Intelligent Expert System of Knowledge Examination of Medical Staff Regarding Infections Associated with the Provision of Medical Care Dmytro Chumachenko1[0000-0003-2623-3294], Victor Balitskii1[0000-0003-0473-1022], Tetyana Chumachenko2[0000-0002-4175-2941], Victoria Makarova2[0000-0003-4104-0052] and Maryna Railian 2[0000-0002-1587-4435] 1 National Aerospace University „Kharkiv Aviation Institute“, Chkalow str., 17, Kharkiv, Ukraine 2 Kharkiv National Medical University, Nauky ave., 4, Kharkiv, Ukraine dichumachenko@gmail.com Abstract. The use of intellectual recommender systems, based on the analysis of the questionnaire of medical workers, generates the rules for their education and detects gaps in knowledge, will allow to effectively solve the issues of pre- vention of IRMC, which will lead to a decrease in morbidity. The aim of the study is to create a software package that allows for an automated assessment of the knowledge of nursing staff on the prevention of hand hygiene in the perfor- mance of professional duties. With the help of the developed program complex, 817 paramedical medical personnel of medical institutions of the Kharkiv re- gion were surveyed. Results are statistically processed. Dependencies of knowledge and skills of medical personnel have been established. Keywords: expert system, knowledge assessment, infections related to medical care. 1 Introduction Infections related to medical care (IRMC) cause negative health and social conse- quences and significant economic losses for patients and health systems around the world [1]. At the same time, good hand hygiene at the right time and in the right way can save the lives of many people [2]. Accurate compliance with the rules of hand hygiene is recommended for the prevention of all IRMC [3]. The management of health care institutions is responsible for ensuring the preven- tion and control of IRMC cases and the prevention of transmission of epidemiologi- cally important pathogens [4]. Medical professionals who are directly involved in the provision of medical care to patients (for example, doctors and nurses), as well as support staff are responsible for the continued use of IRMC prevention and control practices, including hand hygiene [5]. Fulfillment of this rule is not without systematic training of hospital personnel, which may include lectures, practical exercises, and the use of training materials available in paper or electronic form [6]. Training can help define guidelines for per- forming medical procedures in accordance with modern requirements, based on evi- dence-based medicine [7]. Classes with medical workers allow to determine the pur- pose of the program for the prevention and control of IRMC, increase staff compli- ance with the requirements of hand hygiene, dictates the need for its evaluation [8]. Providing feedback to department staff increases the effectiveness of IRMC preven- tion and control programs [9]. A high level of computerization of medical activity is the automation of health care management [10-11]. An automated management system is a means of collecting, processing, storing, storing and transmitting medical information, designed to auto- mate both the management process itself and the professional activities of each medi- cal worker [12]. The modern works of many scientists are devoted to the development of intellectual problem-oriented systems and their application to medical domain [13– 16]. Quality computer aided systems with intelligent interaction with medical workers will solve a lot of problems in medical staff every day work [17]. The aim of this research is to create a software package that allows an automated assessment of the nursing staff’s knowledge of hand hygiene prevention in the per- formance of professional duties. 2 Questionnaire Development In order to assess the quality of knowledge of nurses at several medical and preven- tive institutions on hand hygiene, as well as to study their knowledge on the rules for performing medical manipulations in the hospital and identifying the reasons that make it difficult for the nurses to follow the rules of hand hygiene. The questionnaire included questions characterizing respondents by sex, age and work experience, as well as questions reflecting the content of the current regulatory document: Order of the Ministry of Health of Ukraine of September 21, 2010 No. 798 "On Approval of the Surgical and Hygienic Handling Guidelines of Medical Personnel". 