Scholarlink

<p>With the rapid advancement of information technology, incomplete data has become prevalent in real-world applications. In practical scenarios, such data is often subject to dynamic changes and continuous evolution. From the perspective of granular computing, how to acquire the valuable knowledge in the context of dynamic incomplete data presents a significant challenge. In this study, we investigate a matrix-based incremental method with the aim of updating incomplete multigranulation three-way regions when adding objects. In incomplete multigranulation spaces, matrix-based operations are introduced to construct the three-way regions. To accommodate dynamically increasing objects, we employ incremental mechanisms for efficient updates of the relation matrix and intermediate matrices. Moreover, an incremental algorithm is specifically designed to handle the addition of new objects. To verify the efficiency of our incremental algorithm, we carry out the appropriate experiments through the utilization of several public datasets. Finally, the comparative analysis confirms the competitive performance of our incremental algorithm.</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

Matrix-based method to update incomplete multigranulation three-way regions with increasing objects

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月29日

查看全文
  • Chengxiang Hu,
  • Haoran Zhang,
  • Xingpeng Kuai,
  • Xiaoling Huang,
  • Xiaojing Hu

摘要

With the rapid advancement of information technology, incomplete data has become prevalent in real-world applications. In practical scenarios, such data is often subject to dynamic changes and continuous evolution. From the perspective of granular computing, how to acquire the valuable knowledge in the context of dynamic incomplete data presents a significant challenge. In this study, we investigate a matrix-based incremental method with the aim of updating incomplete multigranulation three-way regions when adding objects. In incomplete multigranulation spaces, matrix-based operations are introduced to construct the three-way regions. To accommodate dynamically increasing objects, we employ incremental mechanisms for efficient updates of the relation matrix and intermediate matrices. Moreover, an incremental algorithm is specifically designed to handle the addition of new objects. To verify the efficiency of our incremental algorithm, we carry out the appropriate experiments through the utilization of several public datasets. Finally, the comparative analysis confirms the competitive performance of our incremental algorithm.