PORTALE DELLA DIDATTICA

Ricerca CERCA
  KEYWORD

Designing AI-Mediated Intelligent Tutor Creation Tools for Educators

keywords ARTIFICIAL INTELLIGENCE, EDUCATION, END-USER DEVELOPMENT, HUMAN AI INTERACTION, HUMAN COMPUTER INTERACTION, INTELLIGENT TUTORING SYSTEMS, PERSONALISATION

Reference persons LUIGI DE RUSSIS

External reference persons Tommaso Calò

Research Groups DAUIN - GR-10 - Intelligent and Interactive Systems - ELITE

Thesis type EXPERIMENTAL

Description The potential of AI to revolutionize user interface development is significant, with recent advances enabling natural language-based creation of complex interfaces. In educational technology, Intelligent Tutoring Systems (ITSs) offer promising opportunities for personalized learning, yet creating these systems remains challenging for educators who lack technical expertise. While existing authoring tools attempt to address this gap, they often result in rigid interfaces or require significant technical knowledge, limiting their practical adoption.

This thesis sets out to explore how AI-mediated interfaces can support educators in creating intelligent tutors through natural language interaction and visual editing. By focusing on the usability aspects of AI-assisted development, we aim to understand how educators interact with and understand these novel creation tools. This research will contribute to the broader understanding of how to design intuitive interfaces for AI-assisted content creation tools.

In particular, the thesis will aim to:
1. Design and develop a prototype of the tool that allows educators to create basic intelligent tutors through natural language interaction, customize AI-generated tutor interfaces through visual editing, and preview tutors before deployment.
2. Evaluate the usability of the tool with educators, including think-aloud sessions with educators to identify interaction barriers, user studies comparing different interface approaches, and semi-structured interviews to understand educators' mental models
3. Provide design recommendations for natural language interfaces for tutor creation, visual editing tools for AI-generated tutors, interface elements that support educator understanding and control.

If satisfactory, the result of the thesis will be released as an open-source project.

See also  teachbench.pdf 

Required skills Experience with web development (JavaScript, React)
Knowledge of UI/UX design principles
Familiarity with usability testing methods
Basic understanding of AI/ML concepts
Interest in educational technology


Deadline 07/01/2026      PROPONI LA TUA CANDIDATURA