MOHAMMAD SHADAB KHAN2024-10-082024-10-082023https://studentrepo.iium.edu.my/handle/123456789/9680The fourth pillar of the United Nations Sustainable Development Goals is 'Quality education' to provide peace and prosperity to the world population; in addition, Special Education Reforms under the Malaysian government scope are influenced by the 'Shared Prosperity Vision 2030'. Special needs education underlines the importance of technological integration in the learning environment to support communication, teaching, and learning needs. Technological advancements assist educators in transforming the educational needs of neurodiverse learners at all academic levels. The early intervention-based teaching and learning process is advocated worldwide, as it has shown positive cues toward neurodiverse learners, especially those with Autism Spectrum Disorder (ASD). However, very little is known about the perception and behavior of teachers toward acceptance of technology-based intervention. Therefore, to bridge the identified gap in academic literature, this study explores the teachers' behavioral intentions concerning the use of technology for teaching children with ASD. A conceptual framework has been developed based on the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) model and other recent studies to address the problem. In this study, constructs such as Performance Expectancy (PE), Personal Innovation (PI), Effort Expectancy (EE), Habit (H), Social Influence (SI), Time (T), Access to Technology (ATT), Voluntariness (V), Self-Efficacy (SE), Facilitating Conditions (FC), Job Relevance (JR), Hedonic Motivation (HM), Risk (RK), and Trust (TR) are measured. The researcher later tested the hypotheses through a survey-based self-administered questionnaire among special education teachers in different parts of Malaysia. This study's theoretical model was evaluated using two different techniques. The first used the PLS-SEM and the Smart-Partial Least Squares software tool. Structural equation modeling was used to evaluate hypothetical relationships. In the second technique, validating hypotheses, assessing their methodology, and gauging their prediction capacity are all tasks that benefit greatly from the application of Machine Learning. In order to look into the correlational routes between constructs, a quantitative research approach of cross-sectional survey method was adopted. A total of 386 replies were compiled using a cluster sampling strategy. 12 out of 17 hypotheses were determined to be statistically significant, and five were found to be insignificant, according to research findings. The findings showed that the following variables had a significant impact on teachers' behavioral intention (BI) to adopt technology for teaching children with ASD and their use behavior (UB): PE, EE, H, ATT, SE, FC, JR, HM, RK, and TR. Whereas BI, H, and FC explain use behavior. The study's findings demonstrated that the suggested model matched the data correctly. Additionally, this study predicts and validates the UTAUT2 model in the context of technology adoption among the teachers of children with ASD, expanding the applicability of the UTAUT2 in IS research. The ML generic algorithms predicted the conceptual model with the best results for the Random Forest algorithm, R2 value for the training (0.993) and testing (0.971) with a minimum error of RMSE = 0.1729 and MAE = 0.1239, in comparison to the predictive power of TAM (R² = 0.67) and UTAUT (R² = 0.73). Finally, this research added to the understanding of technology acceptance among teachers responsible for teaching children with ASD. The study's identification of crucial elements would influence stakeholders, school administrators, and policy and lawmakers to design and implement an inclusive special needs education system and use technology effectively to teach children with ASD, particularly in Malaysia.enJOINTLY OWNED WITH A THIRD PARTY(S) AND/OR IIUMAutism; Special Education Teachers;PLS-SEM; Machine Learning;UTAUT2; Behavioral Intentiondoctoral thesis