Analysis of Online Learning Behaviors of Higher Vocational College Students : Taking Python Data Preprocessing Course as an Example

The development of online education has brought about a massive amount of online learning behavior data. How to fully utilize and analyze these data is a research hotspot of concern in higher vocational education. This study is based on the online learning behavior data of $\mathbf{1 7 9}$ higher vocational college students in the Python data preprocessing course, and observes, analyzes the correlation and K-Means clustering characteristics of online learning behavior. The relevant analysis results indicate that online learning behaviors have an impact on students’ test scores. The clustering analysis results indicate that the research subjects can be divided into four groups, excellent, good, speculative, and needing effort. The students in different groups have different learning styles and characteristics. Finally, this study proposes two suggestions: improving the construction of online course resources, and differentiating management for different types of learning behavior students.