Teaching artificial intelligence through machine learning

This research examines the implementation of machine learning concepts in elementary education through a sample of 1,009 fifth-grade students. The intervention involved a set of structured activities related to machine learning using Teachable Machine and RAISE (built on Scratch 3.0). A pre-experimental research design was employed, combining descriptive analysis with statistical inference. Specifically, a Student’s t-test was applied to analyze the first dimension, while the Wilcoxon test was used for the second dimension. The results indicate that elementary school students improved their understanding of machine learning and the ways in which artificial intelligence models are developed. Furthermore, students with prior experience using Scratch in school obtained higher scores and reported greater motivation compared to those without experience in block-based programming environments. The findings suggest that interactive learning activities focused on machine learning are effective for motivating students and facilitating their comprehension of AI, including how it is trained and generated. Additionally, these activities increased engagement and enjoyment during the sessions. Overall, the study demonstrates that implementing pedagogical designs aimed at introducing machine learning and artificial intelligence in primary education is both feasible and beneficial.