Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: 2nd International Conference on Novel and Intelligent Digital Systems, NiDS 2022
Год издания: 2023
Идентификатор DOI: 10.1007/978-3-031-17601-2_1
Ключевые слова: cognitive visualization, decision explanation, decision making, intelligent tutoring systems, personalized learning space
Аннотация: The paper describes an approach to support methodological decision-making processes and explain them in the Intelligent Tutoring Systems (ITS), relying on cognitive visualization tools. The importance of XAI problems in the framework of automatic analysis of hypotheses concerning the interpretation of the learning situation is noteПоказать полностьюd. In order to combine the structural and functional approaches within a single logic of analysis, various graphic notations are considered. Vertical (for different scopes of learning situation coverage) and horizontal (for different aspects of analysis) transitions between decision-making levels are shown for the proposed decision-making model. A combination of the method of Cognitive Maps of Knowledge Diagnosis (CMKD, structural aspect) and the method of Unified Graphic Visualization of Activity (UGVA, functional aspect) is taken as the basis. The steps of the process of superimposing data from the digital educational footprint on invariant graphic notations are described. The importance of ensuring the isomorphism of displaying the parameters of the learning situation when switching between maps and images is noted. Visualizations of CMKDs and anthropomorphic UGVA images are presented to illustrate examples from the real learning process. The advantages of the cross-cutting visual support for both the decision-making process and the synthesis of explanatory texts are shown. In conclusion, the limitations of the approach are pointed out, and the directions for further research are outlined. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Журнал: Lecture Notes in Networks and Systems
Выпуск журнала: Vol. 556 LNNS
Номера страниц: 3-12
ISSN журнала: 23673370
Издатель: Springer Science and Business Media Deutschland GmbH