doi: 10.18178/ijiet.2026.16.9.2710
Opinion Mining for Educational Data-driven Decision-making in Computing-supported Interventions for Children with Autism
Muslihah Wook1,*
,
Irma Syarlina Che Ilias2,* ,
Suzaimah Ramli1 ,
Noor Afiza Mat Razali1 ,
Nor Asiakin Hasbullah1 ,
Norulzahrah Mohd Zainudin1 ,
Nor Fyadzillah Mohd Taha3
- 1Department of Computer Science, Faculty of Defence Science and Technology, National Defence University of Malaysia, Kuala Lumpur, Malaysia
- 2Computer Engineering Technology Section, Malaysian Institute of Information Technology, Universiti Kuala Lumpur, Kuala Lumpur, Malaysia
- 3Department of Science and Technology Maritime, Faculty of Defence Science and Technology, National Defence University of Malaysia, Kuala Lumpur, Malaysia
- Manuscript receivedNovember 26, 2025
- revisedDecember 29, 2025
- acceptedApril 15, 2026
- publishedSeptember 21, 2026
Abstract
Parental feedback regarding interventions at autism therapy centres is increasingly accessible via online platforms. However, it remains predominantly qualitative and underutilised in data-driven educational decision-making environments. To bridge this gap, the present study introduces an opinion mining framework that could convert unstructured parental testimonials into actionable insights to inform computer-supported interventions for children with autism spectrum disorder. The principal aim of this research was to scrutinise parental feedback for emergent thematic patterns and sentiment trends capable of guiding evidence-based educational and therapeutic strategies. By employing purposive sampling, 118 testimonials were gathered from Google Reviews and Telegram channels spanning from July 2024 to October 2025, encompassing both textual and image-based contents. The Optical Character Recognition (OCR) software was utilised to extract texts from images, followed by conventional preprocessing. Latent Dirichlet Allocation (LDA) and Aspect-Based Sentiment Analysis (ABSA) facilitated thematic extraction and aspect-specific sentiment appraisal, with the application of a logistic regression model for sentiment classification. The analysis discerned five prevalent themes, namely progress, behaviour, routine, communication, and school interaction, with 81.4% of feedback evincing positive sentiments, which were particularly pronounced in the learning, behaviour, and speech domains. The sentiment classifier registered 83% accuracy alongside an F1-score of 0.91 for positive sentiments. These outcomes affirmed the efficacy of opinion mining in advancing data-informed educational decision-making processes and perpetual refinement within autism-focused therapeutic and educational paradigms.
Keywords
- opinion mining
- sentiment analysis
- OCR
- autism intervention
- Latent Dirichlet Allocation (LDA)
- Aspect-Based Sentiment Analysis (ABSA)
- topic modelling
- data-informed education
How to Cite
Muslihah Wook, Irma Syarlina Che Ilias, Suzaimah Ramli, Noor Afiza Mat Razali, Nor Asiakin Hasbullah, Norulzahrah Mohd Zainudin, and Nor Fyadzillah Mohd Taha, "Opinion Mining for Educational Data-driven Decision-making in Computing-supported Interventions for Children with Autism," International Journal of Information and Education Technology, vol. 16, no. 9, pp. 2530-2540, 2026. https://doi.org/10.18178/ijiet.2026.16.9.2710
Copyright & License
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).