Tool: AISSA

Visual support through slides is a key element for the success of oral presentations in the university environment. However, providing detailed and formative feedback on this material during the preparation phase poses a challenge for teaching staff due to overcrowded classrooms and time constraints. This Bachelor's Thesis presents the design and implementation of AISSA (AI-based Student Slides Analysis tool), an interactive web tool that automates the delivery of formative feedback on academic presentations using Generative Artificial Intelligence. Developed with a web architecture and a hybrid database model, the system extracts the features of each slide from the .pptx file and analyzes them by connecting to Large Language Models (LLMs). Using restrictive prompt engineering, the AI provides structured evaluations that strictly follow the teaching rubrics. Through this workflow, students upload preliminary versions of their work and view a breakdown of their performance with specific comments on interactive dashboards, fostering autonomous learning so that students can improve their slides prior to the final presentation.

Related Articles

• Becerra, A., Gomez, D., & Cobos, R. (2026). AISSA: Implementation and Deployment of an AI-based Student Slides Analysis tool for Academic Presentations. arXiv preprint arXiv:2605.04729.

Authors: Alvaro Becerra, Ruth Cobos, Diego Gómez