. EOS classification. 89 CONCLUSIONS ON THE SECOND CHAPTER

CHAPTER 3. A LEARNING SYSTEM BUILT BY

THE PRINCIPLE OF OPERATION OF EXPERT-EDUCATIONAL SYSTEMS ORIENTED TO SOLVING PROBLEMS OF BODY MOVEMENT BY INCLINE

NOAH PLANE

3.1. Software tools that teach solving physical problems.

3.2. Construction and operation of a training system built on the principle of operation of expert-training systems, focused on solving problems about the movement of a body along an inclined plane.

3.3. Tasks solved with the help of the developed expert-training system.

CONCLUSIONS ON CHAPTER THREE

CHAPTER 4

4.1. Carrying out and analysis of the main results of the search experiment.

4.2. Carrying out and analysis of the main results of the teaching and control pedagogical experiment.

CONCLUSIONS ON THE FOURTH CHAPTER

Recommended list of dissertations

• Methodology for the use of expert systems to adjust the learning process and evaluate the effectiveness of teaching staff 1997, candidate of pedagogical sciences Snizhko, Elena Aleksandrovna

• Didactic computer environment as a component of technology for the formation of generalized skills of students in the implementation of experimental research 2002, candidate of pedagogical sciences Koksharov, Vladimir Leonidovich

• Computer technology for the preparation and conduct of training sessions 1999, candidate of pedagogical sciences Sedykh, Svetlana Pavlovna

• Didactic specificity of information technologies in the educational process of secondary school: Based on the material of the astronomy course 2002, Candidate of Pedagogical Sciences Rysin, Mikhail Leonidovich

• Principles of construction and use of expert learning systems in the course "Theoretical Foundations of Informatics" 2000, candidate of pedagogical sciences Kudinov, Vitaly Alekseevich

Introduction to the thesis (part of the abstract) on the topic “Computer training systems built on the principle of operation of expert training systems: Development and application in teaching the solution of physical. tasks"

Traditionally, the learning process in general and the process of teaching physics, in particular, are considered as two-way, including the activities of the teacher and students. The active use of computers in the educational process makes it a full-fledged third partner in the learning process. Computers provide practically unlimited opportunities for the development of independent creative thinking of students, their intellect, as well as independent creative activity of students and teachers.

Active work on the search for new forms and methods of teaching began in the 60s. Under the leadership of Academician A.I. Berg organized and carried out work on the problems of programmed learning, the introduction of technical teaching aids and learning machines. Programmed learning was the first step towards enhancing learning activities. Deep research on the theory and practice of programmed learning was carried out by V.P. Bespalko, G.A. Bordovsky, B.S. Gershunsky, V.A. Izvozchikov, E.I. Mashbits, D.I. Penner, A.I. Raev, V.G. Razumovsky, N.F. Talyzina and others.

The issues of effective use of computers in the educational process and research on the development of effective methods and means of computer training remain relevant at the present time. Relevant work in this area is being carried out in our country and abroad. However, a unified view on the use of computer technology in the field of education has not yet been formed.

The initial period of using computers in the learning process is characterized as a period of intensive development of the ideas of programmed learning and the development of automated learning systems. The developers of automated learning systems proceeded from the assumption that the learning process can be carried out through a well-organized sequence of frames of training and control information. The first experiments on the use of computers in the educational process were embodied in the form of educational programs with a deterministic learning scenario. This class of educational programs has the following disadvantages: a low level of adaptation to the individual characteristics of the student; reducing the task of diagnosing a student's knowledge to the task of determining whether his answers belong to one of the classes of reference answers; large labor costs for the preparation of educational material.

An alternative approach to the process of computerization of learning is the creation of so-called learning environments. The learning environment implements the concept of learning through discovery. The fundamental difference between this approach and the one discussed above is that in this case the student is treated as some kind of autonomous system capable of having its own goals. For this class of educational programs, the following features are characteristic: the learning environment provides the student with educational materials and other resources necessary to achieve the learning goal set by the teacher or by himself; lack of control of the student's actions by the system. The main purpose of the learning environment is to create a favorable, "friendly" environment or "world", "traveling" through which the student acquires knowledge.

Research in the field of the psychology of thinking, achievements in the field of artificial intelligence and programming technologies have expanded the scope of the computer in the educational process, and have made it possible to test new concepts of intellectualization of computer learning in practice.

A sharp increase in the amount of information in the educational process makes new demands on the cybernetic approach to learning, and, consequently, on pedagogical software. They should help to effectively solve the main task - managing the learning process using feedback based on a detailed diagnosis of students' knowledge, identifying the causes of their errors while simultaneously explaining the variant proposed by the computer for solving the educational problem. The noted features are most effectively implemented, first of all, by training systems built on the principle of operation of expert-training systems, which determines the relevance of the theoretical and practical study of this problem.

The introduction of expert systems into the educational process is a natural logical continuation of the computerization of education, its qualitatively new stage, laying the foundations for informatization of education. This process became possible thanks to the deep researches carried out on the issues of computerization of education by scientists and teachers. Considering that the use of expert systems for solving problems in physics has yielded positive results, research on the development and application of expert systems is relevant not only in scientific but also in pedagogical activities, including teaching physics.

The use of training programs built on the principle of expert training systems in the learning process will give a new qualitative leap in education. Their introduction into the practice of teaching will allow: to change the style of teaching, turning it from informational and explanatory into cognitive, educational and research; reduce the time required to acquire the necessary knowledge.

The object of the research is the process of teaching physics.

The subject of the research is the process of learning to solve problems in physics using a learning system built on the principle of operation of expert learning systems, and the formation of a common method for solving problems among students.

The purpose of the work was to develop and create a learning system built on the principle of expert learning systems, focused on solving physical problems of a certain class, and to study the possibility of forming a common method for solving students when teaching solving problems in physics using data from specially developed pedagogical software tools. .

The hypothesis of the study is as follows: the introduction of learning systems into the learning process, built on the principle of expert learning systems, will lead to a more effective assimilation by students of the general way of solving problems in physics, which will increase their academic performance, deepen their knowledge of physics and will contribute to an increase in the quality of knowledge in the subject being studied.

