Українська   Русский
DonNTU   Masters' portal

Abstract

Content

An expert system is a software tool that uses expert knowledge to provide highly efficient solutions to non-formalized problems in a narrow subject area. The basis of the ES is the knowledge base (KB) about the subject area, which is accumulated in the process of building and operating the ES. The accumulation and organization of knowledge is the most important property of all ES.[1]

1. Relevance of the topic

The need for ontology is associated with the impossibility of a fully automatic processing of natural language texts by existing means. In different communities, different designations are often found. for the same concepts. Therefore, for high-quality word processing, it is necessary to have a detailed description of the problem area with many logical connections that show the relationship between the terms of the area. The use of ontologies makes it possible to present a natural language text in such a way that it becomes suitable for automatic processing. Information and propaganda system. terms between all users of the project. Also the wide application of the problem of ontological analysis. Within these tasks with the help of ontological research accumulate valuable information about functioning of complex systems. This analysis usually begins with drawing up glossary of terms that is used in discussion and characteristics objects and processes that make up the system under consideration, as well as the creation systems of precise definitions of these terms. In addition, the main logical relationships between supplying terms and concepts. The result this analysis is a glossary of terms, their precise definitions and relationships between them. Use the collected information in the reorganization process a new system or building new systems[2].

The reasons why there is a need to use ontologies:

2. Purpose and objectives of the study, planned results

The aim of the proposed work is to solve two problems:

1) Analysis of modern ontology editors from the point of view of:

2) Determination of the current directions of development of ontology editors and setting the task of creating new generation ontology editors.

Formulation of the problem:

Intelligent patterns, presented in the form of ontologies and built on the basis of UML class diagrams, as:

Those at the entrance:

At the exit:

Also, this implementation would allow synthesizing program code based on the production approach.

3. Research and development overview

The topic under study is popular not only in national but also in international scientific communities.

This section will provide an overview of research in the field of creating intelligent patterns, ontology.

3.1 Review of international sources

Ontology-Based Expert Systems — Reproducing Human Learning, Rahul Matkar, Ajit Parab [ 9 ] focuses on the learnability of expert systems. This article introduces an elite system known as expert systems that attempts to mimic the behavior of the Human Expert. Expert systems work according to the concept of "Knowledge Base". This knowledge base is created by the knowledge engineer after conducting a series of interviews with the people specialist. The Inference Engine uses facts from the Knowledge Base to find a solution to a problem. The performance of an expert system depends entirely on the quality of the knowledge base and the inference engine. The main problem to consider when designing expert systems is the ability to learn things on your own. Expert systems copy the approach of human experts to problem solving, similarly, expert systems can also "reproduce human learning behavior." By studying a new fact, people use their existing knowledge and try to respond appropriately to the new fact. Likewise, an expert system, if given a baseline — strict set rules to be followed and the ability to infer connections between different facts (ontology) during learning, they can also extract or learn new facts. in the same way Human Experts learn or expand their knowledge.

An Ontological Approach to Diagnosis and Classification for a Health and Nutrition Expert System [ 10 ] presents how to make an ontology-based expert system easy to use and free apply to community sustainability issues. The ontology itself plays an important role in a variety of knowledge and management practices that can facilitate communication between expert domains and users. The scope of this research is health and nutrition, which is expected to help people understand the anxiety they are experiencing. The result of this research is a model expert system and mobile applications that will help users overcome health and nutritional challenges using an ontology approach. The aim of this study is to develop an application based on an ontology method to make it easier for people to find information about expert systems.

3.2 Review of national sources

In the article Ontologies in knowledge-based systems: the possibilities of their application, Smekhun Ya.A. [4], explores the main aspects and roles of development, and the use of ontologies in knowledge-based systems to describe basic concepts in specific subject areas.

In the work Ontological approach and its use in knowledge representation systems, C.H. Shcheglov [5], the ontological approach and its use in knowledge representation systems are considered. The main role in the description of knowledge is assigned to ontologies, which are used in the design of knowledge bases, the creation of expert systems and decision support systems, the development of environments focused on the sharing of information by several users, and the development of various search engines.

