A project at the Universitat Jaume I is studying the use of artificial intelligence to enhance education and social welfare

The use of technology to improve quality of life and education is one of the great challenges of the 21st century. In this context, the project led by Rafael Berlanga, professor in the Department of Computer Languages and Systems, and Lledó Museros Cabedo, senior lecturer in the Department of Computer Science and Engineering at the Universitat Jaume I, investigates how to apply explainable artificial intelligence (XAI) to promote healthy habits, enhance cognitive abilities and foster social inclusion.

This research, included in the XAI4SOC-UJI project and funded by the 2021 State Scientific Research Plan, responds to the call of the United Nations General Assembly's Decade of Healthy Ageing (2021-2030), which promotes the empowerment of older people to remain active citizens, while educating young people in values and habits that improve well-being. In this context, XAI4SOC-UJI combines advanced technologies, such as cognitive video games and conversational systems, to help adolescents in particular to develop spatial reasoning skills and to recognise and manage emotions.

Among the advances achieved so far, the project has developed new methodologies that allow AI-based systems to explain their decisions in a transparent manner tailored to each individual. Specifically, this includes common sense reasoning techniques, the use of ontologies (structures that help AI organise and relate concepts within a specific topic) to better understand context and generate clear explanations, as well as multimodal interfaces that offer accessible explanations in various formats (text, audio, image). In addition, pilot tests have already been conducted with students and older adults, demonstrating the usefulness of these technologies in improving cognitive abilities and promoting well-being. The project also involves collaboration with secondary schools and the spin-off SemanticBots for the development of conversational systems, and has already implemented prototypes such as XAI4SOC-cogGames and XAI4SOC-emoBook.

However, challenges remain, including the validation of these technologies in real-world settings and adaptation to new European regulations on AI, such as the AI Act and the Z-Inspection® guidelines to ensure trustworthy AI systems. In addition, work is underway to expand the open-source library that integrates all the algorithms developed, as well as to evolve the XAI4SOC-emoBook and XAI4SOC-cogGames platforms.

Rafael Berlanga is a professor at the Universitat Jaume I and coordinates the TKBG - Temporal Knowledge Bases research group, founded in 1997 and made up of researchers with extensive experience in artificial intelligence, text and data mining, and social network analysis. Alongside him, Lledó Museros Cabedo, senior lecturer in the Department of Computer Science and Engineering, an expert in qualitative reasoning, robotics and cognitive processes, is participating as co-principal investigator on the project. The group is currently working on areas such as explainable AI, large language models (LLM) and cognitive systems applied to sustainability and well-being, and offers consulting services in conversational systems and cognitive modelling. The group's entire track record and lines of work can be found at: https://krono.act.uji.es/.

With this research, the Universitat Jaume I reaffirms its commitment to applied research and knowledge transfer, contributing to the development of digital tools that improve well-being and promote a more inclusive society that is better prepared for the challenges of the future.

This research is part of the PID2021-123152OB-C22 project, funded by MICIU/AEI/10.13039/501100011033 and FEDER/EU

Papers related to the project: https://repositori.uji.es/search?query=PID2021-123152OB-C22