Scholarlink

<p>Schools and colleges use students’ grades as a primary way to evaluate how well their teachers are performing and to assess the overall quality of their student body. Academic and practical performance are two key indicators of a student’s success in school. Most research suggests that students share common traits while overlooking their unique characteristics. The main aim of this paper is to develop a framework for understanding how generative artificial intelligence (AI) technologies can enhance classroom learning and support students’ success. On the other hand, understanding what each student learns best depends on their individual characteristics, which are important for improving academic performance. This study employs machine learning (ML) techniques, including Logistic Regression (LR), Random Forest (RF), Gradient Boosting (GB), K-nearest neighbors (KNN), and Support Vector Machine (SVM) to classify students and predict their performance. Neural networks (NN) are used to achieve this goal. We tested the algorithms on a standard dataset of 2,392 instances of high school students from Kaggle. We analyzed students’ scores to see how they fit into different groups. The method involves handling missing values, normalizing the data, and adjusting variables as necessary. The RF selection method identified the most important traits. The GB method achieved 92% accuracy on the test, while the NN predicted student performance with an MSE of 0.0665 and an R<sup>2</sup> of 93.27%.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predicting the Academic Performance of the Students: Machine Learning Algorithms

  • 期刊论文

  • 发表日期：

    2025年12月11日

查看全文
  • Asma Abdulmana Alhamadi

摘要

Schools and colleges use students’ grades as a primary way to evaluate how well their teachers are performing and to assess the overall quality of their student body. Academic and practical performance are two key indicators of a student’s success in school. Most research suggests that students share common traits while overlooking their unique characteristics. The main aim of this paper is to develop a framework for understanding how generative artificial intelligence (AI) technologies can enhance classroom learning and support students’ success. On the other hand, understanding what each student learns best depends on their individual characteristics, which are important for improving academic performance. This study employs machine learning (ML) techniques, including Logistic Regression (LR), Random Forest (RF), Gradient Boosting (GB), K-nearest neighbors (KNN), and Support Vector Machine (SVM) to classify students and predict their performance. Neural networks (NN) are used to achieve this goal. We tested the algorithms on a standard dataset of 2,392 instances of high school students from Kaggle. We analyzed students’ scores to see how they fit into different groups. The method involves handling missing values, normalizing the data, and adjusting variables as necessary. The RF selection method identified the most important traits. The GB method achieved 92% accuracy on the test, while the NN predicted student performance with an MSE of 0.0665 and an R2 of 93.27%.