An intelligent decision support system for cultural property identification

Размещено на http://www.allbest.ru/

An intelligent decision support system for cultural property identification

Martynenko A.A.

Moroz B.I.

НнІта I.G.

Анотація

В статті розглядається проблема ідентифікації культурних цінностей, питання розробки методів і моделей організації та обробки даних і знань в інтелектуальній системі підтримки прийняття рішень ідентифікації культурних цінностей. Також зазначається складність та комплексність підходу вирішення поставленої задачі, визначені перспективи та шляхи подальшого дослідження даної предметної області.

Ключові слова: експертна система, системи пі дтримки прийняття рішень, інтелектуальний аналіз даних, ідентифікація культурних цінностей.

Аннотация

identifying cultural value

В статье рассматривается проблема идентификации культурных ценностей, вопросы разработки методов и моделей организации и обработки данных и знаний в интеллектуальной системе поддержки принятия решений идентификации культурных ценностей. Также отмечается сложность и комплексность подхода решения поставленной задачи, определены перспективы и пути дальнейшего исследования данной предметной области.

Ключевые слова: экспертная система, системы поддержки принятия решений, интеллектуальный анализ данных, идентификация культурных ценностей.

Summary

The article considers the problem of identifying cultural property, the development of methods and models for organizing and processing data and knowledge in an intelligent decision support system for cultural property identification. The complexity of the approach for solving the stated problem, prospects and ways of further research of this subject area are noted and identified.

Key words: expert system, decision support system, data mining, cultural property identification.

Introduction. Problem statement. The problem of research and development of methods and models of processing data and knowledge using intelligent decision support systems (IDSS) for the identification of cultural property (CP) is an urgent and promising task [1]. Despite the rapid development of information and intelligent technologies, there is currently no one expert system (ES) or decision support system (DSS) to systematically address the identification and analysis of cultural property. As stated by the authors in their work [1], systems of this type can additionally solve a wider range of tasks: finding and storing information about the CP, and can also be used to popularize this issue in society. It should also be noted that most of the tasks and problems of identifying cultural property are either in the process of being partially resolved or are being formulated.

Literature review. The rapid development of this area of focus has been observed in recent years. Classical works on the design and development of the principles of creating DSS are the works of Pospelova, Popova, Zade, Waterman, Bakaeva, Ulyanovskaya and others. The ideas of these works has been reflected in a number of studies which examine the problems of organizing knowledge bases, models and methods of working with fuzzy or incomplete data.

The conducted analysis of the works in this area leads to the conclusion that, despite the great development and attempts of designing the ES and the DSS, there is no definitive and comprehensive solution to the problem of the identification of cultural property with the help of implementation of the DSS.

It should also be noted that there were attempts to solve problems of this type, but they were narrowly targeted and partially solved. For example, some of the problems is solved using fuzzy set theory which cannot fully achieve the desired quality due to a number of constraints.

Previously unsolved parts. At present, there is no universal model for development of IDSS. The introduction and use of modern tools and approaches in creating systems of this type, the use of a systematic approach, according to the authors, is more effective.

Objectives of the research. To improve the quality and efficiency of solving the practical problem of identification of CP through the implementation and use of IDSS created on the basis of modern methods, models, technologies and tools.

Main part. Recently, scientists and researchers have proposed more modern and effective methods and means of object identification and the construction of modern IDSS. It should also be noted that modern solutions help to avoid or reduce a number of disadvantages that existed in previous solutions.

At the same time, there are still models of different ES and DSS that are developed under uncertainty using probability theory and fuzzy logic [9], expert systems with uncertain knowledge, using subjective probability theory and Bayes theorem as a basis for uncertainty management (Bayesian trust networks as one of the directions of construction and organization of work of expert systems, representation of knowledge using Bayesian trust network and conditional event independence), systems based on Demster-Schaeffer theory [7, 8], etc.

Also, in recent years, the use of such a powerful tool as neural networks (NN) has become very popular [10, 11]. In this case, the use of NN can be applied to the recognition, identification of CP objects and the construction of the IDSS.

Unlike the traditional use of NN for solving only the problems of pattern recognition and formation, the DSS can solve the following tasks:

pattern recognition and formation;

receiving and storing of knowledge (empirically found regular relationships of images and influences on the object of control);

assessment of the quality characteristics of the images;

decision making (choice of influences).

Consider the scheme proposed by the authors in their work (fig. 1.) [2].

Fig. 1. The IDSS scheme of CP identification

If to look at the proposed scheme in more detail, the following points can be noted:

"User". The following categories may act as a user:

experts in the field of cultural values and art objects;

museums;

educational institutions;

organizations for the creation and filling the catalogs of cultural values and art objects;

state bodies for control over cultural values and objects of art;

customs;

average users of the system (guests of the resource);

art critics;

historians;

auctions of cultural property and art objects;

other systems and information resources.

"Software and hardware system". In this case, this part of the system can be a separate, functionally logical unit or module (optional for this type of system). The system can be an automated scanning/photographing system or a system that performs measurements of various parameters of cultural objects and transmits for processing and analysis to the "Query and Answer Block" through the appropriate interfaces (software and hardware).

