NEXT-GENERATION EDGE CLOUD-TO-THING CONTINUUM MODEL FOR EVALUATING CHILDREN'S MOTOR SKILLS USING DEEP LEARNING ALGORITHM.

Young children can enhance their movements and develop motor abilities through play and education with guidance and practice. Understanding young children's physical development and identifying potential growth areas require an analysis of their motor skills. In this study, we assess young children's motor skills using a unique deep residual technique on the cloud-to-thing continuum. Seven kids were used to identify autistic movements. The dataset was gathered from the Tamimi Centre for Autism in Saudi Arabia. The data preprocessing to standardize the data's scale using min–max normalization and removing noise-relevant features is extracted using principal component analysis (PCA). This step is crucial for ensuring the quality and reliability of the data used for subsequent analysis. The bumble bees mating optimization with deep residual algorithm (BBMO-DRN) is designed to handle the complexities of motor skill assessment. Finally, we explore the potential of cloud–fog–edge computing in data storing and processing young children's motor skill development. The results showed that the proposed method performs better compared to the existing methods. This will make it possible to evaluate how well the suggested solution works in terms of accuracy, precision, recall, F 1-score, and scalability. As a result, this proposed approach in evaluating young children's motor skills is to improve data storage in the cloud to the thing continuum. [ABSTRACT FROM AUTHOR]