Dear all,
you are cordially invited to join the master's thesis
defense of Jasmin Fritz on 12.12.22 at 14:00 in room 05.013 (CogSys Lab).
Jasmin Fritz will defend her thesis Prototypical and Near-miss Explanations for Intelligent Tutor Systems in Complex Visual Domains – Learning
to Discriminate Plants
Abstract:
An intelligent tutoring system (ITS) replicates a human tutor to support its students. Artificial intelligence, Socratic dialogs and individualized feedback are already common properties in such ITSs. The ITS presented in this paper contains these properties and teaches its students in complex visual domains. It is designed for third grade students in primary schools in Bavaria and finds its use in their local history and science class. The students should be able to discriminate trees of the forest based on their characteristics and visual features. This application consists of a knowledge base of plants, a student model which tracks the knowledge and learning process of the student, a tutoring module which asks questions of different difficulty levels, provides individualized feedback when needed and a user interface which is pedagogically adapted to children. The focus is particularly on the explanatory feedback which gives the student confidence in the system’s decisions. With the help of near-miss examples packed in Socratic dialogues, the student is guided to the correct solution of the problem.