Exploring the Key Issues and Practical Paths of Modernizing the Governance of Vocational Education for Deep Learning

Abstract

Abstract Deep learning algorithms are widely used in various fields due to the increasing popularity of education modernization, and the Ministry of Education has expressed a requirement to apply these algorithms to the governance of education in vocational schools in order to strengthen their teaching management. This paper constructs a student portrait model based on an improved K-means algorithm to monitor and analyze students’ daily behaviors. Firstly, we collect and integrate data from various sources. The dataset was preprocessed using data preprocessing methods, including data cleaning, data transformation, and data normalization. The Canopy algorithm was used to determine the number of clusters, and the number of clusters and cluster centers obtained by the Canopy algorithm were used as input parameters for the K-mean algorithm. The Maximum Minimum Distance algorithm was used to select sample points as far as possible for the K-means algorithm. Finally, we verify the effectiveness of the improved clustering algorithm and analyze the two dimensions of students using it. The findings show that students of type I in the learning behavior-oriented clustering visited the library an average of 22.54 times a month. There are a small number of students who spend more time online, averaging 48.45 hours per month. The majority of students’ categorical data and real-life learning behaviors coincide. This provides a basis for vocational school educators to optimize decision-making and teaching methods, indicating that the model in this paper is applicable to modern vocational education governance.