Article Details

Abstract

To explore and reveal the relationship between learning behavior and academic performance of higher vocational students, this study selects the original data of learning behavior and academic performance of 1,295 students from 19 majors in typical higher vocational colleges in southern Xinjiang. First, Python is used to clean and transpose the data to form a basic dataset. Then, grade points for each module are calculated based on teaching components including public courses, professional courses, elective courses, and graduation design (thesis), and the students' academic performance is determined according to the regulations of student status management in higher vocational colleges. In the dimension of learning behavior, the data of students' learning behavior in the library and dining behavior in the canteen are normalized and combined with academic performance to form a "learning behavior and academic performance" dataset. Subsequently, Python is used to implement clustering and outlier algorithms, and data mining is carried out on the dataset. On this basis, a decision tree model is generated, and the model is tested, evaluated, and visualized. This study provides a useful reference for exploring the relationship between students' learning behavior and academic performance.

Keywords

clustering; outlier; decision tree; ROC curve; AUC

References

1. [1] Xinhua News Agency The Central Committee of the Communist Party of China and the State Council have issued the "Outline of the Plan for Building an Education Strong Country (2024-2035)" [R/OL] https://www.gov.cn/gongbao/2025/issue_11846/202502/content_7002799.html.
2. [2] Li Hongyu, He Jun, Mo Juqing Analysis of Student Learning Behavior and Grade Prediction Based on Data Mining: A Case Study of Ideological and Political Courses [J]. Modern Information Technology, 2025, 9 (12): 147-151
3. [3] Ding Guoyong, He Bin Empirical Study on Factors Influencing Academic Performance of College Students [J]. Heilongjiang Higher Education Research, 2020, 38 (10): 145-151
4. [4] Li Juntang A Study on the Impact of High School Humanities and Sciences on College Students' Academic Performance [D]. Dalian: Liaoning Normal University, 2012
5. [5] Guo Yuting, Wu Daguang Re understanding the essence and function of credit system and GPA system [J]. Research on Educational Development, 2024, 44 (23): 1-9
6. [6] Huang Yaping, Zhang Yanglei, Wang Xiying Strategic Action and Generative Logic of High achieving Students under the Credit GPA System [J]. Jiangsu Higher Education, 2024 (6): 58-66
7. [7] Cheng Qiao Research on the Mechanism of the Impact of Average GPA on Students' Academic Performance [J]. Military Higher Education Research, 2024, 47 (4): 63-67
8. [8] Li Yanzhi Analysis of Student Learning Behavior: Connotation Examination and Method Exploration - Discussion on the Paradigm Shift of Learning Behavior Analysis Methods under the Background of Big Data [J]. Modern Vocational Education, 2018 (28): 198
9. [9] Ikotun A M，Ezugwu A E，Abualigah L，et al. K-Means Clustering Algorithms: A Comprehensive Review, Variants Analysis, and Advances in the Era of Big Data [J].Information Sciences，2023，622：178-210.
10. [10] Zhang Lina, Zhang Xingrui, Ma Li, etc K-Means clustering algorithm based on heuristic crossover strategy optimization [J]. Journal of Jilin University (Science Edition), 2025, 63 (6): 1663-1672

About the authors

Li Yi (1977-), male, Han nationality, from Jingxing, Hebei Province, professor, master's student, research direction: vocational education, data mining; Xu Ping (1987-), female, Han ethnicity, from Changji, Xinjiang, teaching assistant, undergraduate, research direction: vocational education and informatization.