Personalized Feedback Enhanced by Natural Language Processing in Intelligent Tutoring Systems

Personalized Feedback Enhanced by Natural Language Processing in Intelligent Tutoring Systems

연구 분야: Artificial Intelligence

학회: International Conference on Intelligent Tutoring Systems

This paper proposes a novel approach for enhancing feedback in intelligent tutoring systems (ITSs) for Java programming using natural language processing (NLP). The proposed approach overcomes the limitations of traditional rule-based feedback generation systems and provides more personalized and relevant feedback to learners. The architecture includes three main components: a natural language parser (that takes as input comments and/or questions of the user that can be inserted through a text box in the user interface.), a feedback generator, and a feedback evaluator. The natural language parser is responsible for converting the unstructured text input of the learner into structured data, which can be analyzed for generating feedback. The feedback generator component then processes this data and generates personalized feedback for the learner based on their specific needs. Finally, the feedback evaluator component assesses the quality of the generated feedback and determines its helpfulness to the learner. The evaluation results are promising, indicating that using NLP techniques can improve the overall performance of intelligent tutoring systems and provide a more personalized learning experience for students.