Scholarlink

<p>During the pandemic, education made a rapid shift from traditional classrooms to online platforms. This change really highlighted the need for effective strategies to keep students engaged in a digital environment. This research addresses the challenges, like the lack of protection of sensitive data, which is accessed by unauthorized individuals and the linkage of users for developing a novel student marks management system. A new model called Leopard Seal K-anonymity Privacy-preserved Hyper-ledger Blockchain Model (LK-APHBM) is introduced. This approach integrates the Leopard Seal Optimization Algorithm (LSOA) for ensuring strong K-anonymity by balancing data utility. The system incorporates data preprocessing, feature selection using the Improved Single Candidate Optimizer (ISCO) and a Memristive Cyclic Hopfield Convoluted Hypergraph (MCHCH)-based deep learning framework to capture complex relationships in sequential mark data. Furthermore, a novel approach integrates the Bootstrap aggregating Multivariate adaptive regression splines Bayesian Local Interpretable Model-agnostic Explanations (BMB-LIME) is a hybrid explanatory module that cleverly combines bagging, Multivariate Adaptive Regression Splines (MARS), and Bayesian analysis. This innovative approach aims to deliver explanations of model predictions that are statistically and locally accurate for student mark management. The proposed LK-APHBM model demonstrates superior performance compared to existing approaches, achieving an F1-score of 97.67%, precision of 99.12%, recall of 99.45%, and accuracy of 98.23%. Additionally, this minimizes the total time, making it highly efficient for real-time implementation. The LK-APHBM model offers a privacy-focused resolution for managing student marks in online education. Additionally, this approach exhibits quicker computation times compared to previous techniques, highlighting sustainability and efficiency for various privacy-preserving student mark management.</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

Leopard Seal K-Anonymity Privacy-Preserved Hyper-Ledger Blockchain Integrated with Deep Learning for Student Mark Management

  • 期刊论文

  • 发表日期：

    2025年11月18日

查看全文
  • Rima Anantkumar Patel,
  • Dharmendra T. Patel

摘要

During the pandemic, education made a rapid shift from traditional classrooms to online platforms. This change really highlighted the need for effective strategies to keep students engaged in a digital environment. This research addresses the challenges, like the lack of protection of sensitive data, which is accessed by unauthorized individuals and the linkage of users for developing a novel student marks management system. A new model called Leopard Seal K-anonymity Privacy-preserved Hyper-ledger Blockchain Model (LK-APHBM) is introduced. This approach integrates the Leopard Seal Optimization Algorithm (LSOA) for ensuring strong K-anonymity by balancing data utility. The system incorporates data preprocessing, feature selection using the Improved Single Candidate Optimizer (ISCO) and a Memristive Cyclic Hopfield Convoluted Hypergraph (MCHCH)-based deep learning framework to capture complex relationships in sequential mark data. Furthermore, a novel approach integrates the Bootstrap aggregating Multivariate adaptive regression splines Bayesian Local Interpretable Model-agnostic Explanations (BMB-LIME) is a hybrid explanatory module that cleverly combines bagging, Multivariate Adaptive Regression Splines (MARS), and Bayesian analysis. This innovative approach aims to deliver explanations of model predictions that are statistically and locally accurate for student mark management. The proposed LK-APHBM model demonstrates superior performance compared to existing approaches, achieving an F1-score of 97.67%, precision of 99.12%, recall of 99.45%, and accuracy of 98.23%. Additionally, this minimizes the total time, making it highly efficient for real-time implementation. The LK-APHBM model offers a privacy-focused resolution for managing student marks in online education. Additionally, this approach exhibits quicker computation times compared to previous techniques, highlighting sustainability and efficiency for various privacy-preserving student mark management.