"Expert". In the given scheme the expert can be both in the role of the user-consultant in the process of the development of the system itself, and in the role of the person who forms/fills the "Data Warehouse" according to the established rules and data formats. An expert group may also act as an expert, since the effectiveness of the system (identification and conceptualization stages) is largely determined by the successful formation of a reputable expert group and obtaining the qualitative knowledge from the group that forms the basis of any IDSS.

The essence of the process of knowledge discovery lies in conducting by experts an intuitive-logical analysis of the problem area with a quantitative assessment of the judgments they formulate. In this scheme, experts can determine the objects and concepts of the subject area (goals, decisions, alternative situations, etc.), measure the characteristics (probability of event completion, coefficients of significance of goals, decisions preferences, etc.) [4].

“Query and Answer Block”. This software or hardware is a subsystem with a user interface by means of which a query is generated to the data warehouse for analysis. This part can be made in the form of a dialog where the user select/set the criteria of the object. After the request is made in the installed view, the subsystem access the data warehouse exactly where information about the object being analyzed is collected. But the main task of this unit is to analyze the data received and generate a response to the user's request.

“Data Warehouse” and “Data Warehouse Formation Unit”. In their work [2], the authors considered the work of these subsystem and noted that special attention should be paid to the issues of justifying the criteria for data warehousing and data consolidation. The issue of creating and filling data warehouses is considered in the work [3]. Consideration should also be given to developing methods and models for evaluating the quality of third-party data when filling the data warehouse.

“External Data Sources” (“Databases (DB)”, “Knowledge Bases (KB)” and “Catalogs”). In this scheme these blocks are conditionally united into one whole and represent the already existing CP object bases, for example these could be:

museum catalogs or databases;

descriptions or databases of Ukrainian and foreign organizations for the control of cultural values and

objects of art;

catalogs or databases of the customs services and the Ministry of Internal Affairs regarding the

description of cultural values and objects of art;

catalogs of auctions of cultural property and art objects;

works of art historians and historians;

DBs or DBs of other systems and information resources.

Conclusions and prospects for further research. The authors analyzed the problems associated with the development of an intelligent expert system for the identification and analysis of cultural value objects, as well as the solutions used. The obrained results allow us to draw the following conclusions:

The problem area is relatively new and not well understood field of research,

The process of identification and analysis of cultural objects, and the objects themselves are not sufficiently described, since cultural values are various objects of different shapes, structures and materials with specific characteristics [1, 5]. It should also be noted that cultural values are constantly created, which makes the process of data generation difficult.

Creating universal models of knowledge bases for intelligent decision support systems remains an open and relevant issue.

Existing models of presenting knowledge and creating databases and knowledge in this field have disadvantages.

Creating a universal intelligent decision support system for cultural values identification is a big complex and extremely time-consuming task that, according to the authors, can be divided into separated tasks that can be solved within the framework of individual scientific research and practical implementation and use in the problems of similar class.

For example, if to refer to the scheme presented in Fig. 1, then the formation and work organization of each block is a separated task.

And if in the case of the block “User”, the task can be solved quite simply and quickly, since the principles of user work with the system already exist and are sufficiently described and implemented in systems of this type, then others present more complex tasks.

For example, the block “Hardware and Software System”, which is not obligatory for this type of systems, but during the rapid development of modern information technologies and variety of software and hardware, this system can greatly facilitate and make the process of identification more effective. For example, performing in an automatic mode high precision measurements of certain parameters/characteristics of objects (size, weight, barcode scanning, high resolution scanning, etc.), which in turn can reduce the time and minimize the effect of human factors. Thus, the development of software and hardware systems of this type that can be included in the IDSS is difficult and more practical engineering work, so the authors of the work have not considered it in detail but it can be considered as a further perspective development and improvement of the system as a whole.

There are various approaches in the organization of work and filling of "Data Warehouse", the development of algorithms and the organization of principles of data warehouse formation, which represent a separate scientific problem of data analysis and processing of large data sets [2]. But it can be noted that for certain systems, storage can be created in different forms and formats, depending on the complexity of the system and requirements for its functionality. For example, it may look like a simple greenhouse, relational database or OLAP cube.

The performed analysis also allowed the authors to specify the task and to determine a possible plan for its solution.

The principles of organizing the work of the “Block of query and answer formation” in the author's paper are of great scientific interest. Since the main purpose of this block is to analyze the obtained data and to formulate a response to the user's request, then in the opinion of the authors, special attention should be given to such tasks as:

Analysis of knowledge representation models and development of a new model for the given problem area if necessary.

Development of a form and a method of querying the data warehouse in accordance with the tasks to be solved and in the context of the specific characteristics of the objects.

Development of an algorithm for searching and sampling data in the data warehouse.

Development of a model for processing and analysis of data obtained from search and sampling (item 3).

Development of criteria for evaluation of the reliability of the results and formation of the answers that the system produces as a result of the analysis.

