Jadwal Sholat

Memuat jadwal sholatโ€ฆ

Ilmu Komputer & AI editorial

Open AccessOA2023

Leveraging Computer Vision for Adaptive Learning in STEM Education: Effect of Engagement and Self-Efficacy

Researchers developed RASEDS, a computer vision system that detects STEM learners' engagement levels using YOLOR and the ICAP framework, and recommends adaptive materials. The study found that RASEDS accurately gauges engagement and that its use significantly improves engagement and self-efficacy, with a positive relationship between the two.
Ting-Ting Wu; Hsin-Yu Lee; Wei-Sheng Wang; Chia-Ju Lin; Yueh-Min Huangยท International Journal of Educational Technology in Higher Educationยท 2023ยท DOI 10.1186/s41239-023-00422-5

The core problem

In STEM education, cultivating problem-solving skills is paramount, yet accurately assessing learners' engagement remains a significant challenge. The authors propose the Real-time Automated STEM Engagement Detection System (RASEDS) to address this issue. RASEDS leverages artificial intelligence, computer vision, and the Interactive, Constructive, Active, and Passive (ICAP) framework to detect and map learners' interactions onto four levels of engagement. This mapping informs the recommendation of adaptive learning materials aimed at boosting both engagement and self-efficacy in STEM activities. The study aims to validate RASEDS' accuracy in gauging engagement and to investigate the effects of adaptive materials on engagement and self-efficacy, as well as the relationship between these two constructs.

Innovation

The study employed a design that involved the development and evaluation of RASEDS. The system uses You Only Learn One Representation (YOLOR) for real-time detection of learners' interactions, which are then classified into the four ICAP levels: Interactive, Constructive, Active, and Passive. Based on the detected engagement level, the system recommends adaptive learning materials tailored to the learner's current state. The evaluation likely involved participants engaged in STEM activities, with data collected on engagement (via RASEDS) and self-efficacy (via surveys). The study tested the accuracy of RASEDS in gauging engagement and assessed the impact of the adaptive materials on engagement and self-efficacy through pre- and post-measures or comparisons. However, specific details on sample size, duration, and instruments are not provided in the raw content.
Introduction
In STEM education, cultivating problem-solving skills is paramount, yet accurately assessing learners' engagement remains a significant challenge. The authors propose the Real-time Automated STEM Engagement Detection System (RASEDS) to address this issue. RASEDS leverages artificial intelligence, computer vision, and the Interactive, Constructive, Active, and Passive (ICAP) framework to detect and map learners' interactions onto four levels of engagement. This mapping informs the recommendation of adaptive learning materials aimed at boosting both engagement and self-efficacy in STEM activities. The study aims to validate RASEDS' accuracy in gauging engagement and to investigate the effects of adaptive materials on engagement and self-efficacy, as well as the relationship between these two constructs.

Why it matters

The findings underscore the transformative potential of AI in facilitating adaptive learning in STEM education. The symbiotic relationship between engagement and self-efficacy is highlighted: engagement and self-efficacy appear to reinforce each other, creating a positive feedback loop. This suggests that interventions targeting one construct may indirectly benefit the other. The use of computer vision and the ICAP framework provides a novel approach to real-time engagement detection, enabling timely and personalized adaptive learning. The study contributes to the growing body of research on AI in education by demonstrating a practical application that enhances both engagement and self-efficacy. However, limitations such as the lack of detailed methodological information and potential generalizability issues should be considered. Future research could explore the long-term effects and scalability of RASEDS in diverse educational settings.

Who should read this

Pembaca Research Digest dan praktisi rumpun ini.

Opening member contentโ€ฆ