Master's thesis defense by Rebecca von der Grün on 20.12.2022 at 16:00 (online)

Master's thesis defense by Rebecca von der Grün on 20.12.2022 at 16:00 (online)

by Bettina Finzel -
Number of replies: 0

Dear all,

you are cordially invited to join the virtual master's thesis defense of Rebecca von der Grün on 20.12.22 at 16:00.

You can join via Zoom as follows:


Meeting-ID: 991 9134 2076
Passwort: 7Y^b@?

Rebecca von der Grün will defend her thesis "Algorithmic Debugging and Analogous Examples for Intelligent Tutoring. An Application to English Grammar Learning"


Online learning and computer programs for learning gain more and more popularity these days. Intelligent Tutoring Systems are one type of learning programs and have several advantages over traditional learning programs. Through their use of techniques from artificial intelligence, they can offer an individualized learning experience through generation of contents within the system and suitable adaption to the student’s needs. The goal of this thesis is the creation of an Intelligent Tutoring System for English grammar, i.e. the transformation of active sentences into passive. In generating the system, the algorithmic debugging approach was implemented to diagnose student errors. This approach takes the student solution and compares it step by step with an expert solution provided by the system itself. Further, the identified and classified errors are taken as a basis for the creation of analogous sentences. Through the presentation of these examples and their solutions, the students are given the opportunity to get aware of their misconceptions. The resulting system is a web application which generates random active sentences for the students to practise their conversion into passive. The system diagnoses correct and incorrect student answers and every time the system finds a mistake, analogous examples are presented and the student can rework their answer. However, due to the underlying lexicon and grammar only a limited number of different sentences and solutions can be generated. Further, to cover all aspects of the grammar domain, some extensions need to be added to the system.