Scholarlink

<p>Cognitive IoT integrates cognitive capabilities into Internet of Things (IoT) systems but faces challenges in efficiently processing massive, complex data streams. Therefore, this study proposes an asymmetrical regression framework that reduces data inaccuracies via total variation regularization, handles missing data using probabilistic clustering, and retains only the most plausible clusters to minimize complexity. Asymmetrical pairs from these clusters feed a linear regression model for sensor data forecasting. Evaluated on 21.25&#xa0;years of environmental data, the method achieves an accuracy of over 99.45%, outperforming traditional models in both efficiency and precision.</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

Asymmetrical regression: a cognitively driven approach to advanced forecasting in cognitive internet of things

  • 期刊论文

  • 发表日期：

    2025年10月3日

查看全文
  • Vidyapati Jha,
  • Priyanka Tripathi

摘要

Cognitive IoT integrates cognitive capabilities into Internet of Things (IoT) systems but faces challenges in efficiently processing massive, complex data streams. Therefore, this study proposes an asymmetrical regression framework that reduces data inaccuracies via total variation regularization, handles missing data using probabilistic clustering, and retains only the most plausible clusters to minimize complexity. Asymmetrical pairs from these clusters feed a linear regression model for sensor data forecasting. Evaluated on 21.25 years of environmental data, the method achieves an accuracy of over 99.45%, outperforming traditional models in both efficiency and precision.