DSLAB-TI: teaching innovation
The Data Science Lab for Teaching Innovation (DSLAB-TI) pioneers the development of active methodologies, gamified educational tools, generative AI assistants, and automated assessment systems to transform data science education.
Featured teaching initiatives
Explore our open-source virtual tutor ViLT and the ¡CONECTA! board game.
Empowering education through Data Science
Data Science Lab for Teaching Innovation (DSLAB-TI) is a recognized teaching innovation group whose primary mission is to leverage data science, machine learning, and interactive visual analytics to elevate university pedagogy.
By collecting structured student interaction metrics through tailored online questionnaires, the group builds machine learning models capable of forecasting student evaluations based on partial coursework results. These predictive models identify vulnerable student profiles early, enabling timely, targeted pedagogical interventions.
Results and predictive indicators are delivered through interactive, role-specific dashboards designed for students and educators. Once validated, models are deployed into production across subsequent academic cycles.
Our innovation actions are inherently scalable and replicable across undergraduate and master's programs at URJC and beyond, requiring only subject-specific questionnaire configurations to unlock predictive learning analytics.
Strategic objectives
- Promote continuous improvement of teaching and learning processes.
- Integrate active methodologies that foster skill acquisition and student motivation.
- Encourage the effective use of the Virtual Classroom by both students and faculty.
- Assess, support, and disseminate innovative educational practices to build a quality benchmark for university educators.
Key action lines
- Optimize student performance and reduce academic failure and dropout rates.
- Promote the use of modern digital technologies and the Virtual Classroom.
- Implement generative AI, predictive modeling, and Big Data analytics applied to instructional enhancement.
- Develop and deploy competency-based assessment models.
- Analyze and improve academic outcomes in blended and distance learning programs.
- Provide personalized academic mentoring and data-driven guidance to students.
- Champion Open Education: creation, adoption, and dissemination of open educational resources.
- Launch educational innovation initiatives dedicated to inclusion and gender equality.
Teaching innovation projects
Competitive teaching innovation grants (Convocatorias de Proyectos de Innovación Docente • URJC).
ViLT: Virtual Intelligent Tutor for supporting URJC students
ViLT proposes an intelligent tutoring system powered by AI and Large Language Models (LLMs) to support students across degree courses. It integrates a dynamic chatbot, contextual vector embeddings of course syllabi, and feedback loops to detect conceptual drift. Built on a modular REST API with dedicated analytics dashboards for educators.
DSExams: Massive and automated generation of multipurpose randomized questionnaires
An automated assessment framework for generating randomized exam questionnaires using the R package {exams} formatted in LaTeX. Enables infinite question variations using probabilistic generation, integrated directly into Moodle with automated grading and detailed student feedback.
¡CONECTA! Educational board game
A collaborative educational board game designed to immerse students in telecommunications engineering and data science project lifecycles. Students work in teams to tackle real-world challenges through problem formulation, infrastructure selection, AI modeling, and practical deployment.
Featured publications in educational innovation
Articles, conference proceedings, and interactive teaching tools authored by DSLAB-TI researchers.
Using an LLM-based framework to analyze student performance
A. Fernández-Isabel, C. Lancho, I. Martín de Diego, Á. Udías, A. Alonso-Ayuso, C. Alfaro, E. L. Cano, F. Ortega, J. Gómez, J. M. Moguerza, M. J. Algar
Explores a framework powered by LLMs to support educational mentoring and personalized tutoring, particularly for students less likely to seek face-to-face assistance.
ChatGPT's performance in university admissions tests in mathematics
Á. Udías, A. Alonso-Ayuso, C. Alfaro, M. J. Algar, M. Cuesta, A. Fernández-Isabel, J. Gómez, C. Lancho, E. L. Cano, I. Martín de Diego, F. Ortega
Evaluates ChatGPT-4.0 on Spanish university entrance exams in math, demonstrating strong competence in probability and statistics while analyzing limitations in algebra.
Tutor virtual inteligente basado en modelos generativos del lenguaje
A. Fernández-Isabel, I. Martín de Diego, E. L. Cano, M. Cuesta, C. Lancho
Presents a fine-tuned LLM tutor tailored to university courses using syllabus materials, offering personalized explanations, exercises, and interactive continuous assistance.
Empowering Academic Performance: Data-Driven Mentoring for Personalized Education
M. Cuesta, C. Lancho, I. Martín de Diego, A. Fernández-Isabel, E. L. Cano, J. M. Moguerza
Introduces a mentoring framework combining learning analytics, questionnaires, and predictive machine learning to forecast academic needs and tailor interventions.
DSExams: Massive and automated generation of randomized multipurpose questionnaires
E. L. Cano, M. Cuesta, C. Lancho, C. Alfaro, M. J. Algar, A. Alonso-Ayuso, A. Fernández-Isabel, J. Gómez, I. Martín de Diego, J. M. Moguerza, F. Ortega, Á. Udías
A scalable automated questionnaire pipeline for assessment and self-evaluation in Data Science, leveraging randomization to prevent rote memorization.
DSGame Kids: Learning Data Science projects through a storytelling board game
A. Fernández-Isabel, I. Martín de Diego, M. Cuesta, C. Lancho, J. M. Moguerza
A physical board game teaching the complete data science lifecycle to computer science students through cooperative storytelling, teamwork, and mission milestones.
Concurso de Monty Hall: una aplicación interactiva con R para explicar probabilidad
E. L. Cano
A gamified Shiny application that explains conditional probability and Bayes' rule by simulating the famous Monty Hall dilemma interactively.
Mejorando la comprensión de conceptos estadísticos mediante aplicaciones interactivas innovadoras
E. L. Cano, M. J. Algar, A. Alonso-Ayuso, J. M. Moguerza, F. Ortega
Interactive Shiny apps that visually simulate frequentist probability, sampling distributions, and central limit theorems to boost student engagement in STEM courses.