Based on the formulated hypothesis, in order to achieve the goal of the study, the following tasks were set and solved:

Analysis of modern methods and means of developing educational programs. Focusing on those that correspond to the goals of the work;

Researching the possibilities of using a computer to implement the formation of a common way for students to solve problems;

Development of the structure and principles of building a training system built on the principle of operation of expert training systems, focused on solving physical problems of a certain class;

Testing the proposed research hypothesis, evaluating the effectiveness of the developed methodology, developed pedagogical software during the pedagogical experiment.

To solve the tasks, the following research methods were used:

Theoretical analysis of the problem based on the study of pedagogical, methodological and psychological literature;

Questioning and questioning of pupils, students, teachers of schools and universities;

Studying the process of learning to solve problems and the developed methodology in the course of attending and conducting classes in physics, observing students, talking with teachers, conducting and analyzing tests, testing students;

Planning, preparing, conducting a pedagogical experiment and analyzing its results.

The scientific novelty of the research lies in:

Development of a training system built on the principle of operation of expert training systems, focused on solving a certain class of problems in physics;

Theoretical and practical substantiation of the possibility of forming a common way for students to solve problems when using the developed pedagogical software in the learning process (a learning system built on the principle of expert learning systems);

Development of the fundamentals of the methodology for using a training system, built on the principle of operation of expert-training systems, in teaching the solution of physical problems.

The theoretical significance of the study lies in the development of an approach to teaching solving problems in physics, which consists in the implementation of control over the activities of students in solving problems through specially developed pedagogical software tools (a learning system built on the principle of operation of expert learning systems).

The practical significance of the study lies in the creation of software and methodological support for classes in physics (a teaching system built on the principle of operation of expert teaching systems), determining its role and place in the educational process and developing the foundations for the methodology for using these pedagogical software tools when conducting classes on solving physical problems. tasks using computers.

The following is submitted for defense:

Substantiation of the possibility of using the developed training system, built on the principle of operation of expert-training systems, in the process of teaching solving problems in physics;

Development of an approach to managing the activities of students through specially developed pedagogical software tools (a teaching system built on the principle of operation of expert teaching systems) when teaching solving problems in physics;

Fundamentals of the methodology for using a training system built on the principle of expert training systems in conducting classes on solving problems in the process of teaching physics.

Testing and implementation of research results. The main results of the study were reported, discussed and approved at the meetings of the Department of Methods of Teaching Physics of the Moscow State Pedagogical University (1994-1997), at the conference of young scientists (Mordovia State University, 1996-1997), at the conferences of the Moscow State Pedagogical University (April, 1996).

The main provisions of the dissertation are reflected in the following publications:

1. Gryzlov S.V. Expert-training systems (literature review) // Teaching physics in higher education. M., 1996. No. 4. - S. 3-12.

2. Gryzlov S.V. The use of expert-learning systems in the process of teaching physics // Teaching Physics in Higher School. M., 1996. No. 5.-S. 21-23.

3. S. V. Gryzlov, A. P. Korolev, and D. Yu. Expert-training system focused on solving a complex of tasks about the motion of a body along an inclined plane // Improvement of the educational process based on new information technologies. Saransk: Mordovian state. ped. in-t, 1996. - S. 45-47.

4. Gryzlov S.V., Kamenetsky S.E. Perspective directions of using computer technology in the educational process of the university and school // Science and school. 1997. No. 2.-S. 35-36.

The structure and scope of the dissertation. The dissertation work consists of an introduction, four chapters, a conclusion, a list of references and an appendix. The total volume is 192 pages of typewritten text, including 25 figures, 8 tables. The list of references includes 125 titles.

Similar theses in the specialty "Theory and Methods of Training and Education (by Regions and Levels of Education)", 13.00.02 VAK code

• Didactic conditions for the use of automated training courses in the process of studying natural science disciplines by high school students 1999, candidate of pedagogical sciences Belous, Natalya Nikolaevna

• Development of object-oriented mathematical and software information technologies for managing individualized learning in a correctional school 2003, Ph.D. Kremer, Olga Borisovna

• Theoretical foundations for the creation and application of didactic interactive software systems in general technical disciplines 1999, Doctor of Pedagogical Sciences Zainutdinova, Larisa Khasanovna

• Methods of teaching geometry in grades 10-11 of a secondary school using a computer 2002, Doctor of Pedagogical Sciences Mehdiyev, Muradkhan Hajikhanovich

• Computerized pedagogical support of the student's actions when working on a branched program 2002, candidate of pedagogical sciences Tsareva, Irina Nikolaevna

Dissertation conclusion on the topic "Theory and methods of training and education (by areas and levels of education)", Gryzlov, Sergey Viktorovich

CONCLUSIONS ON THE FOURTH CHAPTER

1. Based on the analysis of possible directions for using a computer in education, the shortcomings of existing pedagogical software tools are identified, the need to create and use teaching software in the educational process, built on the principle of expert training systems, is substantiated.

2. A methodology has been developed for conducting classes using the developed software tools (a training system built on the principle of operation of expert training systems).

3. During the search experiment, the content was determined and the structure of the developed pedagogical software was adjusted.

4. Conducting a search experiment made it possible to develop the final version of the methodology for conducting classes using the developed training system, aimed at developing a common way for students to solve problems.

5. The conducted comparative analysis of the results of the control pedagogical experiment testifies to the significant influence of the proposed methodology for conducting classes on solving physical problems using the developed pedagogical software on the formation of a common way of solving problems among students.

Thus, the validity of the hypothesis put forward about the greater effectiveness of the proposed methodology for conducting classes on solving physical problems using the developed pedagogical software compared to the traditional one has been proved.

CONCLUSION

1. Studied and analyzed pedagogical, methodological and psychological literature and dissertation research on the methodology of using a computer in the learning process. On this basis, it was revealed that the most effective pedagogical software tools are educational programs built on the principle of operation of expert-training systems.

2. Expert-training systems, focused on the formation of a common way of solving students, are the most effective means of teaching problem solving.

3. The prospects for the use of expert-training systems in the educational process are determined, directions for the use of expert systems in the learning process are proposed.

4. The structure of the training system, built on the principle of operation of expert-training systems, focused on the formation of a common way of solving problems among students, is proposed and justified.

5. A training system has been developed, built on the principle of operation of expert-training systems, focused on solving a set of problems about the movement of a body along an inclined plane. Management of students' activities in the course of solving a problem with the help of the developed training system is implemented through: a) computer simulation, which makes it possible to identify the essential properties and relationships of the objects referred to in the problem; b) heuristic tools that provide students with the opportunity to plan their actions; c) step-by-step control of the student's actions by the teaching system and presentation, at the request of the student, of a reference solution to the problem, developing the ability to evaluate one's actions, choose the criteria for this assessment.