3 Expert System Development At present, computer equipment in solving operational problems, as a rule, is used only in the case when the algorithms for solving problems are strictly formalized [18]. In this case, the initial data for input into the computer is prepared by highly qualified specialists [19]. As a result, the optimal solution of problems cannot be fully imple- mented using simple and accurate algorithms, and therefore is automated without the creative participation of man. Such tasks are usually referred to as artificial intelli- gence (AI) [20]. One of the applied areas of AI is the creation of expert systems (ES), designed to reduce the intellectual burden on humans [21-22]. The ES is designed to solve prob- lems in a specific problem area. In them, the source of knowledge is the experts who enable the programmer to develop appropriate strategies (programs) for solving spe- cific problem tasks [23-24]. It is advisable to resort to the development of ES in cases where:  the problem can not be strictly solved with a rigid (known) algorithm [25];  to solve problems requires an intuitive approach [26];  preparation of initial data requires encyclopedic knowledge and experience of highly qualified specialists [27];  the necessary phased verification and correction of intermediate parameters [28];  in the course of the decision change of circumstances (conditions) is possible [29];  in the course of solving the problem, the level of knowledge and experience of the staff changes [30];  time for performance of work changes [31]. When composing such programs, it is easiest to use heuristic methods (rules) that work according to the principle: “if condition A is set, then you should follow rule A” [32]. It is also possible to represent knowledge using semantic networks, as well as a frame [33]. To solve problems in a specific problem area, you can use the ES shell (skeleton and inference system) from another problem area, filling it with new necessary knowledge [34]. In such a shell, you can enter all the problems and ways to resolve them, both in the form of rules and in the form of complexly organized interacting structures. So, for example, when creating the ES, it is advisable to use a shell in the field of medical diagnostics to identify failed elements of the RTS [35]. The ES can significantly reduce the cost of manpower and resources, reduce the requirements for the qualifications of service personnel, as well as minimize the time spent on solving similar tasks [36]. At the same time, human activity will be reduced to the ability to work with computer programs and to collect the necessary initial data. The first order logic language, the syntax and semantics of which we will define in the next section, is based on the concepts of objects and relations. It became extreme- ly important for mathematics, philosophy and artificial intelligence precisely because these spheres of knowledge (and in fact the main part of everyday human existence) can be quite productively regarded as relating objects and relations between them [37]. The first-order logic also makes it possible to express facts about some or all of the objects in the Universe. This makes it possible to present general laws, or rules, such as the following statement: "In the squares adjacent to the square where the vampus is located, there is an unpleasant smell" [38]. The main difference between propositional logic and first-order logic is that each of these languages makes a different ontological contribution to the description of reality, that is, they represent the nature of reality in different ways [39]. For example, in propositional logic it is assumed that there are only facts that are or do not belong to this world. Each fact can be in one of two states: to be true or false. In the first order logic, broader assumptions are made, namely, that the world consists of objects between which there may or may not be some relationship [40]. Some variants of special-purpose logic allow for an even greater ontological contribution; for example, in temporal logic it is assumed that the facts take place at specific time intervals and these intervals (which can be considered infinitesimal or finite) are ordered [41]. Therefore, in variants of special-purpose logic, a first-class status is provided for some types of objects of a special kind (and axioms about these objects), and their definitions are not simply entered into the knowledge base. In the logic of high order, the objects themselves are treated as relations and functions, are considered in the logic of the first order. This allows you to formulate statements about all relation- ships, for example, if you need to determine exactly what the concept of a transitive relation means. Unlike most variants of special-purpose logic, high-order logic is strictly more expressive than first-order logic, in the sense that some statements of higher-order logic cannot be expressed using any finite number of statements of first- order logic. Logic can characterize its epistemological contribution to cognition; this refers to the possible states of knowledge that it allows to express for each fact. And in propo- sitional logic, and in logic of the first order, any statement is fact, and the agent either trusts with the statement that this expression is true, or trusts with the statement that it is false, or has no opinion on this matter. Therefore, in such variants of logic there are three possible states of knowledge relating to any utterance. The main syntactic elements of first-order logic are symbols denoting objects, rela- tions, and functions. Therefore, the symbols themselves are divided into three types: constant symbols denoting objects; into predicate symbols denoting relations, and functional symbols denoting functions. Semantics must link sentences with models in order to be able to determine the truth. To solve such a problem, an interpretation is needed, which determines which particular objects, relations