Approval of the developed methods and models of organization and processing of data and knowledge in the intelligent decision support system for cultural values identification on the example of some certain type of objects of cultural values.

Список бібліографічних посилань

1. А.А. Мартиненко, Б.І. Мороз, І.Г. Гуліна «Інтелектуальна система підтримки прийняття рішень ідентифікації культурних цінностей». IV Всеукраїнська науково-практична конференція «Перспективні напрямки сучасної електроніки, інформаційних і комп'ютерних систем (MEICS-2019)». Дніпро, 27-29 листопада 2019 р.

2. А.А. Мартиненко, Б.І. Мороз, І.Г. Гуліна «Сховища даних системи підтримки прийняття рішень ідентифікації культурних цінностей» XV міжнародна конференція з проблем використання інформаційних технологій в освіті, науці та промисловості. Дніпро, 5-6 грудня 2019 р.

3. Орешков В.И. Паклин Н.Б. «Консолидация данных -- ключевые понятия» режим доступу https://www.cfm.ru/itm/olap/cons.shtml.

4. Хабаров С.П. «Интеллектуальные информационные системы», режим доступу http://www.habarov.spb.ru/new_es/index.htm.

5. Закон України "Про вивезення, ввезення та повернення культурних цінностей", режим доступу https://zakon.rada.gov.ua/laws/show/ru/1068-14/.

6. Ульяновська Ю.В. «Моделі та методи обробки даних в єдиній автоматизованій інформаційній системі митної служби», автореферат дисертації на здобуття наукового ступеня кандидата технічних наук, 05.13.06 - автоматизовані системи управління та прогресивні інформаційні технології, Харків - 2005.

7. Е.А. Ганцева, В.А. Каладзе, А.М. Поляков. «Формирование экспертного вывода на основе теории свидетельств» режим доступу https://cyberleninka.ru/article/n/formiгovanie-ekspertnogo-vyvoda-na-osnove-teorii-svidetelstv/viewer.

8. А.С. Луцаевский, Т.С. Чайникова «Параметрическая идентификация правила Инагаки с помощью генетических алгоритмов». Збірник наукових праць Харківського університету Повітряних Сил ім. І. Кожедуба, випуск 3(15), 2007.

9. Джарратано Джозеф, Райли Гари. «Экспертные системы: принципы разработки и программирование», 4-е издание.: Пер. с англ. -- М.: ООО “И.Д. Вильямс”, 2007. -- 1152 с.

10. Кононюк А.Ю. «Нейроні мережі і генетичні алгоритми» - К.: «Корнійчук». 2008. - 446 с.

11. Демиденко М.А. «Системи підтримки прийняття рішень»: навч. посіб. / М.А. Демиденко; Нац. гірн. ун -т. -- Електрон. текст. дані. - Д.: 2016. - 104 с.

References

1. A.A. Martynenko, BI Moroz, I.G. Hulina "An Intelligent Decision Support System for Identifying Cultural Property". IV All-Ukrainian Scientific-Practical Conference "Promising directions of modern electronics, information and computer systems (MEICS-2019)". Dnipro, November 27-29, 2019.

2. A.A. Martynenko, BI Moroz, I.G. Hulina "Data Warehouse of Decision Support System for Identification of Cultural Property" XV International Conference on Information Technology Use in Education, Science and Industry. Dnipro, December 5-6, 2019.

3. Oreshkov V.I. Paklin N.B. “Data Consolidation - Key Concepts” access mode https://www.cfm.ru/itm/olap/cons.shtml.

4. Khabarov S.P. “Intelligent Information Systems”, access mode http://www.habarov.spb.ru/new_es/index.htm.

5. Law of Ukraine "On Exports, Imports and Returns of Cultural Property", access regime https://zakon.rada.gov.ua/laws/show/en/1068-14/.

6. Ulyanovskaya Yu.V. «Models and methods of data processing in a single automated information system of the customs service», dissertation author's abstract on obtaining a scientific degree of Candidate of Technical Sciences, 05.13.06 - automated control systems and progressive information technologies, Kharkiv - 2005.

7. E.A. Gantseva, V.A. Koladze, A.M. Poles. "Formation of expert opinion based on the theory of evidence" access mode https://cyberleninka.ru/article/n/formirovanie-ekspertnogo-vyvoda-na-osnove-teorii-svidetelstv/viewer.

8. A.S. Lutseevsky, TS Chaynikov "Parametric identification of the Inagaki rule using genetic algorithms". Collection of scientific works of Kharkov University of the Air Force. I. Kozheduba, Issue 3 (15), 2007.

9. Jarratano Joseph, Riley Gary. "Expert Systems: Principles of Development and Programming", 4th Edition.: Transl. with English. - M.: LLC “ID. Williams”, 2007. - 1152 p.

10. Kononyuk A.Yu. Neural Networks and Genetic Algorithms - K.: Korniychuk. 2008. - 446 p.

11. Demidenko M.A. "Decision Support Systems": Tutorial. tool. / M.A. Demidenko; Nat. mines Univ. -- Electron. text. data. - D.: 2016. - 104 p.

Размещено на Allbest.ru