6. The methodology for conducting classes on solving problems using the developed pedagogical software, their role and place in the educational process is determined. The main provisions of this technique are as follows: a) students' independent choice of tasks for mastering the general way of solving problems of a certain class; b) the use of the developed pedagogical software tools (a learning system built on the principle of expert learning systems) to form a common method for solving problems; c) a combination of independent problem solving by each student with a collective discussion of the solution plan; d) selection of an algorithm for solving problems of this class based on the generalization of already solved problems.

7. The results of the pedagogical experiment showed that the formation of a common way of solving problems among students in the experimental groups, where training was carried out using the developed pedagogical software tools (a training system built on the principle of expert training systems), is much higher than in control groups , where training was carried out using the most common types of computer programs (simulating and teaching), which confirms the reliability of the hypothesis put forward.

List of references for dissertation research Candidate of Pedagogical Sciences Gryzlov, Sergey Viktorovich, 1998

1. Alekseeva E.F., Stefanyuk V.L. Expert systems (state and perspective) // Izvestiya AN SSSR. Technical cybernetics. 1984.- No. 5. pp. 153-167.

2. Anatsky N.M., Levin N.A., Pospelova L.Ya. Implementation of the expert system "IPILOG" / Materials of the V All-Union Seminar "Development and application of PC software in the educational process": Abstracts. report Ordzhonikidze, 1989. - S. 27-28.

3. Anderson J.R., Reiser B.J. LISP teacher // In the book. Reality and forecasts of artificial intelligence: Sat. articles; per. from English. / Ed. V.L. Stefanyuk. M.: Mir, 1987. - S. 27-47.

4. Antonyuk L.S., Cherepina I.S. On the use of active teaching methods in junior courses // Programmed learning, 1988. -Vol. 25.-S. 98-101.

5. Aristova L.P. Automation of the teaching of schoolchildren. M.: Enlightenment, 1968. -139 p.

6. Babansky Yu.K. The choice of teaching methods in secondary school. M.: Pedagogy, 1981. - 176 p.

7. Baikov F.Ya. Problem-programmed tasks in physics in high school. A guide for teachers. M.: Enlightenment, 1982. - 62 p.

8. Balobashko N.G., Kuznetsov V.C., Smirnov O.A. Providing the educational process with computing resources. M .: Research Institute of Higher Problems. school - 1985. 44 p.

9. Bespalko V.P. Fundamentals of the theory of pedagogical systems. Voronezh: Publishing House of the Voronezh University, 1977. - 304.

10. Bespalko V.P. Programmed learning (didactic foundations). M., 1970. - 300 p.

11. Bobko I.M. Adaptive pedagogical software. -Novosibirsk: NSU publishing house, 1991. 101 p.

12. Bugaenko G.A., Burkova S.A. Solving one problem of increased difficulty // Physics at school. No. 4. - 1991. - S. 43-46.

13. Bunyaev M.M. Scientific and methodological foundations for the design of branched interactive learning systems: Dis. for the degree of Cand. ped. Sciences. 1992. - 350 p.

14. Vlasova E.Z. Prospects for the use of expert systems in the educational process // Secondary special education. 1991. - No. 4. - S. 21.

15. Vlasova E.Z. Development of knowledge bases of expert systems in the methodological training of physics students: Dis. for the degree of Cand. ped. Sciences. St. Petersburg, 1993. - 211 p.

16. Gvaramiya M. Experience in the development of computer teaching aids in physics. Informatics and education. 1990. - No. 6. - S. 79.

17. Gergey T., Mashbits E.I. Psychological and pedagogical problems of the effective use of computers in the educational process. Voprosy psikhologii. 1985. - No. 3. - S. 41-49.

18. Gershunsky B.S. Computerization in education: problems and prospects. M.: Pedagogy, 1987. - 264 p.

19. Glushkov V.M. Computing technology and control activation problems. In: The Future of Science. Perspectives. Hypotheses. Modern problems. Issue. 4. - M.: Knowledge, 1971.

20. Golitsina I., Narkov I. Computer at physics lessons // Informatics and education. 1990. - No. 3. - S. 31.

21. Gottlieb B. Computer and didactic support // Informatics and education. 1987. - No. 4. - S. 3-14.

22. Gottlieb B. The structure of AOS // Informatics and education. 1987. - No. Z.-S. 11-19.

23. Grabar M.I., Krasnyanskaya K.A. Application of mathematical statistics in pedagogical research. Nonparametric methods. -M., Pedagogy, 1977. 136 p.

24. Gryzlov S.V. Expert-training systems (literature review) // In Sat. Teaching physics in higher education. No. 4. - M., 1996. - S. 312.

25. Gutman V.I., Moshchansky V.N. Algorithms for solving problems in mechanics in high school: A book for teachers. M.: Enlightenment, 1988. -95 p.

26. Davydov V.V. The problem of developmental education: The experience of theoretical and experimental psychological research. M.: Pedagogy, 1986. - 240 p.

27. Dalinger V. Interactive training programs and requirements for them // Informatics and education. 1988. - No. 6. - S. 35-37.

28. Danovski P., Dovgyallo A.M., Kirova K.N. et al. Automated learning systems based on SPOK // Modern Higher School.-1983.-№ 1.-S. 171-178.

29. Denisov A.E., Bushuev S.D. Programmed learning and computerization of the educational process at the university // Programmed learning, 1988.-Vol. 25.-S. 3-9.

30. Didactics of high school: Some problems of modern didactics. / Ed. M.N. Skatkin. M.: Enlightenment, 1982. - 319 p.

31. Driga V.I., Pankov M.N. On the issue of didactic requirements for the compilation of software and pedagogical tools / In Sat. Computer and Education / Ed. Razumovsky V.G. M.: APN USSR, 1991 -117 p.

32. Emelyanov V.V., Ukhanova T.V., Yasinovskii S.I. Using artificial intelligence methods in flexible production systems: A textbook for the course "Organizational management of the GPS" / Ed. V.V. Emelyanov. M.: Publishing House of MSTU, 1991. - 36 p.

