Reform and Practice of OBE Programming Laboratory Instruction Integrating Learning Behavior Data Mining and Personalized Case Recommendations
Year: 2026
Type: conference-paper
Abstract: Addressing challenges in programming laboratory instruction—such as the difficulty in meeting students’ diverse needs, the haphazard allocation of teaching resources, and the lack of data-driven support in evaluation systems—this study integrates the OBE philosophy with learning behavior data mining techniques to construct a framework for personalized teaching reform. Student learning behavior data was collected via an online platform, standardized using Z-scores, and then analyzed using the K-Means clustering algorithm to segment 254 students into groups. The optimal number of clusters, K=4, was determined using the rule of thumb, and the validity of the clusters was verified through the coefficient of variation (mean 0.32). This enabled data-driven, personalized case recommendations. The study indicates that there are significant differences in learning behaviors among different student groups. Personalized case recommendations can enhance students’ programming abilities, achieving a deep integration of data mining and OBE teaching, and providing a replicable practical solution for the precise and personalized development of programming laboratory instruction.
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: hybrid
OpenAlex ID: W7213156286