and functions certain constant, predicate and functional symbols refer to. The truth of any utterance is determined with the help of some model and some in- terpretation of the symbols of this utterance. Therefore, the logical consequence, ad- missibility, and other properties of statements are defined in terms of all possible models and all possible interpretations. It is important to note that the number of ele- ments of the problem area in each model can be unlimited, for example, the elements of the problem area can be integers or real numbers. Therefore, a limited number of possible models, as well as the number of interpretations. After determining the logic that allows the use of objects, it becomes quite natural to create tools that allow you to express the properties of entire collections of objects, rather than sorting these objects by name. Quantifiers allow you to do this. The first- order logic includes two standard quantifiers, called quantifiers of universality and existence. 4 Program Realization of Developed Expert System In order to assess the quality of knowledge of nurses at several medical and preven- tive institutions on hand hygiene, as well as to study their knowledge on the rules for performing medical manipulations in the hospital and identifying the reasons that make it difficult for the nurses to follow the rules of hand hygiene. The questionnaire consisted of questions characterizing respondents by sex, age and length of service, and questions reflecting the content of the current document regulates: Order of the Ministry of Health of Ukraine of September 21, 2010 No. 798 "On Approval of the Surgical and Hygienic Handling Guidelines of Medical Personnel" To automate the assessment of knowledge and data collection from medical staff, a web application has been developed, the platform for which .net core is chosen, since during development most of the necessary components of the application can be downloaded as separate modules through the package manager NuGet. This allows you to reduce the number of excess dependencies and the total size of the finished product. Also, a project based on .NET Core is fairly easy to transfer to the cloud. Microsoft Azure already supports hosting .NET Core projects in both Application Services and virtual machines. .NET Core allows small projects to take full advantage of the enterprise-level platform, while providing convenient and development tools, as well as low-cost infrastructure. Also a project based on .NET Core is best suited for computing and analytical tasks. The developed web applications used technologies such as HTML, CSS, LESS, JavaScript,. NET CORE, Angular, Grunt, MS SQL, Material, the main advantage of which is the low threshold of entry and ease of use. Using the Angular framework reduces development time. Also Angular allows you to create one-page applications. According to the logic of their work, they reduce the load on the server and can pro- vide a much richer interface for the end user. There are opportunities to use these technologies to create hybrid mobile applications. Grunt was chosen to compile the project, because it can be used to easily compile Less, compress the CSS and minify JavaScript in order to keep the files as small as possible. MS SQL is the generic name for Microsoft SQL Server 2005/2008/2012/2014/2016 Express Edition. It is a reliable server with excellent fea- tures, high speed and maximum security. MS SQL is installed on the central server, and all other computers are connected to this server. Well suited for storing test re- sults for further analysis, as MS SQL is a high-end SQL server, it takes care of man- aging the database, its security and stability. MS SQL guarantees a high level of data protection and virtually no problems. This web application is implemented in the form of a web page on which the user is asked to answer the question of the developed questionnaire. The result of the pro- gram are recommendations based on the analysis of the received answers of the user of the system in the process of questioning. Recommendations are the result of an expert system developed in a web application. The expert system is the direction of research in the field of artificial intelligence to create computing systems that can make decisions similar to the decisions of experts in a given subject area. Expert systems have one big difference from other artificial intelligence systems: they are not designed for solving some universal tasks, such as neural networks or genetic algorithms. Expert systems are designed for high-quality problem solving in a specific area of developers, in rare cases - areas. The expert system incorporates two main blocks:  knowledge base;  machine logical inference. The knowledge base contains the user's answers in the questionnaire process, the cor- rect answers to the questionnaire questions and recommendations are based on expert opinion. The logical inference machine determines the correct answers of the user and, based on the logic of second-order predicates, determines the recommendations for the system user. 5 Results A survey of 817 paramedical medical workers of medical institutions was conducted. Before the survey, the respondents