Using machine learning to predict student outcomes for early intervention and formative assessment

利用机器学习预测学生学习成果,以进行早期干预和形成性评估

阅读:1

Abstract

The increasing importance of early prediction of student performance has led to research into machine learning models that can be used to assess student outcomes more accurately.This study focused on developing a predictive model based on machine learning algorithms to evaluate student performance and provide early intervention mechanisms. Create a new predictive model using machine learning algorithms to assess student performance and identify the key variables that influence success. The proposed model aims to serve as an early warning system to detect potential academic failures and suggest interventions. A questionnaire was developed to collect data from the students. Four machine learning algorithms, C5.0, CART, Support Vector Machine (SVM) and Random Forest, were used to analyze the data. The effectiveness of each algorithm was evaluated with a focus on performance accuracy. Among the four algorithms, Random Forest achieved the most consistent results in the cross-validation metrics. However, C5.0 provided higher accuracy on the test set and CART showed the highest training performance, indicating performance conflicts, which are analyzed in more detail in the Discussion section. Based on these findings, a new classification model is proposed that includes the most important variables that significantly influence student success. This model was developed to detect academic failure at an early stage and enable timely intervention. The proposed predictive model provides a valuable tool for early identification of at-risk students and can support formative assessments. By identifying students who are likely to fail, the model provides opportunities for interventions to improve their academic outcomes. It is expected to help educators respond more effectively to student needs, ensure equity in the classroom, and provide cost-effective solutions for education policymakers.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。