33. Eslyamov S.G. Methods and tools that ensure the effective use of expert systems in education: Abstract of the dissertation for the degree of candidate of technical sciences: 05.25.05. Kyiv, 1993.- 16 p.

34. Jablon K., Simon J.-C. The use of computers for numerical simulation in physics. M.: Nauka, 1983. - 235 p.

35. Zak A.Z. How to determine the level of development of thinking of a student. -M.: Knowledge, 1982. 98 p.

36. Ibragimov O.V., Petrushin V.A. Expert-training systems. - Kiev, 1989. 21 p. - (Prepr. / Academy of Sciences of the Ukrainian SSR. Institute of Cybernetics named after V.M. Glushkov; 89-47).

37. V.A. Didactic bases of computer training in physics. L.: LGPI, 1987. - 256 p.

38. Cab drivers V.A., Zharkov I.V. Dialogue between student and machine // Physics at school. 1985. - No. 5. - S. 48-51.

39. Cab drivers V.A., Revunov D.A. EVT at physics lessons in high school. Moscow: Enlightenment, 1988. - 239 p.

40. Ilyina T.A. Pedagogy: Course of lectures. Textbook for students ped. universities. M.: Enlightenment, 1984. - 202 p.

41. Cybernetics and learning problems. / Ed. A.I. Berg. M.: Progress, 1970. - 390 p.

42. The computer gains intelligence: TRANS. from English / Ed. B.J.I. Stefanyuk. -M.: Mir, 1990. 240 p.

43. Kondratiev A.S., Laptev V.V. Physics and computer. L .: Publishing House of Leningrad State University, 1989. - 328 p.

44. Konstantinov A.B. Computer as a Theorist: Symbolic Calculations and Principles of Artificial Intelligence in Theoretical Physics / Experiment on Display. M.: Nauka, 1989. - S. 6-44.

45. Korzh E.D., Penner D.I. Programmed tasks in physics for grade VIII. Vladimir: V PI, 1984. - 81 p.

46. ​​Krug G.K., Kabanov V.A., Chernykh A.V. Instrumental interactive learning systems on microcomputers // Microprocessor devices and systems. 1987. - No. 3. - S. 29-30.

47. Kuznetsov A., Sergeeva T. Educational programs and didactics // Informatics and education. 1986. - No. 2. - S. 87-90.

48. Kuznetsov A. Basic principles of using computers in the learning process. / Sat. Theoretical and applied problems of computerization of education. Kazan, 1988. - 184 p.

49. Lanina I.Ya. Formation of cognitive interests of students in physics lessons. M.: Enlightenment, 1985. - 128 p.

50. Lobanov Yu.I., Brusilovsky P.L., Sedin V.V. Expert learning systems. - M., - 56 p. - (New information technologies in education: review, inform. / NIIVO; Issue 2)

51. Lyaudis V.Ya. Psychological principles of constructing dialogue teaching systems // In the collection of articles. Psychological-pedagogical and psychological-physiological problems of computer training. M.: Publishing House of the Academy of Sciences of the USSR. - 1985.- 162 p.

52. Marcellus D. Programming expert systems on Turbo Prolog: Per. from English. M.: Finance and statistics, 1994. - 256 p.

53. Maryasina E.D. Analysis of the correctness of answers in automated learning systems using interpretive models // Control systems and machines. 1983. - No. 1. - S. 104-107.

54. Maslov A., Tairov O., Trush V. Physiological and hygienic aspects of the use of personal computers in the educational process. Informatika i obrazovanie. 1987. - No. 4. - S. 79-81.

55. Mashbits E.I. Dialogue in the learning machine. Kyiv: Vishcha school, 1989. -182 p.

56. Mashbits E.I. Computerization of education: problems and prospects. M.: Knowledge, 1986. - 80 p.

57. Mashbits E.I. Psychological and pedagogical problems of computerization of education. M.: Pedagogy, 1988. - 215 p.

58. Methods of studying the topic "Electric field" in the course of physics of a secondary school based on problematic programmed tasks:

61. Mitrofanov G.Yu. Expert systems in the learning process. M.: TSNTI of civil aviation, 1989. - 32 p.

62. Mikhalevich V.M., Dovgyallo A.M., Saveliev Ya.M., Kogdov N.M. Expert-training systems in the complex of computer learning tools // Modern Higher School. 1988. - No. 1 (61). - S. 125-136.

63. Monakhov V.M. Psychological and pedagogical problems of ensuring computer literacy of students // Questions of psychology. 1985.- No. 3. S. 14-22.

64. Morozova N.V., Ionkin V.P. Using frame systems to control students' knowledge // In the book. Methods and means of informatization of education and research / Mosk. eq.-st. in-t. M., 1992.- S. 43-49.

65. Nevdava L., Sergeeva T. On promising trends in the development of pedagogical software // Informatics and education.- 1990.-№6.-S. 79.

66. Nikolov B.C. Development of tools for creating educational expert systems: Dis. for the degree of Cand. Phys.-Math. Sciences. M., USSR Academy of Sciences, 1988. - 183 p.

67. Nilson N. Principles of artificial intelligence / Per. from English. -M.: Radio and communication, 1985. 373 p.

68. Novikov V.N. On one task of increased difficulty // Physics at school. No. 5. - 1989. - S. 124-128.

69. Novitsky L.P., Feidberg L.M. Expert-training system for a personal computer // In the book: Methods and means of cybernetics in the management of the educational process of higher education: Sat. scientific tr. / Moscow. ex-st. in-t. M.; 1992. - S. 43-49.

70. Pedagogy of the school. / Ed. I.T. Ogorodnikov. M.: Enlightenment, 1978.-320 p.

71. Prospects for the development of computer technology: In 11 books: Reference, allowance / Ed. Yu.M. Smirnova. Book. 2. Intellectualization of computers / E.S. Kuzin, A.I. Roitman, I.B. Fominykh, G.K. Khakhalin. M.: Higher. school, 1989. - 159 p.

72. Petrushin V.A. Architecture of expert learning systems / In the book. Development and application of expert-training systems: Sat. scientific tr. M.: NIIVSH, - 1989. - S. 7-18.

73. Petrushin V.A. Intelligent learning systems: architecture and implementation methods (review) // Izvestiya AN. Technical cybernetics, No. 2 1993. - S. 164-189.