were explained the purpose of the survey and the rules for filling out the questionnaire. The survey was conducted on a voluntary basis. Respondents chose the answer option on their own. The results of the study were processed statistically. An analysis of the data obtained during the survey was conducted. The results were tested for validity. Validity characterizes the suitability of the test to measure a certain amount. It should be noted that it is impossible to talk about the validity of the test without specifying the conditions for its use. Also, validity means the test's suitability to measure the property for which it is intended to determine. This test is aimed at assessing the level of knowledge of nursing staff regarding the prevention of hand hygiene in the performance of professional duties. The measured property in this case is the level of knowledge of medical personnel. During the test, the level of knowledge has not changed. The “position” criterion does not affect the passing of the test, only the level of knowledge of nursing staff influenced, and is a measurable property. It can be argued that this test is valid, and subsequent analysis of the data obtained will carry a qualita- tive and believable assessment of the knowledge of nursing staff. The questionnaire was divided into two sets of questions: “Skill” and “Knowledge”. The “Skill” block was formed to reveal the level of practical skills of medical per- sonnel. For example: How often do you follow hand hygiene before and after contact- ing a patient? The block "Knowledge" was formed to test the level of knowledge of medical per- sonnel. For example: What are the stages of surgical treatment of hands? One of the goals of the work was to determine the relationship between these two blocks. For further analysis, all the results of the survey were recoded accordingly. The answers to the first block of questions were grouped into two categories: “The skill level is above average” and “The skill level is less than average.” The answers to the second block of questions were grouped into categories: “Correct” and “Incor- rect”. The obtained data was processed in the R language in the RStudio software environment. RStudio is available in open source versions and has useful features for both beginners and experienced R developers, including code completion, source execution, search history, and support for developing Sweave documents. The result was a data set consisting of the average values of each group for each block of ques- tions. It was found that, on average, 791 people answered the first block of questions (at the level above the average), at the level below the average — 26 people. 492 people answered the questions of the second block correctly, 326 people incorrectly. A total of 817 people were interviewed. Since all data are presented on a categorical scale, the analysis was performed us- ing the Pearson Chi-square test. As a result, p-value = 1 was obtained (since p-value> 0.05), so it can be argued that there is no direct connection between the first block of questions and the second. It was revealed that the older the employee, the more experienced he is and his lev- el of knowledge about the prevention of hand hygiene does not depend on such two signs as “age” and “experience”. Both signs are represented by a categorical scale, the Pearson Chi-Square test was used to determine the significance of differences. This is a non-parametric method that allows assessing the significance of differences between the actual (identified as a result of the study) number of cases or qualitative character- istics of the sample falling into each category, and the theoretical amount that can be expected in the studied groups with the validity of the null hypothesis. As a result, it was found that p-value <2.2e-16 (no difference). It was also found that medical staff adhere to the rules governing the requirements for handling hands, only if these rules are available. It was also calculated that p-value <2.2e-16 (there is a connection), so it can be argued that medical staff wear medical gloves if necessary and follow the rules of hand hygiene after they are removed. It can be argued that most medical personnel adhere to hand hygiene before and after con- tact with the patient and the surfaces of objects and equipment, since the dependence was confirmed by the Pearson Chi-Square criterion (p-value <2.2e-16). It can be ar- gued that medical personnel are aware of the rules established at the medical institu- tion and based on the questionnaire, it is possible to identify the reasons for non- compliance with the established rules and give an individual assessment for each interviewed employee, general statistics and recommendations of the medical institu- tion administration to resolve the problems. The results showed that the level of knowledge of medical personnel is insuffi- cient. 