A fuzzy clustering approach for cloud-based personalized distance music education and resource management
Lei Jiang
DOI: https://doi.org/10.1007/s00500-023-09525-7
IF: 3.732
2023-12-27
Soft Computing
Abstract:Advancements in information technology (IT), the Internet of Things (IoT), and other emerging technologies have led to an overflow of information and big data. The emergence of these technologies and fast internet connectivity has led to the growing popularity of distance learning in academic circles and society. The goal of this paper is to improve online music learning through the use of cloud computing and fuzzy clustering. A significant amount of data regarding students and their activities is collected. Following data cleaning and organization, students are grouped according to their preferred style of music learning using a technique called fuzzy C-mean (FCM) clustering. Each group is then given a customized learning plan that matches them with appropriate music resources. To put this into practice, a simple online music learning platform is created, enabling teachers and students to sign in, give instructions, view results, and log out with ease. To ensure that our approach remains effective even under high user loads, we designed the system to function properly for up to 400 concurrent users. An educational resource database is used to find the best learning materials for students, and FCM helps them pick the right resources. To make everything more organized, a resource library is designed using a simple language called XML. This library has different categories for learning materials and follows international standards to make sure everything works together smoothly. The performance of the system is measured in terms of scalability (8/10 response time, 7/10 throughput, 9/10 resource utilization), usability (8/10 user satisfaction, 7/10 task completion time, 6/10 error rate), effectiveness (9/10 learning outcomes, 8/10 student engagement, 9/10 retention rates). It has a 99% uptime and no security incidents have been reported. This study shows that our method can improve online music education, making it more personalized and effective for each student. The steps we followed, like using the FCM clustering method and creating a user-friendly system, can be applied to other areas of online learning to make them better too.
computer science, artificial intelligence, interdisciplinary applications