.png){kind=icon}
返回
Efficient hyperparameter tuning for predicting student performance with Bayesian optimization
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-023-17525-w.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Higher education is crucial as it introduces students to various fields and then guides them to the next steps. Student's academic performance is critical and could lead to failure if it is not monitored to find the strengths and weaknesses of students and the factors that affect them. That is why the student academic prediction method should be improved so teachers can predict their students' performance. A lot of research tried to improve the prediction accuracy but had problems with imbalanced data and how to tune the algorithm. For this case, we proposed two different machine learning algorithms that handle imbalanced data by applying the Synthetic Minority Oversampling Technique and employing a hyperparameter tuning algorithm to increase the prediction during the training process in the machine learning models. The machine learning models we used are Random Forest and Decision Tree. Models were further tuned using Grid Search, Random Search and Bayesian Optimization Hyperparameter Tuning. After we compared them, the results showed that Synthetic Minority Oversampling Technique and Bayesian Optimization combined with the Decision Tree algorithm outperformed models for student academic prediction.
Keyword:
Student academic prediction
Machine learning
Synthetic minority oversampling technique
Bayesian optimization
Hyperparameter tuning
期刊
IF:
3
论文数:
1.9W
被引数:
3.2W
机构
引用论文
分享收藏
Utilizing early engagement and machine learning to predict student outcomes.png)
COMPUTERS & EDUCATION
OAAI
COMPUTERS & EDUCATIONIF10.5
OAAI分享收藏
查看更多内容