Scholarlink

<p>Remote learning in South Asia faces significant challenges related to student engagement, quality management, and predictive analytics for satisfaction assessment. Traditional approaches struggle with noisy data, suboptimal feature selection, and insufficient predictive accuracy, limiting their effectiveness. We have proposed this paper to overcome the challenges faced by such systems in previous designs. To find practical methods for improving educational opportunities in Southeast Asia, the research concentrates on forecasting student satisfaction and optimizing quality management measures for remote learning in the area. We proposed a Minimum Redundancy Maximum Relevance-based Whale Optimization Algorithm (MRMR-WOA) for feature selection, Long Short-Term Memory and Convolutional Neural Networks (CONV-LSTM) for feature extraction, and (Random Forest classifier-based improved quantum particle swarm optimization (RFC-IQPSO) for the prediction model. The approaches opted in this study to overcome the issues faced in previous technologies are tested using different metrics like: Accuracy, Precision, F-measure, Recall, True Positive, and False Positive, to analyze its performance.</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 Student Satisfaction and Quality Management for Remote Learning in Southeast- Asia

  • 期刊论文

  • 发表日期：

    2025年6月4日

查看全文
  • Jiacheng Zhong

摘要

Remote learning in South Asia faces significant challenges related to student engagement, quality management, and predictive analytics for satisfaction assessment. Traditional approaches struggle with noisy data, suboptimal feature selection, and insufficient predictive accuracy, limiting their effectiveness. We have proposed this paper to overcome the challenges faced by such systems in previous designs. To find practical methods for improving educational opportunities in Southeast Asia, the research concentrates on forecasting student satisfaction and optimizing quality management measures for remote learning in the area. We proposed a Minimum Redundancy Maximum Relevance-based Whale Optimization Algorithm (MRMR-WOA) for feature selection, Long Short-Term Memory and Convolutional Neural Networks (CONV-LSTM) for feature extraction, and (Random Forest classifier-based improved quantum particle swarm optimization (RFC-IQPSO) for the prediction model. The approaches opted in this study to overcome the issues faced in previous technologies are tested using different metrics like: Accuracy, Precision, F-measure, Recall, True Positive, and False Positive, to analyze its performance.