In the article Formation of the knowledge base of the expert system based on ontology using the original language of knowledge representation, A. V. Akhaev, I. A. Khodashinsky. [6], a production-type knowledge representation language is considered for generating recommendations based on an ontology. The syntax of the proposed knowledge representation language is described, the process of forming rules based on the ontology is considered, examples of developed rules are given.

In the article Ontological approach to building the knowledge base Superhard materials, V.N. Kulakovsky, A.A. Lebedev, K.Z. Gordashnik, E.M. Chistyakov, I. V. Skvortsov [7], a complexly structured subject area of superhard materials, the components of the STM knowledge base and the process of its development based on the metaontology Superhard materials (STM) are determined.

In the work Providing weak connectivity of the expert system and the ontological knowledge base by adding a serving layer, Ryaskov A.S. [8], the assessment of the current architectural implementation of communication between expert systems and knowledge bases is considered, the drawbacks of — the main disadvantage is the need to rewrite the interface layer between the expert system and the knowledge base in case of any change in the data exchange protocol between them, the goal is to reduce the connectivity of the expert system and the knowledge base.

In the article Analysis of ontology editors from the point of view of class representation and production rules, Kutelev R., Pahota M., Grigoriev A. [3], a comparative analysis of ontology editors has been carried out. The analysis of the representation of classes, patterns. The possibility of using productions for class synthesis in ontology editors has been studied.

3.3 Browse local sources

Consider the work of other masters.

In the work of Lev Olegovich Vorobyov Software synthesis of object-oriented design patterns [11] is the development of a new way of automating the programming process based on the use of SOLID principles and applying ontologies.

Nikita Olegovich Bilyk's work Models and algorithms for updating knowledge of expert systems based on the ontological approach [12] describes the process of developing an expert system and introducing an ontology into it. The author discusses the benefits of this approach.

4. Expert system structure

The generalized structure of the expert system is shown in Figure 1.

Результаты экспериментальных исследований

Figure 1 — Generalized structure of the expert system (5 frames, 1 frame in 1.2 seconds, 5 repetitions)

4.1 Spatial knowledge

Objects have two properties: spatial and nonspatial. Spatial properties define objects in three categories, namely position of objects, shape of objects and size of objects. Nonspatial properties represent color and category of the object. These properties support the scheduling of robot tasks. Spatial presentation is necessary to represent the concept of space and form in a robotic environment. The AI ??system is based on the development of a high-level robot task, in which all knowledge the domain are represented using spatial functions. But robot navigation carried out with incomplete knowledge of the subject area, unknown objects and unfamiliar the location of objects in the subject area. To solve this problem, spatial an ontology-based representation defines both an explicit and implicit task specification in the domain and also provides a map for the simulated area. Spatial representation explicitly defines spatial entities such as the location of the object, the shape of the object, and defines the position of the object in the domain. An implicit specification controls the behavior of the robot and movement of action to reach a destination in space. This spatial the view is based on three main categories, namely spatial essence, spatial relationships and fuzzy information that help the robot to work successfully and plan within one's environment[13]

4.2 Semantic knowledge

Semantic knowledge can represent general knowledge such as concepts, their relationship, and how they are semantically related. Semantic knowledge provides instructions and detailed information required to execute in an intelligent system. Semantic knowledge allows you to output new information that allows the robot to perform a large set of tasks. The robot must have sufficient knowledge to perceive deterministic environment and access to methods of performing actions using this semantic knowledge. Task scheduling has a sequence of ordered actions for achieving a high-level goal[14].

4.3 Temporary knowledge

Temporal logic is used to represent temporal information, which includes both qualitative and quantitative information. Quantitative time information expresses points in time associated with events such as start time or endings. Qualitative temporal information expresses events through temporal relationships, which define the sequential order between events. This temporary information the system helps in ordering tasks using the event start time, duration events and coincidence of two events, which is very important for robots to act in the environment environment.

