КОГНІТИВНО-ОРІЄНТОВАНА UML-ФОРМАЛІЗАЦІЯ ОСВІТНІХ ПРОГРАМ З МАШИННИМ ІНТЕЛЕКТОМ.
Saved in:
| Title: | КОГНІТИВНО-ОРІЄНТОВАНА UML-ФОРМАЛІЗАЦІЯ ОСВІТНІХ ПРОГРАМ З МАШИННИМ ІНТЕЛЕКТОМ. |
|---|---|
| Alternate Title: | COGNITIVELY-ORIENTED UML FORMALIZATION OF EDUCATIONAL PROGRAMS INCORPORATING MACHINE INTELLIGENCE. |
| Authors: | Лях, І. М.1 igor.lyah@uzhnu.edu.ua, Пригара, М. П.1 mykhailo.prygara@uzhnu.edu.ua, Дитко, Т. В.1 taras.dytko@uzhnu.edu.ua, Коляно, Я. Ю.2 yaroslav.y.koliano@lpnu.ua |
| Source: | Informatics & Mathematical Methods in Simulation / Informatika ta Matematičnì Metodi v Modelûvannì. 2026, Vol. 16 Issue 3, p515-527. 13p. |
| Subjects: | Artificial intelligence in education, Cognitive load, Instructional systems, Educational technology, Artificial intelligence, Educational programs |
| Abstract: | This article presents the results of the development and justification of a cognition-oriented UML model for formalising the structural-logical diagrams of educational programs, integrated with machine intelligence technologies. The relevance of the research stems from the rapid transformation of the EdTech environment, accompanied by the rise of ‘digital obesity’ and a critical increase in the cognitive load on learners. The fragmentation of educational content and the lack of dynamic adaptation mechanisms lead to a decline in motivation and an increased risk of academic dropout. The proposed solution is based on the creation of a student’s ‘cognitive twin’, formalized via a UML metamodel, which includes the EducationalProgram, CognitiveState and LearningPathOptimizer classes. This allows the learning process to be transformed from a linear sequence of modules into a dynamic ecosystem capable of real-time self-correction. The methodological basis of the work is the synergy of principles of software engineering, cognitive psychology and deep learning methods. To identify the user’s psychophysiological state, a multimodal approach was applied, combining subjective metrics from the multidimensional NASA-TLX (6D) scale with objective indicators from computer vision and semantic analysis. The use of the MiniLM-L6-v2 model to generate 384-dimensional content embeddings enabled the mathematical determination of the ‘cognitive cost’ of switching between topics, whilst the implementation of the YOLOv8n algorithm ensured highly accurate attention monitoring through ROI detection and the generation of dynamic heatmaps. A key technological outcome of the research is the development of an Adaptive Behavior Protocol (ABP), which implements a closed-loop control system (real-time cognitive loop). Thanks to the use of SignalR technologies and the ONNX Runtime environment, critically low data processing latency has been achieved – less than 500 ms – making the content adaptation process seamless and imperceptible to the user. The protocol covers a five-stage pipeline: from event detection in the LMS to the execution of an adaptive action via LTI Deep Linking. Experimental testing on a sample of 120 students demonstrated high classification reliability for four cognitive states (Flow, Overload, Boredom, Frustration) with an average F1-score of over 0,90. The paper pays particular attention to issues of cybersecurity and interoperability. Risks associated with the compromise of API keys and the manipulation of cognitive data records (CDR tampering) have been analyzed. The system architecture, built on the LTI 1.3 and xAPI standards, ensures the security of personalized profiles and the ability to transfer them between heterogeneous educational platforms. The research results confirm that the implementation of cognitiveoriented formalization allows for a 26% increase in course completion rates and a 19% reduction in subjective overload, creating a new paradigm for intelligent and secure digital education. [ABSTRACT FROM AUTHOR] |
| Copyright of Informatics & Mathematical Methods in Simulation / Informatika ta Matematičnì Metodi v Modelûvannì is the property of Odessa Polytechnic University and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| FullText | Links: – Type: pdflink Text: Availability: 0 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 194922489 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: КОГНІТИВНО-ОРІЄНТОВАНА UML-ФОРМАЛІЗАЦІЯ ОСВІТНІХ ПРОГРАМ З МАШИННИМ ІНТЕЛЕКТОМ. – Name: TitleAlt Label: Alternate Title Group: TiAlt Data: COGNITIVELY-ORIENTED UML FORMALIZATION OF EDUCATIONAL PROGRAMS INCORPORATING MACHINE INTELLIGENCE. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Лях%2C+І%2E+М%2E%22">Лях, І. М.</searchLink><relatesTo>1</relatesTo><i> igor.lyah@uzhnu.edu.ua</i><br /><searchLink fieldCode="AR" term="%22Пригара%2C+М%2E+П%2E%22">Пригара, М. П.</searchLink><relatesTo>1</relatesTo><i> mykhailo.prygara@uzhnu.edu.ua</i><br /><searchLink fieldCode="AR" term="%22Дитко%2C+Т%2E+В%2E%22">Дитко, Т. В.</searchLink><relatesTo>1</relatesTo><i> taras.dytko@uzhnu.edu.ua</i><br /><searchLink fieldCode="AR" term="%22Коляно%2C+Я%2E+Ю%2E%22">Коляно, Я. Ю.</searchLink><relatesTo>2</relatesTo><i> yaroslav.y.koliano@lpnu.ua</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Informatics+%26+Mathematical+Methods+in+Simulation+%2F+Informatika+ta+Matematičnì+Metodi+v+Modelûvannì%22">Informatics & Mathematical Methods in Simulation / Informatika ta Matematičnì Metodi v Modelûvannì</searchLink>. 