Influencing Factors of Vocational College Students’ Career Development: A GA-Optimized KNN-Random Forest Model

Abstract

College students can effectively realize personal value and promote social progress by attaching great importance to career planning and development. This study collected 404 valid data points through the career aspiration questionnaire, career exploration questionnaire, and career adaptability questionnaire and used the K-nearest neighbor (KNN) random forest algorithm with hyperparameter optimization based on genetic algorithms (GA) for influencing factor analysis. First, data cleaning was performed, followed by basic frequency analysis and difference analysis. Next, the KNN algorithm was introduced to establish a random forest model, with parameters for the random forest determined through hyperparameter analysis using a genetic algorithm (GA). Subsequently, data classification and value assignment were conducted to predict the importance ranking of influencing factors. Finally, the model’s accuracy (Acc), precision (P), recall (R), and F1 score were evaluated and compared with the unimproved Random Forest algorithm. Results demonstrated that the KNN Random Forest algorithm significantly improved Acc compared to traditional methods. The study found that in terms of career aspiration and adaptability, the top three factors were participation in part-time jobs, internships, or social practices; grade; and career counseling experience. For career exploration, the top three factors were grade, participation in part-time jobs, internships, or social practices, and firstgeneration college student. Additionally, there were variations in the importance rankings of each dimension in career aspiration, career exploration, and career adaptability. Higher vocational colleges need to further strengthen guidance on internships and practical training, as well as management of career counseling, to more effectively promote students’ career development.