Provas de CAT do aluno Pedro Miguel Orvalho Marques da Silva

Thesis Title: MENTOR : Automated Feedback for Introductory Programming Exercises

Data: 22 de julho de 2022

Hora: 14H00

Sessão de Zoom para Discussão da CAT :

https://videoconf-colibri.zoom.us/j/87636028565?pwd=OJDb6IR0gxuYJI7SkSOzpg49DFvu9O.1

Thesis Abstract:

The increasing demand for programming education has given rise to all kinds of online evaluations such as Massive Open Online Courses (MOOCs) focused on introductory programming assignments (IPAS), especially over the last years due to the coronavirus outbreak. As a consequence of a large number of enrolled students, one of the main challenges in these courses is to provide valuable and personalized feedback to students. This personalized feedback can be provided as a list of possible repairs to a student’s program. Semantic repairs are fixes that correct the program by checking its execution on a set of tests provided by the lecturer or by comparing the student’s program against a reference solution using program analysis. The main drawback of current state-of-the-art semantic repair tools is that these tools can only fix incorrect programs using correct implementations with exactly the same control flow graph as the faulty program. This thesis proposal presents MENTOR, a semantic program repair tool capable of Automated Feedback for Introductory Programming Exercises. MENTOR is a clustering-based program repair approach that takes advantage of students’ submissions from previous years to help repairing incorrect submissions. Our work will further the state-of-the-art with three main contributions. First, MENTOR will use invariant-based program clustering. Additionally, it will be able to repair an incorrect program using a correct implementation with a different control flow graph. Lastly, we will improve MaxSMT-based program repair encodings applied to IPAS.