International Journal of Human Movement and Sports Sciences Vol. 12(4), pp. 636 - 645
DOI: 10.13189/saj.2024.120404
Reprint (PDF) (757Kb)

Analysis of Human Movement Learning Using Machine Learning Techniques

Ahmed Abatal ^1,*, Hasnaa Rezki ², Ahmed Elhamal ³, Adil Korchi ^4
1 Faculty of Sciences and Techniques, Hassan Premier University, Morocco
² Faculty of Sciences, Hassan II University Ain Chock, Casablanca, Morocco
³ Faculty of Letters and Humanities, Sidi Mohamed Ben Abdellah University, Morocco
⁴ Faculty of Juridical, Economic and Social Sciences, Chouaib Doukkali University, Morocco

ABSTRACT

This research explores the transformative potential of integrating machine learning with human movement learning in education, revolutionizing personalized learning experiences. An advanced system, nestled within a dynamic 3D virtual environment, serves as an analysis tool for human motion and a support mechanism for educators and learners. Central to this investigation is addressing challenges in movement modeling and analysis, particularly in diverse educational contexts like music instrument teaching. Through strategic integration of machine learning algorithms, the system anticipates actions, discerns user situations, and offers refined models, providing valuable feedback for learners and aiding teachers in assessments. Beyond technical proficiency, the ambition is to offer a holistic tool enriching human movement analysis across various learning scenarios, aiming to pave the way for personalized learning to become fundamental in education. In Lute lessons, capturing diverse movements from instructors and students is crucial. Multiple renditions of movements by experts contribute to building robust target gesture models. Filtering steps ensure data integrity, identifying and removing errors swiftly. Challenges in automating the detection of playing styles for intelligent music coaching are addressed by introducing a multimodal dataset covering four Lute techniques, enhancing the development of intelligent software. This dataset encompasses a rich array of human movement data within music instrument teaching, recorded in a 3D virtual environment. Each data point is annotated with contextual information, facilitating nuanced analysis of the relationship between human movement and musical expression. This serves as a foundation for applying machine learning algorithms in personalized music education.

KEYWORDS
Movement Detection, Motion, Machine Learning, Clustering, Unsupervised Learning

Cite This Paper in IEEE or APA Citation Styles
(a). IEEE Format:
[1] Ahmed Abatal , Hasnaa Rezki , Ahmed Elhamal , Adil Korchi , "Analysis of Human Movement Learning Using Machine Learning Techniques," International Journal of Human Movement and Sports Sciences, Vol. 12, No. 4, pp. 636 - 645, 2024. DOI: 10.13189/saj.2024.120404.

(b). APA Format:
Ahmed Abatal , Hasnaa Rezki , Ahmed Elhamal , Adil Korchi (2024). Analysis of Human Movement Learning Using Machine Learning Techniques. International Journal of Human Movement and Sports Sciences, 12(4), 636 - 645. DOI: 10.13189/saj.2024.120404.