Home A2C Workshop Current Meeting of A2C Workshop
Current Meeting of A2C Workshop

International Workshop

Advances & Challenges in Computing (A2C)

A Regular meeting
Date: 20 October 2021 (Wednesday)
Place: online (a link will be put in invitation)
Time: 18:00-20:00 (East European time, Kyiv)

AGENDA:

1. Speaker: Yelyzaveta Hnatchuk

PhD, Associate Professor
Computer Engineering & Information Systems Department
Khmelnytskyi National University


E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Topic: Intelligent Information Technology for Supporting the Medical Decision-Making Considering the Legal Basis

Abstract: The decision support system (DSS) and information technology (IT) in the field of medical law are designed to provide the possibility of free and automated verification of all significant conditions of the contract and provide recommendations for the conclusion of the contract or not. Modern decisions for support the adoption of medical decisions on legal basis have shown that none of the known decisions does meet all the necessary criteria in the complex. Therefore, an urgent problem for Ukraine is the development and implementation of intelligent information technology to support medical decision-making, taking into account the legal basis, which is the purpose of this study. The proposed intelligent information technology to support medical decision-making based on legal basis provides support for decision-making on the possibility of using reproductive technologies (the possibility of surrogacy and/or in vitro fertilization), the possibility of donation and transplantation, the possibility of concluding contracts for therapeutic services, contracts for dental services and general contracts for medical services. In addition, intelligent information technology to support medical decision-making based on the legal basis automates the semantic parsing of contracts and draws conclusions about the possibility or impossibility of concluding a contract, as well as provides a request stating the reasons for impossibility to conclude a contract (for example, indicating missing essential conditions), if it was concluded that such a contract cannot be concluded.

Resume: In 2003, Yelyzaveta Hnatchuk graduated with honour from Khmelnitskyi National University. In 2008, she defended the PhD Thesis. Since 2003, she has been working at Khmelnitskyi National University. She has more than 50 scientific publications. Research interests include hybrid systems of computational intelligence: adaptive, neuro-, fuzzy-, real-time systems, including problems connected with control, identification, and forecasting, diagnostics, fault detection in technical and medical objects. She is the IntelITSIS workshops organizing committee member, member of 1 editorial board.

2. Speaker: Eduard Manziuk

 

 


(PhD, Khmelnytskyi National University, Ukraine)

ORCID ID: https://orcid.org/0000-0002-7310-2126

E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Topic: Theoretical and applied grounds of human-oriented information technologies on the principles of ethics and trustworthy artificial intelligence
Presentation in Ukrainian

Abstract: The widespread dissemination and application of artificial intelligence (AI) systems requires the development of formalized approaches and the construction of basic principles of functioning of the subject areas of the use of AI. This need is realized in the development of recommendations, regulations, and standards to maximize the benefits from the use of AI and minimize the possible risks. The regulatory framework is built on a human-centric basis. According to the developed standards should form the basis for further activities aimed at the application of AI and be applicable at all stages of the creation of practical solutions. Therefore, an important step is to formalize the requirements, principles and provisions of legal and ethical norms in the form of practical templates of practical application. With this method, the study developed models and ontologies of the standardized concept of trustworthiness for AI. This allowed to determine the basic concepts, which allow forming a position of trust, are a substantial part of the concept of AI trustworthiness, determine the necessity of its existence and pose a threat to it. On the basis of the subject domain ontology the models were developed and further decomposition of structural meaningful concepts was performed. Further defined the characteristics of the concept of trust formation. Machine learning methods have been proposed that allow dividing into interpreted and non-interpreted decisions within the framework of the concept of trust. Among the non-interpreted decision a method for improving the values of the target quality function using the grouping of an ensemble of models by the correlation of the decisions made is proposed. Also proposed a method for data features regularization in order to identify atypical features and feature outliers in the non-interpreted decisions based on the grouping of decisions in a single-class classification. Thus developed the model of confidence in AI allowed to determine the basic requirements in accordance with which were developed methods of obtaining decisions based on ensembles with maximization of the target function and the method of analysis of transient data features.



3. Information about research projects and conferences: Anatoliy Sachenko

 

Doctor of Technical Sciences, Professor, scientific advisor of Research Institute for Intelligent Computer Systems.

E-mail:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it