Enhancing the Willingness of Adopting AI in Education Using Back-propagation Neural Networks

The use and application of artificial intelligence (AI) in Taiwanese higher education bring about new possibilities and challenges.Using AI will effectively change the entire internal structure of the Taiwan Institute of Higher Education; for instance, AI can be utilized in education to investigate how teachers enrich their knowledge, how students learn, and how higher education institutions make accurate and timely decisions.Timely responses are critical for higher education.Hence, AI applications in higher education are important from an educational perspective.In this research, we employ Statistical Package for the Social Sciences (SPSS) and smart partial least squares regression (Smart PLS) software as the primary analytics tools; the former is used to analyze fundamental statistics, whereas the latter is used to investigate the structural model.We also explore how stakeholders can adopt AI applications using back-propagation neural networks and structural equation modeling for analysis.A framework and model were developed for this study and 408 respondents analyzed, and we concluded that the model can explain the increase in the willingness to adopt AI in higher education.education to ensure a good quality of education. (4)Many researchers consider that Taiwanese higher education urgently needs to implement the latest technologies such as AI. (5)I technology enables the customization of learning, where the unique needs of diverse student groups are addressed.This approach ensures that each student benefits from an educational experience specifically tailored to their individual requirements. (6)In higher education, AI plays a pivotal role in enhancing learning experiences (7) by offering personalized and individualized learning strategies.Various AI applications are being developed to personalize these experiences, (8) although current AI technologies might need more time to fully adapt to these advanced requirements.In the interim, chatbots provide personalized support by addressing specific student issues with targeted solutions.As AI technology evolves, these chatbots will increasingly respond to individual student inquiries with greater accuracy, (9) extending support beyond traditional classroom settings.AI-driven systems also assist students in accessing admission information and navigating administrative processes.Furthermore, AI is instrumental in creating intelligent textual materials, (10) and thus contributes significantly to the advancement of higher education in numerous ways. (11)s students face an increasing volume of homework and the need for updated learning skills, the application of modern technologies such as AI becomes crucial. (12)However, the effectiveness of AI in education hinges on its adoption by key stakeholders, including students, teaching staff, and nonteaching personnel such as administrative staff.Despite its importance, there is a noticeable scarcity of detailed research on AI adoption within the Taiwanese higher education sector. (13)In this study, we aim to identify the factors affecting the adoption of AI in higher education and will explore the following research questions in this context.Structural equation modeling (SEM) can be used to process multiple variables simultaneously, allowing observational errors between arguments and variables; this method is particularly suitable for sociological and psychological studies involving analyses with multiple variables that cannot be measured accurately or directly.SEM can also be used to investigate various facets, latent variables, and the relationship between these facets and observed variables, including their direct and indirect effects, allowing researchers to understand the causation and interaction between different variables.The back-propagation neural network (BPNN) is an AI technology that imitates the biological nervous system and solves complex problems through learning and self-adaptation. (14,15) In education, neural networks can be applied in adaptive learning, personalized learning, student assessment, learning analysis, and other fields, which are significant to education reform and teaching innovation. (14,16)Adaptive learning is a learning method that dynamically adjusts the learning process in accordance with the characteristics and needs of learners, (16) and BPNN can realize personalized learning recommendations and promote adjustments for learners by analyzing students' learning history, behaviors, hobbies, and other data, (17) which will not only improve the learning effect but also allow learners to participate more actively and autonomously in the learning process.On the other hand, personalized learning is a learning approach that addresses individual differences among learners, (18) where BPNN can implement personalized design and adjustment of learning content and methods by analyzing students' learning habits, styles, abilities, and other factors; this can better meet the needs of learners and improve learning effects. (14)