Scholarlink

<p>Smart sensors, actuators, and the Industrial Internet of Things (IIoT) uses additional devices to enhance manufacturing and industrial operations. Secure High Order Clustering Algorithm by Quick Search, or SHOCFS was proposed to address this issue where primary components are optimum density peak detections. The algorithm’s choice of density points is constant and significant concern. To overcome this issue, proposed work introduced SEAHO (Secure Energy Aware High-Order) clustering where inputs are intelligent grid datasets for feature selections with secure clustering. Proposed work steps are (1) data encryption using Paillier cryptosystem, (2) SA-Bi-LSTM (Supervised Attention based Bi-Directional Long Short-Term Memory) feature selection to choose best features from IIoT, (3) Dynamic Programming with Gaussian Process Regression (DPGPR) to achieve secure energy conscious communication, (4) SEAHO clustering algorithm by fast search with Decision Tree based Salp Swarm Optimization (SEAHO-DT-SWO) to get optimal density peaks on hybrid clouds in IIoT, (5) Finally, electricity demand forecasting. The results of experiments that proposed technique can help in accurate and effective clustering of big data compared to other techniques. The proposed approach evaluation by metrics like quality of clustering centers, RI (Rand Indices), speedup ratios, and Encryption Times where results demonstrate proposed method’s superiority in scalability and clustering efficiency when compared to other methods.</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

Efficient Resource Allocation in Cloud-Based IIOT Through Secure Energy-Aware High-Order Optimal Density Selection Using Supervised Attention-Based Bi-directional Long Short-Term Memory and Decision Tree-Based Salp Swarm Optimization

  • 期刊论文

  • 发表日期：

    2026年3月28日

查看全文
  • G. Kanagaraj,
  • S. Lekashri,
  • A. N. Duraivel,
  • R. Ramya

摘要

Smart sensors, actuators, and the Industrial Internet of Things (IIoT) uses additional devices to enhance manufacturing and industrial operations. Secure High Order Clustering Algorithm by Quick Search, or SHOCFS was proposed to address this issue where primary components are optimum density peak detections. The algorithm’s choice of density points is constant and significant concern. To overcome this issue, proposed work introduced SEAHO (Secure Energy Aware High-Order) clustering where inputs are intelligent grid datasets for feature selections with secure clustering. Proposed work steps are (1) data encryption using Paillier cryptosystem, (2) SA-Bi-LSTM (Supervised Attention based Bi-Directional Long Short-Term Memory) feature selection to choose best features from IIoT, (3) Dynamic Programming with Gaussian Process Regression (DPGPR) to achieve secure energy conscious communication, (4) SEAHO clustering algorithm by fast search with Decision Tree based Salp Swarm Optimization (SEAHO-DT-SWO) to get optimal density peaks on hybrid clouds in IIoT, (5) Finally, electricity demand forecasting. The results of experiments that proposed technique can help in accurate and effective clustering of big data compared to other techniques. The proposed approach evaluation by metrics like quality of clustering centers, RI (Rand Indices), speedup ratios, and Encryption Times where results demonstrate proposed method’s superiority in scalability and clustering efficiency when compared to other methods.