With the rapid development of the information age, new technological means are constantly applied to the field of education and produce good results. This paper uses data mining algorithms to extract and analyze data on college English education and explore how to adjust the teaching methodology. On the basis of improving the inherent defects of the DPC algorithm, this paper determines the clustering center through a scoring function, uses the K-neighbor method to form the cluster backbone, and proposes a nearest-neighbor-based label propagation method to propagate the labels to their adjacent nodes to complete data mining. After mining and analyzing the current situation of English education, it was found that the age group over 30 years old accounted for 77.6%, female teachers accounted for 74.6% of the total number of teachers, teachers with the titles of lecturer and assistant professor accounted for 58%, and 33% of the students were not too satisfied or dissatisfied with the content of the English course. It shows that the English teaching team of universities in this province has shortcomings such as gender imbalance, insufficient young and senior teachers, and less than satisfactory content of English courses, which need to be improved in several aspects.