74. Petrushin V.A. Modeling the state of knowledge of a trainee in intelligent learning systems // In the book. Development of computer learning technologies and their implementation: Sat. scientific tr. / AN Ukrainian SSR. Institute of Cybernetics im. Glushkova, Kyiv, 1991. - S. 26-31.

75. Povyakel N.I. Goal formation in the psychological support of computer user software. M.: Publishing House of Moscow State University, 1975. -S. 79-81.

76. Popov E.V. Communication with computers in natural language. M.: Nauka.-1982. - 360 p.

77. Popov E.V. Expert systems: Solving non-formalized problems in dialogue with a computer. M.: Science. Ch. ed. Phys.-Math. lit., 1987. - 288 p.

78. Building expert systems. Ed. F. Hayes-Roth M.: Mir, 1987.-442 p.

79. Workshop on the development of pedagogical software for secondary schools. / Uch. allowance ed. V.D. Stepanova. M.: publishing house Prometheus, 1990. - 79 p.

80. Representation and use of knowledge: Per. from Japanese. / Ed. X. Ueno, M. Ishizuka. M.: Mir, 1989.

81. The use of expert systems in teaching physics: Methodological recommendations. / Comp. E.Z. Vlasova, prof., Dr. Ph.-M. Sciences V.A. Cabbers. St. Petersburg, 1992. - 50 p. - (Cybernetics. Pedagogy. Educology. / Russian Pedagogical University named after A.I. Herzen. From "Education").

82. Putieva A. Issues of developing education using computers // Questions of psychology. 1987. - No. 1. - S. 63-65.

83. Raev A.I. Psychological issues of programmed learning. Leningrad: LGPI im. Herzen, 1971. - 96 p.

84. Development and application of expert learning systems. // Sat. scientific tr. M.: NIIVSH, 1989. - 154 p.

85. Revunov A.D., Izvozchikov V.A. Electronic computers at physics lessons in high school. Moscow: Enlightenment, 1988. - 257 p.

86. Richmond W.K. Teachers and machines: (Introduction to the theory and practice of programmed learning). M., 1968. - 278 p.

87. Savchenko N.E. Mistakes in the entrance exams in physics. -Minsk, Higher. school, 1975. - 160 p.

88. Sergeeva T. New information technologies and the content of education // Informatics and education. -1991. No. 1.

89. Sergeeva T., Chernyavskaya A. Didactic requirements for computer training programs // Informatics and education. -1986. -No. 1.-S. 48-52.

90. Talyzina N.F. Theoretical problems of programmed learning. M.: Publishing House of Moscow State University, 1969. - 133 p.

91. Talyzina N.F. Management of the learning process. M.: Publishing House of Moscow State University, 1975.-343 p.

92. Tarasov JI.B., Tarasova A.N. Questions and tasks in physics (Analysis of characteristic errors entering technical universities). Textbook allowance, 3rd ed., revised. and additional - M.: Higher. school, 1984. - 256 p.

93. Tikhomirov O.K. Psychological structure of the dialogue "Man-computer" // Bulletin of Moscow State University. Ser. 14. Psychology. - 1984. - No. 2. - S. 1724.

94. Usova A.V., Bobrov A.A. Formation of educational skills and abilities of students in physics lessons. M.: Enlightenment, 1988. - 112 p. (Library of the teacher of physics).

95. Usova A.V., Tulkibaeva N.N. Workshop on solving physical problems: Textbook. allowance for students of physics and mathematics. fak. M.: Enlightenment, 1992. - 208 p.

96. Fedoseenko M.Yu. The choice of means of knowledge representation in expert-training systems // In the book: Development and application of expert-training systems: Sat. scientific tr. M.: NIIVSH, 1989. - S. 43-48.

97. Chekulaeva M.E. The use of computers as a means of developing the thinking of students in teaching physics: Abstract of the dissertation for the degree of candidate of pedagogical sciences: 13.00.02. -M., 1995.- 17 p.

98. Man and computer technology / Ed. V.M. Glushkov. Kyiv, Naukova Dumka, 1971.

99. Man and computer technology. / Under the total. ed. V.M. Glushkov. Kyiv, 1971.-294 p.

100. Schukina G.I. Activation of cognitive activity of students in the educational process. M.: Enlightenment, 1979. - 160 p.

101. Aiken K. Teachers and Computer. What is the key component? // Paper Presented at ABS (Automatization of the Educational System) in Secondary and High Schools. Institute Kurchatova. M., 1989, May 26. - P. 37-41.

102. Anderson J.A. Psychology and intelligent tutoring / Artif. Intel. and Educ.: Proc. 4th Int. Conf. AI and Educ., Amsterdam, 24-26 May, 1989. -Amsterdam etc., 1989. P. 1.

103. Andriole S.J. The Promise of artificial intelligence // J. Syst. Manag. -1985.-Vol. 36.-№7.-P. 8-17.

104. Bodnar Gy. A mesterseges intelligencia es a szakerforendzerek // Minosed es Megbizhatosag, 1988. No. 3. - P. 11-17.

105. Bork A. Learning with Personal Computers. Cambridge: Harper and Row, 1987. - 238 p.

106. Brown I.S., Burton R.R. Diagnostic Models for Procedural Bugs in Basic Mathematical Skills // Cognitive Science. 1978. - V. 2. - P. 155192.

107 Burton R.R. Diagnosing bugs in simple procedural skills // Intern. J. Man-Machine Studies. 1979. - No. 11.

108. Cumming G., Self J. Collaborative intelligent educational systems / Artif. Intel. and Educ.: Proc. 4th Int. Conf. AI and Educ., Amsterdam, 2426 May, 1989. Amsterdam etc., 1989. - P. 73-80.

109. Dutta A. Reasoning with imprecise knowledge in expert system // Int. sci. (USA). 1985. - Vol. 37. - No. 1-3. - P. 3-24.

110. Elson-Cook M. Guided discovery tutoring and bounded user modeling // Self J. (Ed.) Artifical intelligence and human learning. Intelligent computer-aided instruction. L.: Chapman and Hall, 1988.

111. Feigenbaum E. On generality and problem solving // Machine Intelligence. 1971. - No. 6.

112. Feigenbaum E.A., Mecorduck P. The 5th generation. Addison Wesley. Mass. 1983.-226 p.

113. Goldstein I.P. The Genetic Graph: a representation for the evolution of procedural knowledge // Intern. J. Man-Machine Studies. 1979. -№11.

