AI Based Analysis of Online Educational Content for ADHD Students

Basaran B., Polat D., Yilmaz G.

10th International Conference on Computer Science and Engineering, UBMK 2025, İstanbul, Turkey, 17 - 21 September 2025, pp.102-107, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/ubmk67458.2025.11206902
  • City: İstanbul
  • Country: Turkey
  • Page Numbers: pp.102-107
  • Keywords: adhd, artificial intelligence, convolutional neural networks, online education
  • TED University Affiliated: Yes

Abstract

This study assesses the effectiveness and suitability of online educational contents for students with Attention Deficit Hyperactivity Disorder (ADHD) by employing a proposed Convolutional Neural Network (CNN) model alongside established models including MobileNetV2, ResNet50, VGG16, DenseNet, and EfficientNet. Utilizing a newly developed ADHD Online Instructor (AOI) dataset - comprising videos and images categorized based on whether the instructor's hand is in motion during online lessons - the models classify the appropriateness of instructional videos for ADHD learners. Results indicate that the proposed CNN model outperforms the others, achieving an accuracy of 0.9984 with fewer training epochs. These findings provide valuable insights for designing more effective and inclusive online instructional materials tailored to the needs of students with ADHD.