Scholarlink

<p>In recent years, the field of educational technology has experienced significant advancements, with innovative methods being developed to assess and improve student performance. Traditional approaches developed for student performance analysis face challenges, such as simple grading or standardized testing, often fail to capture the complex, multifaceted nature of learning. These methods do not manage dynamic factors like student persistence, engagement, and cognitive development, which play a crucial role in shaping learning outcomes. To address this limitation, an enhanced deep learning approach was developed. The data are collected from students’ academic records. Initially, the data undergoes pre-processing through decimal scaling normalization, the relative density factor, and missing value imputation using extreme gradient boosting. Multi-Dimensional Scaling (MDS) is applied to extract relevant features from the pre-processed data. The GritNet approach predicts students’ academic performance, with its hyperparameters optimized using the Hippopotamus algorithm to enhance accuracy. The experimental results demonstrate the effectiveness of this integrated approach, achieving impressive performance metrics and attaining 95% of accuracy, 93% of precision, 93% of recall, 96% of specificity, which highlights its ability to provide highly reliable predictions of student outcomes. These results suggest that the combined techniques provide a powerful tool for enhancing student performance analysis and personalizing learning pathways through adaptive gamified learning, compared to existing systems.</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

Advanced Student Performance Analysis Through Adaptive Gamified Learning: Integrating Gritnet, Hippopotamus Optimization, and Multi-Dimensional Scaling for Predictive Insights

  • 期刊论文

  • 发表日期：

    2026年5月2日

查看全文
  • Margret Vijay,
  • J. P. Jayan,
  • R. I. Heaven Rose

摘要

In recent years, the field of educational technology has experienced significant advancements, with innovative methods being developed to assess and improve student performance. Traditional approaches developed for student performance analysis face challenges, such as simple grading or standardized testing, often fail to capture the complex, multifaceted nature of learning. These methods do not manage dynamic factors like student persistence, engagement, and cognitive development, which play a crucial role in shaping learning outcomes. To address this limitation, an enhanced deep learning approach was developed. The data are collected from students’ academic records. Initially, the data undergoes pre-processing through decimal scaling normalization, the relative density factor, and missing value imputation using extreme gradient boosting. Multi-Dimensional Scaling (MDS) is applied to extract relevant features from the pre-processed data. The GritNet approach predicts students’ academic performance, with its hyperparameters optimized using the Hippopotamus algorithm to enhance accuracy. The experimental results demonstrate the effectiveness of this integrated approach, achieving impressive performance metrics and attaining 95% of accuracy, 93% of precision, 93% of recall, 96% of specificity, which highlights its ability to provide highly reliable predictions of student outcomes. These results suggest that the combined techniques provide a powerful tool for enhancing student performance analysis and personalizing learning pathways through adaptive gamified learning, compared to existing systems.