114. Murray W.R. Control for intelligent tutoring systems: a blackboard-based dynamic instructional planner / Artif. Intel. and Educ.: Proc. 4th Int. Conf. AI and Educ., Amsterdam, 24-26 May, 1989. Amsterdam etc., 1989.-P. 150-168.

115. Newell A. Heuristic programming: illstructured problems // Progress in operation processing. New York: Wiley and Sons, 1969. - V. 3. - P. 362414.

116. Simon H. The structure of illstructured problems // Artificial Intelligence. 1974. - V. 5. - No. 2. - P. 115-135.

117. Sleeman D. Some challenges for intelligent tutoring systems / IJCAI 87: Proc. 10th Joint Conf. Artif. Intel., Milan, Aug. 23-28, 1987. P. 11661168.

118. Sleeman D. Assessing aspects of competence in basic algebra // Sleeman D., Brown J.S. (eds) Intelligent Tutoring Systems. New York: Academic Press, 1982.

119. Souldin Y. Optimum teaching system Illusion or reality? /East-West: Intern. conference "Human-Computer Interaction", Moscow, August 3-7, 1993: Dokl. T. 1. - M., 1993. - S. 59-72.

120. Tompsett C.R. Education, training and knowledge base design // Expert syst. 1988. - V. 5. - No. 4. - P. 274-280.

121. Weip S. The Computer in School: Machine as Humanizer // Symposium: Harvard Educational Review, 1989. Vol. 59. - No. 1. - P. 61.

122. Yazadani M. Guest editorial: expert tutoring systems // Expert Syst. -1988. V. 5. - No. 4. - P. 271-272.

Please note that the scientific texts presented above are posted for review and obtained through recognition of the original texts of dissertations (OCR). In this connection, they may contain errors related to the imperfection of recognition algorithms. There are no such errors in the PDF files of dissertations and abstracts that we deliver.

Expert teaching system

Introduction

Currently, in connection with the rapid development of Internet technologies, there are more and more interactive services for Internet and Intranet networks, such as distance learning. The distance learning system is a fairly popular form of education in the world in those countries that have a fairly high level of development of communication tools based on computer technology. The training of modern specialists requires the organization of the educational process using these new information technologies and using knowledge-based systems - expert systems (ES).

The use of ES to assess the level of knowledge of students in testing systems determines an important block of computer programs - expert-training systems (ETS).

Expert-training systems are computer programs that have the main components of ES, but which have an additionally expanded explanation component. Such systems are based both on the knowledge of software experts and on the knowledge of experts in teaching methods. In addition, they have a component of adapting the presentation of educational material to the student, depending on his preparedness. And at least there are several learning strategies, the level of detail of which depends on the activity of the student in the dialogue with the system.

The use of EOS as a testing tool to determine the quality of a student's knowledge is also of great importance in teaching. Since in such testing the student is not affected by the subjective factor, that is, the test results do not depend on the personal characteristics of the examiner and the testee. And the use of unified tests allows the teacher to objectively assess the level of preparation of students.

1. Relevance of the topic

In connection with the widespread use of computers, the role of computer training is increasing, the methodology of which increases the intellectual abilities of the student and the independence of decision-making. And such qualities are most in demand in a competitive economy and contribute to educational andprofessional growth. There are problems of creating effective learning systems, as well as the creation of new forms and ways of presenting educational material, the search for new pedagogical techniques and teaching aids. One of the ways to increase the effectiveness of training, assimilation of information and reduce the cost of the learning process itself is the development and use of automated expert training systems. At present, there are many terms denoting an automated expert learning system, which, in fact, are similar.

The most popular of them are distance learning systems, computer training system and others. To explain the full meaning of the above terms, the following definition can be given.
An expert learning system (ETS) is a complex of software and hardware and educational and methodological tools built on the basis of the knowledge of experts in the subject area (qualified teachers, methodologists, psychologists), which implements and controlslearning process. The purpose of such a system is that, on the one hand, it helps the teacher to teach and control the student, and on the other hand, the student learns independently.

2. Purpose and objectives of the study, planned results

The aim of the study is to develop a computer expert teaching system that will help increase the amount of acquired knowledge and the efficiency of information perception, as well as reduce the time for studying the subject, including the time spent by the teacher on presenting information and instilling practical skills in students.

The main objectives of the study:

1. Development of ontological model of EOS;
2. EOS structure development;
3. Justification and choice of computer means of implementation;
4. Implementation of active components in the EOS (games, interactive systems, direct access to communication, for example, via Skype with the manager);

Object of study: expert teaching system.

Subject of study: models, structures and functions of EOS.

Scientific novelty consists of a new approach to the design of ETS, based on the modeling of the student's activity and the use of artificial intelligence methods.

As part of the master's work, it is planned to obtain relevant scientific results in the following areas:

1. Modeling of learning processes.
2. EOS structure design for Internet and Intranet.

Planned results of the work: a prototype of an expert training system that will improve the quality of training and reduce training time.

3. Survey of scientific researches.

Since the issues of researching expert learning systems and improving the effectiveness of learning in this system are an important part of solving complex problems with the help of expert systems. EOS have been widely studied by both foreign and domestic specialists.

3.1. Review of international sources

First teaching system Plato based on a powerful computer company " Control Data Corporation ” was developed in the USA in the late 50s and has been developed for 20 years. The creation and use of training programs have become truly massive since the early 80s, when personal computers appeared and became widespread. Since then, the educational applications of computers have moved into the mainstream along with word processing and graphics, pushing mathematical calculations into the background.

ECSI was also founded in 1972 and has since established itself as a leading provider of services to the education industry. The company specializes in developing products and services to enhance the learning experience for students and their parents. ECSI currently serves over 1,300 schools, colleges and universities across the country, offering a wide range of fully customized, intuitive learning systems.

3.2. Review of national sources

Modern training systems include TrainingWare, eLearning Server 3000 v2.0, eLearningOffice 3000, IBM Workplace Collaborative Learning and HyperMethod 3.5 systems from HyperMethod, which is the largest Russian developer of ready-made solutions and software in the field of multimedia, expert training and e-commerce.

4. Expert learning systems

An expert learning system (ETS) is a computer program built on the basis of the knowledge of subject matter experts (qualified teachers, methodologists, psychologists) that implements and controls the learning process. The purpose of such a system is that, on the one hand, it helps the teacher to teach and control the student, and on the other hand, the student learns independently.

