Saturday 15 March 2025
Reading comprehension is a crucial skill for students, but it can be a challenging task for teachers to provide personalized feedback to their pupils. A new system developed by researchers in Japan aims to tackle this issue by generating customized feedback for short-answer reading comprehension questions.
The system uses an innovative approach called the Answer Diagnostic Graph (ADG), which represents the logical structure of a prompt text and identifies gaps in understanding between student responses and model answers. This allows the system to provide tailored feedback templates that can help students identify areas where they need improvement.
In an empirical study, the researchers tested their system with 40 students from two high schools in Japan. The participants were asked to answer two reading comprehension questions and then re-answer them after receiving feedback generated by the system. The results showed that the students who received the personalized feedback made significant improvements in their responses compared to those who did not receive feedback.
The study also found that the feedback had a positive impact on the students’ emotional aspects, such as motivation and satisfaction. Many of the participants reported that they felt more confident in their ability to understand the prompt text and were motivated to revise their responses.
One of the limitations of the system is its reliance on oracle response node estimations and scoring results, which may not be feasible in real-world educational settings. However, the researchers are working to improve the technological aspect of the system to achieve sufficient performance for future studies.
The development of this personalized feedback system has implications for reading education, as it could help teachers provide more effective support to their students. The system’s ability to identify gaps in understanding and provide tailored feedback could also be applied to other subjects and educational contexts.
Overall, the study demonstrates the potential of AI-generated feedback to improve student learning outcomes and engagement. As the technology continues to evolve, it will be interesting to see how educators adapt this approach to enhance reading comprehension skills in their students.
Cite this article: “AI-Powered Feedback System Improves Reading Comprehension Skills”, The Science Archive, 2025.
Reading Comprehension, Personalized Feedback, Ai-Generated, Answer Diagnostic Graph, Logical Structure, Student Responses, Model Answers, Empirical Study, Motivation, Satisfaction
