Skip to main content Skip to navigation

MOSAIC-F: A Framework for Enhancing Students' Oral Presentation Skills through Personalized Feedback

Project Overview

The document discusses the application of generative AI in education through the introduction of the MOSAIC-F framework, which integrates Multimodal Learning Analytics and collaborative assessments to enhance students' oral presentation skills. This innovative framework employs a combination of peer and professor assessments alongside multimodal data collection to generate personalized, AI-driven feedback. By amalgamating human evaluations with rich data inputs, MOSAIC-F aims to provide comprehensive and actionable insights that improve students' learning experiences and engagement. The findings suggest that this approach not only fosters individual skill development but also promotes collaborative learning environments, ultimately leading to better educational outcomes. Through its emphasis on personalized feedback and data-driven assessment, MOSAIC-F highlights the transformative potential of generative AI in shaping effective educational practices.

Key Applications

MOSAIC-F Framework

Context: Improving oral presentation skills of final-year engineering students at Universidad Autónoma de Madrid.

Implementation: A case study involving peer and professor assessments, multimodal data collection from various sensors, and AI-generated personalized feedback.

Outcomes: Enhanced oral presentation skills through detailed feedback, improved self-reflection, and actionable insights for future presentations.

Challenges: Subjectivity in evaluations, potential inconsistencies in feedback, and the challenge of integrating automated feedback with personal assessments.

Implementation Barriers

Technical Barrier

Integration of AI-generated feedback with human assessments can lead to inconsistencies or inaccuracies.

Proposed Solutions: Implement human oversight in the feedback process to ensure pedagogical validity and trustworthiness.

Perceptual Barrier

Students often perceive AI-generated feedback as less trustworthy and contextually sensitive compared to human feedback. Involving students and educators in the design of AI-based systems can enhance usability and acceptance.

Proposed Solutions: Involve students and educators in the design of AI-based systems to enhance usability and acceptance.

Project Team

Alvaro Becerra

Researcher

Daniel Andres

Researcher

Pablo Villegas

Researcher

Roberto Daza

Researcher

Ruth Cobos

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Alvaro Becerra, Daniel Andres, Pablo Villegas, Roberto Daza, Ruth Cobos

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies