EKM 2025 Abstracts

Area 1 - Educational Knowledge Management

Full Papers

Paper Nr:	5
Title:	Ontology-Based Framework for Personalized Home-Based Rehabilitation in Cerebral Palsy Care
Authors:	Rahma Haouas Zahwanie, Lilia Cheniti-Belcadhi and Saoussen Layouni
Abstract:	In the domain of cerebral palsy rehabilitation, advances in machine learning and semantic technologies of-fer promising solutions to enhance treatment strategies. This paper focuses on developing an ontology-based framework to support rehabilitation programs for children with cerebral palsy, addressing the need for personalized, home-based exercise programs (HEP). These programs aim to improve recovery by enabling patients to engage in tailored exercises outside clinical settings. However, the effectiveness of HEP depends on accurate monitoring and feedback, as improper execution of exercises can hinder progress. To address this challenge, we propose an intelligent system framework that integrates ontology-driven knowledge representation to oversee rehabilitation programs. The system analyzes patient profiles and progress data, recommending a personalized rehabilitation plan consisting of targeted exercises supported by healthcare professionals. The ontology serves as the backbone of this framework, enabling semantic representation of rehabilitation concepts and facilitating the management and improvement of cerebral palsy treatment pathways. Furthermore, this approach enhances patient outcomes by providing structured, context-aware rehabilitation plans while promoting interoperability and knowledge sharing across healthcare systems. By embedding the ontology within the framework, we enable greater reusability, semantic comprehension, and adaptability to multilingual healthcare environments. This work highlights the critical role of ontologies in advancing rehabilitation strategies for cerebral palsy and improving access to high-quality, personalized care.
Download

Paper Nr:	6
Title:	Evaluating the Adherence of Synthetic Digital Educational Content to Kolb’s Learning Theory
Authors:	Lidiane Castro Silva, Eduardo Alves Lima Caldas, Marcos Vinícius de Freitas Borges, Adson Roberto Pontes Damasceno and Francisco Carlos de Mattos Brito Oliveira
Abstract:	The use of generative AI can be a powerful ally in combating dropout rates in online courses. This study explores the application of artificial intelligence (AI) to personalize educational content, aligning texts with the learning styles identified by David Kolb (Converging, Diverging, Assimilating, and Accommodating). This research proposes a generative AI algorithm capable of creating texts tailored to these styles, specifically designed for distance education (DE), in which personalization is essential due to the diversity of learning profiles and the lack of face-to-face interaction. Besides the initial development of the generator programs, this study reports on the proposal of a methodology used to validate the quality of the generated texts and their adequacy to Kolb’s learning styles. The methodology was applied by six experts. The results show a high alignment of the texts with the Converging, Diverging, and Accommodating styles (100% on the Content Validity Index), with room for improvement in the Assimilating style (83%). The research highlights the technical feasibility of the proposed approach, both from the perspective of generative AI and the methodology for certifying the quality of synthetically generated material.
Download

Paper Nr:	7
Title:	Managing Data Heterogeneity for Ontology-Driven Models: Application to Gamified E-Learning Contexts
Authors:	Yara Gomaa, Christine Lahoud, Marie-Hélène Abel and Sherin Moussa
Abstract:	Data heterogeneity within gamified e-learning systems exposes a challenge for ontology-driven models, specifically ontology-based recommender systems. These systems can help teachers who are unfamiliar with gamification by offering personalized recommendations to gamify their pedagogical resources. Yet, developing such systems requires collecting and integrating diverse data about users, resources, and game elements, originating from multiple sources, like learning management systems and educational repositories, each with varying formats and inconsistent semantics. This paper proposes an approach to manage the complexities of collecting and preparing heterogeneous data for an ontology-driven model within gamified e-learning contexts. A full overview is provided on the data workflow, which consists of two main phases: (1) Data collection, which combines automated techniques through APIs and web scraping, and (2) Data Integration by means of mapping the collected data into our Teacher in Gamified e-learning Context (TGC) ontology to produce coherent and semantically enriched structure. The resulting data repository facilitates semantic queries, inference, and knowledge enrichment, overcoming challenges like cold-start scenarios and supporting the dynamic generation of personalized recommendations. This proposed approach aims to establish a robust approach that addresses the challenges of data heterogeneity, ensuring consistent and meaningful integration for ontology-based recommender systems in gamified e-learning contexts.
Download

Paper Nr:	8
Title:	Designing a Multimodal Interface for Text Simplification: A Case Study on Deepfakes and Misinformation Mitigation
Authors:	Francisco Lopes, Sílvia Araújo and Bruno Reynaud Sousa
Abstract:	The complexity of scientific literature often prevents non-experts from understanding science, limiting access to scientific knowledge for a broader audience. This paper presents a proof-of-concept system designed to enhance the accessibility of primary scientific literature through multimodal graphical abstracts (MGAs). The system simplifies complex content by integrating textual simplification, visual elements, and interactive components to support a broader range of learners. Using a case study on deepfake detection, the paper demonstrates the system’s functionalities. Key features include automated text simplification, audio narration, content summarization, and information visual representation. The example presented suggests the use of the generated MGA to improve AI literacy and misinformation resilience by promoting direct engagement with scientific content. Future work will focus on assessing the quality of text simplification and validating its effectiveness through user studies.
Download