Quality development of vocational education in the context of big data

Abstract

Taking vocational education as an entry point, this paper utilizes factor analysis to deal with the problem of multicollinearity of vocational education indicators so that the synthesized factors are no longer correlated and lay a good foundation for vocational education cluster analysis. The distance measure and similarity coefficient describe the relationship between vocational education indicators. To increase computational efficiency and handle large and multidimensional data sets more effectively, the advantages of K-mean and grid clustering algorithms are combined to create the GBKM cluster analysis algorithm. Higher education institutions in Province H are chosen as representative samples in order to examine the variables influencing the development of high-quality vocational education, categorize the degree of that development, and forecast the trend of that development. The findings demonstrate that the GBKM cluster analysis algorithm's accuracy difference between the predicted and actual data on the development trend of vocational education ranges from [0,0.1]. The algorithm's absolute error value is 0.182, which falls within an acceptable error range and suggests that the algorithm proposed in this paper is capable of more accurately predicting the development trend of vocational education and may have some theoretical and practical implications for fostering high-quality development of vocational education. It has theoretical and practical implications for advancing the development of vocational education to a high standard.