eStudy: a Novel Human-Centric AI-Based Platform to Customize Educational Contents on the Scholars' Needs

View/ Open
Author
Melloni, Daniele
Cramerotti, Sofia
Zambotti, Francesco
Rivelli, Nicoletta
De Martin, Nicoletta
Yeguas-Bolívar, Enrique
Calabrò, Giuseppe
Alcalde-Llergo, José M.
Zingoni, Andrea
Publisher
IEEEDate
2026Subject
Artificial intelligenceLarge language models
Retrieval-augmented generative
Personalized learning
Metacognition
Universal design for learning
METS:
Mostrar el registro METSPREMIS:
Mostrar el registro PREMISMetadata
Show full item recordAbstract
Personalized learning enhances student engagement, comprehension, and retention by adapting education to individual needs. Whereas AI offers powerful tools for delivering personalized content through data analysis and adaptive feedback, many current solutions lack a human-centric approach, risking reduced student autonomy and pedagogical misalignment. To address this, eStudy has been developed as an AI-powered educational platform focused on learner agency and inclusivity. It combines automation with reflective learning strategies, offering features such as content summarization, conceptual maps, guided text production, and multi-level text simplification. These tools support diverse learners, including those with specific learning needs or language barriers. Built on advanced large language models, eStudy uses retrieval-augmented generation and prompt engineering to deliver reliable and adaptive content. Grounded in Universal Design for Learning and student-centered principles, it promotes autonomy, metacognition, and meaningful interaction. Early testing shows positive results, positioning eStudy as a promising step toward ethical, effective AI in education.
Description
Embargado hasta 23/01/2028
