author: Tim Blok
title: ZITA - A self learning tutoring assistant
keywords: software engineering, education, static analysis, machine learning
topics: Case studies and Applications , Software Technology , Other
committee: Doina Bucur ,
Ansgar Fehnker
started: November 2018
end: May 2019

Abstract

Static analysis tools can be helpful to immediately get feedback on the code you write. However, if you want to spot very specific errors (to use in a programming course, for example), you'd have to write your own rules by hand. We use machine learning on ASTs to automatically create these rules, and check how well they perform compared to normal static analysis tools