6 Conclusions A software package has been developed that allows for automated interviewing of health workers and greatly simplifying its analysis. It has been established that in medical institutions where the study was conducted there are clearly developed rules for the treatment of hands based on current regulatory documents. In-patient work- shops are held on hand hygiene. Most respondents know the algorithms of washing and hygienic antiseptics of hands, less than half know the sequence of actions in the surgical treatment of hands. A statistical analysis was carried out which allows conclusions to be drawn regard- ing hand hygiene in medical institutions. It has been established that in medical insti- tutions where questionnaires were conducted there are clearly developed rules for the treatment of hands, based on current regulatory documents. It is known that in-patient training seminars on hand hygiene are held. Among the respondents, the majority knows the algorithms of washing and hygienic antiseptics of hands, less than half know and can list the sequence of actions during surgical treatment of hands. The results of the survey of nursing staff showed that when conducting training seminars for health workers, hand hygiene rules should be given special attention to the prevention of CB, including providing health workers with moisturizing protective creams, focusing on the correct implementation of all stages of hygienic and surgical treatment of hands; it is necessary to strengthen the control over the uninterrupted provision of hospital departments with alcohol antiseptics and liquid soap, and to revise the load standards for nurses to ensure the quality of medical care for patients. The results of the survey were tested for validity and it was found that the data ob- tained are valid. In the future, based on these results, it is possible to develop a soft- ware package, in the form of a web application, which will be aimed at individual assessment of each respondent, general statistics and recommendations for the admin- istration of the medical institution to eliminate insufficient knowledge and possible factors that interfere with hand hygiene. References 1. Haque, M., Sartelli, M., McKimm, J., Abu Bakar, M.: Health care-associated infections - an overview. In: Infection and drug resistance, vol. 11, pp. 2321–2333 (2018) 2. Boyce, J.M., Pittet, D.: Guideline for hand hygiene in health-care settings. Recommenda- tions of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA hand hygiene task force. Society for Healthcare Epidemiolo- gy of America/Association for Professionals in Infection Control/Infectious Diseases Soci- ety of America. In: MMWR Recommendations and Reports, vol. 51(RR-16), pp.1-45 (2002) 3. Dellit, T.H. et al.: Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship. In: Clinical Infectious Disease, Vol. 44, Iss. 2, pp. 159-177 (2007) 4. Minhas, S., Kotwal, A., Singh, M.: Infection Control in Health Care Facilities. In: Medical journal, Armed Forces India, vol. 67(1), pp. 7–8 (2011) 5. Danasekaran R, Mani G, Annadurai K.: Prevention of healthcare-associated infections: protecting patients, saving lives. In: International Journal of Community Medicine. Public Health, Vol. 1(1), pp. 67–68 (2014) 6. Hasske, E., Beil, M., Keller, K.: Competency-based Education and Training of medical staff. A Programm of the Medical Academy Waldbreitbach: Concept - Implementation - Materials. In: GMS journal for medical education, vol. 34(4), Doc41 (2017) 7. Masic, I., Miokovic, M., Muhamedagic, B.: Evidence based medicine - new approaches and challenges. In: Acta informatica medica : AIM : journal of the Society for Medical In- formatics of Bosnia & Herzegovina : casopis Drustva za medicinsku informatiku BiH, vol. 16(4), pp. 219–225 (2008) 8. Burton, D.C. et al.: Trends in catheter–associated urinary tract infections in adult intensive care units–United States, 1990–2007. In: Infection Control and Hospital Epidemiology, Vol. 32 (8), pp.748–756 (2011) 9. Polyvianna, Yu., Chumachenko, D., Chumachenko T.: Computer Aided System of Time Series Analysis Methods for Forecasting the Epidemics Outbreaks, In: 2019 15th Interna- tional Conference on the Experience of Designing and Application of CAD Systems (CADSM), pp. 7.1-7.4 (2019) 10. Dotsenko, N., Chumachenko, D., Chumachenko, I.: Modeling of the processes of stake- holder involvement in command management in a multi-project environment. In: 2018 IEEE 13th International Scientific and Technical Conference on Computer Sciences and Information Technologies, CSIT 2018 – Proceedings, pp. 29-32 (2018) 11. Dotsenko, N., Chumachenko, D., Chumachenko, I.: Project-oriented management of adap- tive commands formation recourses in multi-project environment. In: CEUR Workshop proceedings, Vol. 2353, pp. 911-923 (2019) 12. Mazorchuck, M., Dobriak, V., Chumachenko, D.: Web-Application Development for Tasks of Prediction in Medical Domain. In: 2018 IEEE 13th International Scientific and Technical Conference on Computer Sciences and Information Technologies (CSIT), Lviv, pp. 5-8 (2018) 13. Chen, J. et. al.: Syndrome Differentiation and Treatment Algorithm Model in Traditional Chinese Medicine Based on Disease Cause, Location, Characteristics and Conditions. In: IEEE Access, vol. 6, pp. 71801 – 71813 (2018) 14. Chumachenko, D.: On Intelligent Multiagent Approach to Viral Hepatitis B Epidemic Pro- cesses Simulation. In: Proceedings of the 2018 IEEE 2nd International Conference on Data Stream Mining and Processing, DSMP 2018, pp. 415-419 (2018) 15. Bazilevych, K. et