An Algorithm Course Grade Warning Model Based on Ensemble Learning

Abstract

This paper analyzes the massive multidimensional teaching process data generated in the blended teaching of algorithm design and analysis courses. In response to the problems of uneven data set categories and distribution differences in curriculum data among students at different levels and majors, the data is processed unbiasedly to construct new curriculum ability features. A model based on ensemble learning is constructed to provide intelligent early warning for students' learning status. Teaching application found that the model can effectively identify students with learning difficulties and provide effective assistance for teachers' differentiated teaching and guiding students' curriculum learning