The main components of the EOS are:

1. knowledge base;
2. output machine;
3. knowledge extraction module;
4. learning module;
5. explanation system;
6. testing module.

Picture 1- Functional model of the EOS structure

(animation: 8 frames, 5 loops, 118 kilobytes)

In this model, the upper part of the ETS is inherited from the ES, and the lower part is the blocks that provide the learning and testing process.

The knowledge base is a repository of knowledge modules. The knowledge module of expert systems is a formalized, using some method of knowledge representation (production system, frames, semantic networks, 1st order predicate calculus), mapping of objects of the subject area, their relationships, actions on objects.

Working with the knowledge base involves the following stages:

1. extracting knowledge from experts;
2. formalization of knowledge;
3. access, processing of knowledge modules.

In the process of learning, expert knowledge can be transferred to the student in the form of a portion of information (textual, graphic, multimedia), as well as knowledge based on experience that cannot be transferred directly to the student, but acquired by him in the course of independent activity].

To transfer knowledge of experts, advanced hypertext technology is widely used - from traditional programs for creating help (help) to modern tools for creating and maintaining Web sites (for example, Dreamweaver MX).

Unlike ES, to build a knowledge base, EES involves not only expert teachers, but also uses knowledge about pedagogical techniques and learning strategies and about the psychological characteristics of a person. Therefore, knowledge modules are formed by many experts. And here it is necessary to take into account the consistency of the opinions of experts and fine-tune the knowledge base, taking into account the competence of experts. Of course, these difficulties can be circumvented if there is an expert who combines the knowledge of a specialist in the subject area, knowledge of the tactics and strategies of teaching and who owns the psychological methods of teaching, that is, a highly qualified teacher.

The learning component is a set of software modules that implement various inference mechanisms to achieve the pedagogical goal in learning. ETS, unlike other computer learning tools, have interactivity: they have a dialogue with the student, which is very attractive for the latter.

Building a dialogue is based on the basic psychological principles of learning:

1. friendly interface;
2. exit the dialogue at any time;
3. timely and motivated help.

Each question asked to the trainee must be carefully considered, if necessary, provide for a more detailed question in order to better understand it.

As a result of the study it was shown that many components of the creation of an ETS depend on the result of training, therefore, to create a knowledge base of an ES, a specialist is needed who has excellent knowledge of the subject area, and is also confident in learning techniques.

5. Client-server technology of an expert learning system for networks InternetandIntranet

The client-server architecture consists of the following components:

a server that fulfills client requests; a client that provides a user interface that sends requests to the server and receives responses from it; network communication software that interacts between a client and a server. The use of client-server technology gives certain advantages when building an ES: the knowledge base is stored on the server and, therefore, it needs to be updated once;
the knowledge base can be accessed by other applications; and the advantage for expert learning systems (ETS) is that you can store content on the server and track training statistics on it.
Client-server ES and EOS for Internet/Intranet networks allow expanding the possibilities of their application in distance education.
Computer training systems allow both the development of ES prototypes and can be used for adapted testing and teaching students over a local network.
The main components of the EOS are the following: knowledge base editor; logical inference machines (direct, reverse, indirect inference, Bayes formula); explanation subsystem; test analyzer; teacher module; learning component.

The main task of expert learning systems is to provide the student with the opportunity to acquire knowledge, skills in developing knowledge base and creating ES prototypes independently, as well as for trained testing.

There are at least five important reasons that prevent the implementation of client-server (distributed) ES:

1. The structural elements of the ES components are not isolated from each other.
2. A KB is not a database for which there are powerful DBMSs (Oracle, InterBase, MySQL, and so on) that use SQL queries.
3. Multi-user access to the KB for editing is simply not acceptable.
4. The logical conclusion and the specifics of creating a knowledge base (different ways of representing knowledge) do not contribute to the need to combine them into a single system. A number of description languages, Web services have been developed for the Symantec Web, but so far there are no proposals for the implementation of inference.
5. Software tools for building ES and KB is exclusive and expensive.

You can, of course, place the ES on a Web server to download to a client machine using the download link and update it on the server, but this is not a client-server solution.

Similarly, one can argue about the use of a three-tier client-server architecture (Server - CORBA - Client), when the knowledge base is hosted on the application server and presented in the form of business decision rules.

Also, the technology of "thin client" (KB, logical inference, explanation system is located on the server, and the dialogue with the ES is supported both on the server and on the client) and "thick client" (KB, logical inference, explanation system are located on the client) are also not suitable. machine, and the conversational interface is maintained by the client and server).

Note that the KB ES is an intellectual property and cannot be made available for free use. And training knowledge bases should be placed on a Web server so that any user of interest can analyze how the ES works and improve their knowledge of the subject area.

Do not forget about server loads in peak situations. No provider will give the server only for the functioning of the ES, since the user's reaction during consultation or explanation is not predictable. And these are important points in the functioning of the ES (consultations can last from minutes to several hours).

The development of EOS for Internet/Intranet networks is quite another matter.

EOS is a computer system built on the basis of the knowledge of subject matter experts (qualified teachers, methodologists, psychologists), which implements and controls the learning process. The purpose of such a system is that, on the one hand, it helps the teacher to teach and control students, and on the other hand, students learn on their own.

The main components of the EOS are the following: KB; output machine; learning module; explanation system; learning testing module.

As a rule, the BR contains:

Psychodiagnostic rules for identifying the psychological types of trainees.

Didactic techniques for learning. The rules represent the accumulated knowledge of teachers to assess the knowledge of students.

The learning rules change the sequence of the content tasks presented. This sequence is a function of many variables: the psychological type of the learner, the level of learning, the current answer of the learner, the level of difficulty of the task, the amount of training.

In connection with the above regarding distributed ES, it is recommended to use the "thick client" technology for training and testing, that is, when all the components of the ETS are located on the client machine, and the results of training and testing are transferred to the server. And there is no need to be afraid that the results can be replaced, given the modern possibilities of encrypting the protocol with a remote server. Why this particular technology? It is known that about 80% of all information perceived by a person - it's visual. Therefore, multimedia technologies (avi-files) are a priority in teaching. If they are located and run onserver - this is a huge load on the server and, as a result, traffic increases to a huge size.

conclusions

ETS, unlike other computer learning technologies, have the ability to implement the learning process according to the individual model of the student. Learning with the help of ES is focused on the extraction of knowledge by the trainees themselves. Namely, such specialists are in demand in the modern labor market. EOS also has its advantages and disadvantages.

