Scholarlink

<p>Intelligent analysis of the enormous amount of heterogeneous data often produced by many connected devices is the primary focus of current Internet of Things (IoT) research. Thus, the analysis of massive heterogeneous data requires the insertion of cognition into IoT architecture, which in turn causes the emergence of a new area called cognitive IoT (CIoT). Several applications in cognitive IoT need a recommendation from the intelligent analysis of massive heterogeneous data. Therefore, this research proposes a recommendation system in which the novelty of the proposed method lies in the two spheres-(i) to manage the large amounts of heterogeneous data, which are further categorized using model-based clustering, the most suitable copula is designed. The next step is to calculate the entropy of each cluster by adding together all the information linked to each sensory observation for each element in the cluster, and (ii) The <InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11276_2025_3945_Article_IEq2.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">\(\pi \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>π</mi> </math></EquationSource> </InlineEquation>-tree is designed for knowledge representation that is like a general tree of O(n) time complexity. At the top of the π-tree is the cluster with the highest positive entropy value, and below it is the cluster with the lower&#xa0;entropy value. The level-order entropy traversal is used for the recommendation system. The carbon monoxide (CO) data from six months is used to evaluate the proposed system experimentally, revealing the efficacy of the proposed system over competing approaches.</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

\({\varvec{\pi}}\)-tree based knowledge representation and recommendation system in cognitive IoT

  • 期刊论文

  • 发表日期：

    2025年3月25日

查看全文
  • Vidyapati Jha,
  • Priyanka Tripathi

摘要

Intelligent analysis of the enormous amount of heterogeneous data often produced by many connected devices is the primary focus of current Internet of Things (IoT) research. Thus, the analysis of massive heterogeneous data requires the insertion of cognition into IoT architecture, which in turn causes the emergence of a new area called cognitive IoT (CIoT). Several applications in cognitive IoT need a recommendation from the intelligent analysis of massive heterogeneous data. Therefore, this research proposes a recommendation system in which the novelty of the proposed method lies in the two spheres-(i) to manage the large amounts of heterogeneous data, which are further categorized using model-based clustering, the most suitable copula is designed. The next step is to calculate the entropy of each cluster by adding together all the information linked to each sensory observation for each element in the cluster, and (ii) The \(\pi \) π -tree is designed for knowledge representation that is like a general tree of O(n) time complexity. At the top of the π-tree is the cluster with the highest positive entropy value, and below it is the cluster with the lower entropy value. The level-order entropy traversal is used for the recommendation system. The carbon monoxide (CO) data from six months is used to evaluate the proposed system experimentally, revealing the efficacy of the proposed system over competing approaches.