Logo des Repositoriums

Auflistung nach Schlagwort "Intelligent Tutoring Systems"

1 - 3 von 3
  • Conference paper
    Evaluating Task-Level Struggle Detection Methods in Intelligent Tutoring Systems for Programming
    (Proceedings of DELFI 2024, 2024) Dannath, Jesper; Deriyeva, Alina; Paaßen, Benjamin
    Intelligent Tutoring Systems require student modeling in order to make pedagogical decisions, such as individualized feedback or task selection. Typically, student modeling is based on the eventual correctness of tasks. However, for multi-step or iterative learning tasks, like in programming, the intermediate states towards a correct solution also carry crucial information about learner skill. We investigate how to detect learners who struggle on their path towards a correct solution of a task. Prior work addressed struggle detection in programming environments on different granularity levels, but has mostly focused on preventing course dropout. We conducted a pilot study of our programming learning environment and evaluated different approaches for struggle detection at the task level. For the evaluation of measures, we use downstream Item Response Theory competency models. We find that detecting struggle based on large language model text embeddings outperforms chosen baselines with regard to correlation with a programming competency proxy.
  • Konferenzbeitrag
    Fehlvorstellungen in der Programmierausbildung: Eine Heuristik für die semi-automatische Annotation von Fehlerkandidaten
    (Proceedings of the Sixth Workshop "Automatische Bewertung von Programmieraufgaben" (ABP 2023), 2023) Fischer, Björn; Panitz, Sven Eric; Dörner, Ralf
    Die zuverlässige Erkennung von Fehlern zu Fehlvorstellungen in der Programmierausbildung stellt eine Herausforderung dar, die mit Deep Learning adressiert werden kann. In dieser Arbeit wird eine Heuristik vorgestellt, die es ermöglicht, die dafür erforderlichen Annotationen weitestgehend automatisch zu generieren. Die Heuristik verbindet Informationen aus der statischen und dynamischen Codeanalyse mit dem Ziel, mögliche Fehlalarme zu reduzieren. Erste Ergebnisse zeigen in unserem Datenfall anhand eines betrachteten Fehlertyps, dass die Heuristik in etwa der Hälfte der Fälle eine automatische Entscheidung treffen kann und dabei eine Genauigkeit von 81 % erreicht. Dies stellt eine erhebliche Verbesserung von etwa einem Drittel gegenüber den Ergebnissen von Pattern Matching dar.
  • Textdokument
    An Intelligent Tutoring System Concept for a Gamified e-Learning Platform for Higher Computer Science Education
    (SEUH 2023, 2023) Meißner, Niklas; Speth, Sandro; Breitenbücher, Uwe
    Intelligent Tutoring Systems (ITSs) are increasingly used in modern education to automatically give students individual feedback on their performance. The advantage for students is fast individual feedback on their answers to asked questions, while lecturers benefit from considerable time savings and easy delivery of educational material. Of course, it is important that the provided feedback is as effective as direct feedback from the lecturer. However, in digital teaching, lecturers cannot assess the student’s knowledge precisely but can only provide information on which questions were answered correctly and incorrectly. Therefore, this paper presents a concept for integrating ITS elements into the gamified e-learning platform IT-REX so that the feedback quality can be improved to support students in the best possible way.