An Ensemble-based Machine Learning Model for Investigating Children Interaction with Robots in Childhood Education

Investing in children's well-being and supporting high-quality pre-school education is a significant component of its promotion (ECE). All children have the right to participate. ECE teachers' thoughts about children's participation were examined to see if they were linked to children's perceptions of their participation. On the other hand, current studies focus on a single categorization method with lower overall accuracy. The findings of this study provided the basis for the development of an ensemble machine learning (ML) approach for measuring the participation of children with learning disabilities in educational situations that were specifically developed for them. Visual and auditory data are collected and analyzed to determine whether or not the youngster is engaged during the robot-child interaction in this manner. It is proposed that an ensemble ML technique (Enhanced Deep Neural Network (EDNN), Modified Extreme Gradient Boost Classifier, and Logistic Regression) be used to judge whether or not a youngster is actively engaged in the learning process. Children's participation in ECE courses depends on both the quantitative and qualitative characteristics of the classroom, according to this research.