RAGDWAREuz - Retrieval-Augmented Generation for a DataWarehouse Academic Retrieval Engine
RAGDWAREuz represents our work for the development of a virtual assistant using RAG (Retrieval Augmented Generation) to facilitate the exploitation of data cubes available in the data warehouse of the University of Zaragoza (DATUZ), offering contextual recommendations and interactive support. In addition, we attempt to create an educational environment with fictitious data cubes and practical exercises to strengthen training in data analysis and multidimensional modeling, which can be applied in the teaching of subjects on this topic. In the future, this work could be extended to encompass other types of data warehouses or institutions.
- User Guide
"Next-gEnerATion dAta Management to foster suitable Behaviors and the resilience of cItizens against modErN ChallEnges (NEAT-AMBIENCE)", funded by MICIU/AEI/10.13039/501100011033 (Agencia Estatal de Investigación). Leading researcher: Sergio Ilarri.
- This work belongs to the I+D+i project PID2020-113037RB-I00, funded by MICIU/AEI/10.13039/501100011033.
- Besides the previous project (NEAT-AMBIENCE), we also thank the support of the Departamento de Ciencia, Universidad y Sociedad del Conocimiento del Gobierno de Aragón (Government of Aragon: Group Reference T64_23R, COSMOS research group).
- Carlos de Vera Sanz (student at the University of Zaragoza) developed a related academic project (TFG - Trabajo Fin de Grado) titled "RAG to facilitate the use of DATUZ and learning about data warehouses" (as well as the associated prototypes), supervised by Sergio Ilarri and María Belén Gracia. Original repositories (last access: July 15, 2025): https://github.com/carloss4dv/RAGDWAREuz and https://github.com/carloss4dv/MDX_LEARN. As of July 15, 2025, the code contained here is available in those repositories.
