Fuzzy knowledge extraction in the development of expert predictive diagnostic systems

Suzdaltsev V.A., Suzdaltsev I.V., Tarhavova E.G.
 pdf (333K)

Expert systems imitate professional experience and thinking process of a specialist to solve problems in various subject areas. An example of the problem that it is expedient to solve with the help of the expert system is the problem of forming a diagnosis that arises in technology, medicine, and other fields. When solving the diagnostic problem, it is necessary to anticipate the occurrence of critical or emergency situations in the future. They are situations, which require timely intervention of specialists to prevent critical aftermath. Fuzzy sets theory provides one of the approaches to solve ill-structured problems, diagnosis-making problems belong to which. The theory of fuzzy sets provides means for the formation of linguistic variables, which are helpful to describe the modeled process. Linguistic variables are elements of fuzzy logical rules that simulate the reasoning of professionals in the subject area. To develop fuzzy rules it is necessary to resort to a survey of experts. Knowledge engineers use experts’ opinion to evaluate correspondence between a typical current situation and the risk of emergency in the future. The result of knowledge extraction is a description of linguistic variables that includes a combination of signs. Experts are involved in the survey to create descriptions of linguistic variables and present a set of simulated situations.When building such systems, the main problem of the survey is laboriousness of the process of interaction of knowledge engineers with experts. The main reason is the multiplicity of questions the expert must answer. The paper represents reasoning of the method, which allows knowledge engineer to reduce the number of questions posed to the expert. The paper describes the experiments carried out to test the applicability of the proposed method. An expert system for predicting risk groups for neonatal pathologies and pregnancy pathologies using the proposed knowledge extraction method confirms the feasibility of the proposed approach.

Keywords: expert system, knowledge acquisition, linguistic variable, membership degree, fuzzy rule
Citation in English: Suzdaltsev V.A., Suzdaltsev I.V., Tarhavova E.G. Fuzzy knowledge extraction in the development of expert predictive diagnostic systems // Computer Research and Modeling, 2022, vol. 14, no. 6, pp. 1395-1408
Citation in English: Suzdaltsev V.A., Suzdaltsev I.V., Tarhavova E.G. Fuzzy knowledge extraction in the development of expert predictive diagnostic systems // Computer Research and Modeling, 2022, vol. 14, no. 6, pp. 1395-1408
DOI: 10.20537/2076-7633-2022-14-6-1395-1408
URL: http://crm-en.ics.org.ru/journal/article/3283/
Creative Commons License This work is licensed under a Creative Commons Attribution-NoDerivs 3.0 Unported License.

Indexed in Scopus

Full-text version of the journal is also available on the web site of the scientific electronic library eLIBRARY.RU

The journal is included in the Russian Science Citation Index

The journal is included in the RSCI

International Interdisciplinary Conference "Mathematics. Computing. Education"

Copyright © 2009–2024 Institute of Computer Science