The Evaluation Model of College English Diagnostic Exercises Based on Machine Learning

Abstract

Online learning is an important way for college students to learn English independently. The evaluation information provided by the previous online teaching platform is more summative evaluation, which cannot make students have a more intuitive and comprehensive understanding of their English learning status and lack of personalized guiding suggestions. Therefore, this paper combines data mining technology with machine learning to build an English diagnostic exercise model that can analyze students’ learning status, the correlation between knowledge points and question types, and predict English achievement, so as to provide students with more comprehensive analysis data information. The experimental results show that the evaluation model of college English diagnostic practice based on machine learning has the classification results of learning state with finer granularity, effectively analyzes the association rules of knowledge points and question types, and has high prediction performance. It can help students fully understand their English learning status, provide students with personalized analysis data and effective guiding suggestions, and enhance students’ English application ability, improving CET-4.