4.4 Possibility of using ontologies in expert systems

One of the most important components of an expert system is the knowledge base. It is from the completeness and consistency of the knowledge available in it depends on the quality of the expert system. Ontologies can be used as a basis for building a base. For example, tasks solved by decision-making systems are poorly formalized, it is very important have detailed, consistent and logical knowledge in a given subject area. Decision-making systems often use huge amounts of knowledge. And since with the help ontology can explicitly describe the semantics of knowledge and data, then it serves as a basis for integration and joint application of different data in solving different problems.

Conclusions

In this work, the research tasks were completed:

Based on the knowledge gained, an expert system will be implemented using logical inference in intelligent patterns, built as ontologies of objects with physical semantics.

References

  1. Введение в экспертные системы. Основные понятия и определения, 2020 Режим доступа: [Ссылка] (дата обращения: 18.11.2020).
  2. Никоненко А.А. Обзор баз знаний онтологического типа // Штучний інтелект.—2009.—№ 4.— С.208-219 Режим доступа: [Ссылка] (дата обращения: 18.11.2020).
  3. Кутелёв Р.С., Пахота М.И., Григорьев А.В. Анализ редакторов онтологий с точки зрения представления классов и продукционных правил // Сборник научных трудов III Международной научно-практической конференции:сб.статей.—Донецк, 2020.— С.21-28 [Ссылка на сборник] (дата обращения: 18.11.2020).
  4. Смехун Я.А. Онтологии в системах, основанных на знаниях: возможности их применения// Международный научно-исследовательский журнал.—2016.—№ 5 (47) Часть 3.— С.173-175, режим доступа:[Ссылка] (дата обращения: 18.11.2020).
  5. Щеглов C.H. Онтологический подход и его использование в системах представления знаний //Известия Южного федерального университета. Технические науки.—2009.—№7.— С.10-17. Режим доступа: [Ссылка] (дата обращения: 18.11.2020).
  6. Ахаев А.В., Ходашинский И.А. Формирование базы знаний экспертной системы на основе онтологии с использованием оригинального языка представления знаний// Электронные средства и системы управления. Материалы докладов международной научно-практической конференции.—2013.—№2.— С.3-7. Режим доступа: [Ссылка] (дата обращения: 18.11.2020).
  7. В.Н. Кулаковский, А.А. Лебедева, К.З. Гордашник, Е.М. Чистяков, И.В. Скворцов. Онтологический подход к построению базы знаний Сверхтвердые материалы// Штучний інтелект.—2008.—№ 2.— С.91-102. [Ссылка] (дата обращения: 18.11.2020).
  8. Рясков А.С. — Обеспечение слабой связности экспертной системы и онтологической базы знаний путём добавления обслуживающего слоя // Кибернетика и программирование.— 2018.— № 2.— С. 75-82. Режим доступа: [Ссылка] (дата обращения: 18.11.2020).
  9. Rahul Matkar, Ajit Parab, Экспертные системы на основе онтологий — воспроизведение человеческого обучения // Proceedings of the First International Conference on Contours of Computing Technology.—2010.—№ 2.— С.43-47. Режим доступа: [Ссылка] (дата обращения: 18.11.2020).
  10. Фриска Н., Деа Ч. и Санти С. Онтологический подход к диагностике и классификации для экспертной системы в области здравоохранения и питания // Ontological Analyses in Science, Technology and Informatics, 2019 Режим доступа:[Ссылка] (дата обращения: 18.11.2020).
  11. Воробьёв Л.О. Программный синтез паттернов объектно-ориентированного проектирования, Донецкий национальный технический университет, 2020 Режим доступа: [Ссылка] (дата обращения: 18.11.2020).
  12. Билык Н.О. Модели и алгоритмы обновления знаний экспертных систем на основе онтологического подхода, Донецкий национальный технический университет, 2013 Режим доступа:[Ссылка] (дата обращения: 18.11.2020).
  13. Ontology based spatial planning for human-robot interaction / Belouaer L., Bouzid M., Mouaddib A.-I. 2020. [Temporal Representation and Reasoning, TIME, 2010 17th International Symposium on, IEEE (2010), pp. 103-110.
  14. Web-enabled robots / Tenorth M., Klank U., Pangercic D., Beetz M. IEEE Roboot. Autom. Mag., 18 (2) (2011), pp. 58-68