2026, Vol. 16 Issue 3, p515-527. 13p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Artificial+intelligence+in+education%22">Artificial intelligence in education</searchLink><br /><searchLink fieldCode="DE" term="%22Cognitive+load%22">Cognitive load</searchLink><br /><searchLink fieldCode="DE" term="%22Instructional+systems%22">Instructional systems</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+technology%22">Educational technology</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+intelligence%22">Artificial intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+programs%22">Educational programs</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: This article presents the results of the development and justification of a cognition-oriented UML model for formalising the structural-logical diagrams of educational programs, integrated with machine intelligence technologies. The relevance of the research stems from the rapid transformation of the EdTech environment, accompanied by the rise of ‘digital obesity’ and a critical increase in the cognitive load on learners. The fragmentation of educational content and the lack of dynamic adaptation mechanisms lead to a decline in motivation and an increased risk of academic dropout. The proposed solution is based on the creation of a student’s ‘cognitive twin’, formalized via a UML metamodel, which includes the EducationalProgram, CognitiveState and LearningPathOptimizer classes. This allows the learning process to be transformed from a linear sequence of modules into a dynamic ecosystem capable of real-time self-correction. The methodological basis of the work is the synergy of principles of software engineering, cognitive psychology and deep learning methods. To identify the user’s psychophysiological state, a multimodal approach was applied, combining subjective metrics from the multidimensional NASA-TLX (6D) scale with objective indicators from computer vision and semantic analysis. The use of the MiniLM-L6-v2 model to generate 384-dimensional content embeddings enabled the mathematical determination of the ‘cognitive cost’ of switching between topics, whilst the implementation of the YOLOv8n algorithm ensured highly accurate attention monitoring through ROI detection and the generation of dynamic heatmaps. A key technological outcome of the research is the development of an Adaptive Behavior Protocol (ABP), which implements a closed-loop control system (real-time cognitive loop). Thanks to the use of SignalR technologies and the ONNX Runtime environment, critically low data processing latency has been achieved – less than 500 ms – making the content adaptation process seamless and imperceptible to the user. The protocol covers a five-stage pipeline: from event detection in the LMS to the execution of an adaptive action via LTI Deep Linking. Experimental testing on a sample of 120 students demonstrated high classification reliability for four cognitive states (Flow, Overload, Boredom, Frustration) with an average F1-score of over 0,90. The paper pays particular attention to issues of cybersecurity and interoperability. Risks associated with the compromise of API keys and the manipulation of cognitive data records (CDR tampering) have been analyzed. The system architecture, built on the LTI 1.3 and xAPI standards, ensures the security of personalized profiles and the ability to transfer them between heterogeneous educational platforms. The research results confirm that the implementation of cognitiveoriented formalization allows for a 26% increase in course completion rates and a 19% reduction in subjective overload, creating a new paradigm for intelligent and secure digital education. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Informatics & Mathematical Methods in Simulation / Informatika ta Matematičnì Metodi v Modelûvannì is the property of Odessa Polytechnic University and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=194922489 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.15276/imms.v16.no3.515 Languages: – Code: ukr Text: Ukrainian PhysicalDescription: Pagination: PageCount: 13 StartPage: 515 Subjects: – SubjectFull: Artificial intelligence in education Type: general – SubjectFull: Cognitive load Type: general – SubjectFull: Instructional systems Type: general – SubjectFull: Educational technology Type: general – SubjectFull: Artificial intelligence Type: general – SubjectFull: Educational programs Type: general Titles: – TitleFull: КОГНІТИВНО-ОРІЄНТОВАНА UML-ФОРМАЛІЗАЦІЯ ОСВІТНІХ ПРОГРАМ З МАШИННИМ ІНТЕЛЕКТОМ. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Лях, І. М. – PersonEntity: Name: NameFull: Пригара, М. П. – PersonEntity: Name: NameFull: Дитко, Т. В. – PersonEntity: Name: NameFull: Коляно, Я. Ю. IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 07 Text: 2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 22235744 Numbering: – Type: volume Value: 16 – Type: issue Value: 3 Titles: – TitleFull: Informatics & Mathematical Methods in Simulation / Informatika ta Matematičnì Metodi v Modelûvannì Type: main |
| ResultId | 1 |