The main disadvantages associated with expert learning systems can be divided into psychological associated with the lack of "live" communication with the teacher, high requirements for self-organization and technical which are caused by the imperfection of content, technologies and telecommunications infrastructure.

The advantages of expert learning systems are:

1. Geographical and temporal advantages.
2. Personalization of the learning process. Opportunity to train various categories of people, including those with disabilities.
3. Expanding the information being studied and increasing the intensity of learning.
4. Optimization and automation of the knowledge transfer process.

Master's work is devoted to the actual scientific problem of automation of an expert training system. As part of the research carried out:

1. The existing expert training systems are analyzed.
2. A study was made of an automated expert training system.
3. The Client-server technology of an expert training system for Internet and Intranet networks is considered.

In accordance with the problem statement, the further direction of research is the selection, development and adaptation of an expert training system, its software implementation and testing.

When writing this abstract, the master's work has not yet been completed. Final Completion: December 2013. The full text of the work and materials on the topic can be obtained from the author or his supervisor after the specified date.

List of sources

1. Brooking A. Expert systems. Principles of work and examples.: Per. from English. / A. Brooking, P. Jones; [Ed. R. Forsyth. - M.: Radio and communication, 1987. - 224 p.

2. - American Association of Piece Intellect American Association for Artificial Intelligence (AAAI).

7. Karpova I.P. Analysis of student responses in automated learning systems / I.P. Karpova // - Information technologies, 2001, No. 11. - p.49-55.

8. Pusilovsky, P., Adaptive and Intelligent Technologies for Web-based Education. In C. Rollinger and C. Peylo (eds.), Special Issue on Intelligent Systems and Teleteaching, Konstliche Intelligenz, 4, 19 - 25.

9. Burdaev V.P. Client-server technology of an expert training system for Internet and Intranet networks. // Artificial intelligence.

11. Andreichikov AV Intelligent information systems. /BUT. V. Andreichikov, O. N. Andreichikova.: Textbook. - M.: Finance and statistics, 2004. - 424 p.

12. Atanov G. A. Education and artificial intelligence, or the foundations of modern didactics of higher education. /G. A. Atanov, I. N. Pustynnikova. - Donetsk: DOU, 2002. - 504 p.

13. Marvin Minsky. The Emotion Machine: Commonsense Thinking, Artificial Intelligence, and the Future of the Human Mind. 2007. - 332 p.

Expert systems are one of the main applications of artificial intelligence. Artificial intelligence is one of the branches of computer science, which deals with the tasks of hardware and software modeling of those types of human activity that are considered intellectual.

The results of research on artificial intelligence are used in intelligent systems that are able to solve creative problems that belong to a specific subject area, knowledge about which is stored in the memory (knowledge base) of the system. Artificial intelligence systems are focused on solving a large class of tasks, which include the so-called partially structured or unstructured tasks (weakly formalized or non-formalizable tasks).

Information systems used to solve partially structured tasks are divided into two types:

Creating management reports (performing data processing: search, sorting, filtering). The decision is made on the basis of the information contained in these reports.

Developing possible alternative solutions. Decision making is reduced to choosing one of the proposed alternatives.

Information systems that develop alternative solutions can be model or expert:

Model information systems provide the user with models (mathematical, statistical, financial, etc.) that help to ensure the development and evaluation of solution alternatives.

Expert information systems ensure the development and evaluation of possible alternatives by the user through the creation of systems based on knowledge obtained from specialists - experts.

Expert systems are computer programs that accumulate the knowledge of specialists - experts in specific subject areas, which are designed to obtain acceptable solutions in the process of information processing. Expert systems transform the experience of experts in a particular field of knowledge into the form of heuristic rules and are intended for advice from less qualified specialists.

It is known that knowledge exists in two forms: collective experience, personal experience. If the subject area is represented by collective experience (for example, higher mathematics), then this subject area does not need expert systems. If in the subject area most of the knowledge is the personal experience of high-level specialists and this knowledge is semi-structured, then this area needs expert systems. Modern expert systems are widely used in all areas of the economy.

The knowledge base is the core of the expert system. The transition from data to knowledge is a consequence of the development of information systems. Databases are used to store data, and knowledge bases are used to store knowledge. The database, as a rule, stores large amounts of data with a relatively low cost, and knowledge bases store small, but expensive information arrays.

The knowledge base is a collection of knowledge described using the chosen form of their presentation. Filling the knowledge base is one of the most difficult tasks, which is associated with the choice of knowledge, their formalization and interpretation.

The expert system consists of:

knowledge base (as part of working memory and rule base) designed to store initial and intermediate facts in working memory (it is also called a database) and store models and rules for manipulating models in the rule base

problem solver (interpreter), which provides the implementation of a sequence of rules for solving a specific problem based on facts and rules, stored in databases and knowledge bases

explanation subsystem, allows the user to get answers to the question: “Why did the system make such a decision?”

a knowledge acquisition subsystem designed both to add new rules to the knowledge base and to modify existing rules.

user interface, a set of programs that implement the user's dialogue with the system at the stage of entering information and obtaining results.

Expert systems differ from traditional data processing systems in that they usually use symbolic representation, symbolic inference and heuristic search for solutions. For solving weakly formalized or non-formalized tasks, neural networks or neurocomputers are more promising.

The basis of neurocomputers is neural networks - hierarchical organized parallel connections of adaptive elements - neurons, which provide interaction with objects of the real world in the same way as the biological nervous system.

Great successes in the use of neural networks have been achieved in the creation of self-learning expert systems. The network is set up, i.e. train, passing through it all known solutions and achieving the required answers at the output. The setting consists in selecting the parameters of the neurons. Often use a specialized training program that trains the network. After training, the system is ready to work.

If in an expert system its creators preliminarily lay knowledge in a certain form, then in neural networks it is not known even to developers how knowledge is formed in its structure in the process of learning and self-learning, i.e. The network is a black box.

Neurocomputers, as artificial intelligence systems, are very promising and can be improved indefinitely in their development. Currently, artificial intelligence systems in the form of expert systems and neural networks are widely used in solving financial and economic problems.

"