al.: Stochastic modelling of cash flow for personal insurance fund using the cloud data storage. In: International Journal of Computing, Vol. 17, Iss. 3, pp. 153-162 (2018) 16. Prokhorov, O. V. et. al.: Regional resources management by agent-based simulation. In: Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu, Vol. 4, pp. 107-114 (2016) 17. Cahan, A., Cimino, J. J.: A Learning Health Care System Using Computer-Aided Diagno- sis. In: Journal of medical Internet research, Vol. 19(3), e54 (2017) 18. Taylor, A. T., Garcia, E. V.: Computer-assisted diagnosis in renal nuclear medicine: ra- tionale, methodology, and interpretative criteria for diuretic renography. In: Seminars in nuclear medicine, Vol. 44(2), pp. 146–158 (2014) 19. Matthias, D., Udo, O.: Expert System for Medical Diagnosis of Hypertension and Anae- mia. In: MAYFEB Journal of Environmental Science, Vol. 3, pp. 12-19 (2017) 20. Metaxiotis, K.S., Samouilidis, J.‐E.: Expert systems in medicine: academic illusion or real power?. In: Information Management & Computer Security, Vol. 8 Issue: 2, pp.75-79 (2000) 21. Chumachenko, D., Chumachenko, K., Yakovlev, S.: Intelligent Simulation of Network Worm Propagation using the Code Red as an Example. In: Telecommunications and Radio Engineering, Vol. 78, Iss. 5, pp. 443-464 (2019) 22. Chumachenko, D., Yakovlev, S.: On Intelligent Agent-Based Simulation of Network Worms Propagation. In: 2019 15th International Conference on the Experience of Design- ing and Application of CAD Systems (CADSM), pp. 3.11-3.13 (2019) 23. Guo, J., Li, B.: The Application of Medical Artificial Intelligence Technology in Rural Ar- eas of Developing Countries. In: Health equity, Vol. 2(1), pp. 174–181 (2018) 24. Jabeen, S.H., Zhai, G.: A prototype design for medical diagnosis by an expert system. In: 7th International workshop on computer science and engineering, China, pp. 1413-1417 (2017) 25. Noguchi, T. et al.: A practical use of expert system "AI-Q" focused on creating training da- ta. In: 2018 5th International Conference on Business and Industrial Research (ICBIR), Bangkok, pp. 73-76 (2018) 26. Lőrincz, T., Szakonyi, B., Lipovits, Á., I. Vassányi: Development of an expert system framework for lifestyle improvement. In: 2017 IEEE 30th Neumann Colloquium (NC), Budapest, pp. 000093-000098 (2017) 27. Kosa, I., et. al.: A fast android based dietary logging application to support the life style change of cardio-metabolic patients, In: Med-e-Tel Conference, Vol. 7, pp. 553-556 (2014) 28. Kurniawan, D.A., Sihwi, S.W., Gunarhadi: An expert system for diagnosing dysgraphia. In: 2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE), Yogyakarta, pp. 468-472 (2017) 29. Nusai, C., Cheechang, S., Chaiphech, S., Thanimkan, G.: Swine-Vet: a Web-based Expert System of Swine Disease Diagnosis. In: Procedia - Procedia Comput. Sci., Vol. 63, No. I, pp. 366-375 (2015) 30. Sihwi, S. W., Andriyanto, F., Anggrainingsih, R.: An expert system for risk assessment of information system security based on ISO 27002. In: 2016 IEEE International Conference on Knowledge Engineering and Applications (ICKEA), Singapore, pp. 56-61 (2016) 31. Teubler, T., Hellbrück, H.: Design of expert systems for autonomous underwater vehicle control. In: OCEANS 2016 MTS/IEEE Monterey, Monterey, CA, pp. 1-6 (2016) 32. Lacave, C., Diez, F.: A review of explanation methods for heuristic expert systems. In: The Knowledge Engineering Review, Vol. 19(2), pp. 133-146 (2004) 33. Menemencioglu, O., Orak, I.M.: A Review on Semantic Web and Recent Trends in Its Applications. In: 2014 IEEE International Conference on Semantic Computing, Newport Beach, CA, pp. 297-303 (2014) 34. Fernando, I., Henskens, F., Cohen, M.: A Domain Specific Expert System Model for Di- agnostic Consultation in Psychiatry. In: 2011 12th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Compu- ting, Sydney, NSW, pp. 3-6 (2011) 35. Leung, K.S., Wong, M.H.: An expert-system shell using structured knowledge: an object- oriented approach. In: Computer, Vol. 23, No. 3, pp. 38-47 (1990) 36. Akanbi, A. K., Masinde, M.: Towards the Development of a Rule-Based Drought Early Warning Expert Systems Using Indigenous Knowledge. In: 2018 International Conference on Advances in Big Data, Computing and Data Communication Systems (icABCD), Dur- ban, pp. 1-8 (2018) 37. Sasikumar, M., Ramani, S., Raman, S.M., Anjaneyulu, K.S.R., Chandrasekar, R.: A prac- tical introduction to rule based expert systems, New Delhi:Narosa Publishing House (2007). 38. Salehi, S., Saffar, M. T., Taghiyareh, F., Badie, K.: A multi context dynamic test bed for simulating real world constraints in agents' teamwork. In: 6th International Symposium on Telecommunications (IST), Tehran, pp. 1201-1206 (2012) 39. Mashtalir, V.P., Yakovlev, S.V.: Point-set methods of clusterization of standard infor- mation. In: Cybernetics and Systems Analysis, Vol. 37, №3, pp. 295-307 (2001) 40. Gerasin, S.N., Shlyakhov, V.V., Yakovlev, S.V.: Set coverings and tolerance relations. In: Cybernetics and Systems Analysis. Vol. 43, №3, pp. 333–340 (2008) 41. Mashtalir, V.P., Shlyakhov, V.V., Yakovlev, S.V.: Group structures on quotient sets in classification problems. In: Cybernetics and Systems Analysis, Vol. 50, №4, pp. 507-518 (2014)