Research on the Construction and Evaluation of National Online Excellent Courses Based on Internet Technology: A Study of Vocational Education Courses Using SVM, K-Means, and Recurrent Neural Networks for Performance Analysis

Due to rapid advancements in information technology, blended learning that combines online and offline teaching methods, has become the prevailing model in modern education. However, evaluation of student performance in a hybrid environment has some challenges, particularly in achieving a comprehensive and personalized assessment. This research study aims at establishing an innovative assessment system to evaluate students’ performance in the blended learning courses, stressing both individual progress and process-oriented evaluation. The study uses a combination of online MOOCs and offline apps for evaluation, incorporating advanced machine learning algorithms, such as Support Vector Machines (SVM), K-Means clustering, and Recurrent Neural Networks (RNN). All these models are applied to analyze and assess students' performance and development in a multi-dimensional manner. The results of the system's implementation in the "Electronic Product Testing and Maintenance" course presented significant improvements in student learning outcomes and individual development, leading to the course being recognized as a National Online Excellent Course by the Ministry of Education in 2023. This research validates the effectiveness of integrating educational assessments and machine learning models and highlights the potential for enhancing both the timeliness and comprehensiveness of evaluations in vocational education.