Abstract: In order to study the factors that affect students’ achievement and better help students learn, teachers carry out teaching and school management, the Data Mining technology (Data Mining) was adopted on the existing original data, and the machine Learning algorithm (Deep Learning) was used to analyze the student data, and a classification prediction model was built. Decision Tree algorithm in machine learning is employed to select data, classify features, and normalize data preprocessing. Label Ecode tag encoding can effectively reduce data dimension and ensure classification accuracy quantization. The data in this paper are from the family background data and school examination data of foreign high school students. Sklearn, Seaborn, Matplotlib and other software are used to realize decision tree algorithm, generate off-line model and academic index evaluation report, so as to better complete the research on influencing factors of student achievement prediction.