การวิเคราะห์เปรียบเทียบประสิทธิภาพ เพื่อการแนะแนวนักศึกษาโดยใช้เทคนิคการเรียนรู้ของเครื่อง: กรณีศึกษาวิทยาลัยอาชีวศึกษาอุดรธานี

Abstract

This research aimed to (1) develop machine learning models for predicting the academic program selection trends of Vocational Certificate (VC) students and (2) compare the performance of Logistic Regression, Random Forest, and Gradient Boosting to identify the most suitable prediction model. The dataset consisted of 1,789 records with 13 attributes collected from Vocational Certificate students at Udon Thani Vocational College. The data were divided into training and testing sets using an 80:20 ratio (1,431 and 358 records, respectively) and categorized into three academic programs: Accounting (504 records), Information Technology (601 records), and Food and Nutrition (684 records). Data preprocessing was performed prior to model development, and feature selection was conducted using Recursive Feature Elimination with Cross-Validation (RFECV). The prediction models were developed using Logistic Regression, Random Forest, and Gradient Boosting, and their performance was evaluated using 10-fold cross-validation and a testing dataset based on Accuracy, Precision, Recall, and F1-score.

The results indicated that the Logistic Regression model achieved the highest predictive performance, with an Accuracy of 92.46%, Precision of 90.78%, Recall of 92.24%, and an F1-score of 91.50%, while requiring only 0.04 seconds for model training. Overall, Logistic Regression outperformed the Random Forest and Gradient Boosting models and was identified as the most suitable model for predicting the academic program selection trends of Vocational Certificate students using the dataset employed in this study.