Scholarlink

<p>In the realm of interdisciplinary group decision-making for product design, identifying the core topics of knowledge exchange and visualizing the extent of interdisciplinarity are crucial for revealing focal points and potential conflicts within the decision-making process. This study addresses the challenge of structuring interdisciplinary knowledge in engineering product design by proposing a novel model that integrates domain-specific embedding and Attention mechanism for named entity recognition. The model proposed refines interdisciplinary engineering context elements and develops an embedding layer with domain-specific features to enhance the precision of knowledge entity extraction. The model quantifies the academic association among entities within the same context, utilizing weighted mutual information to build a knowledge network with contextual attributes. Applied to ship cabin design decision-making dialogues, the method extracts knowledge application contexts and assesses interdisciplinary degree, revealing core themes and potential conflicts. The approach effectively encapsulates engineering knowledge within disciplinary attributes, visualizes the exchange of disciplinary knowledge, and thereby enhances group decision-making efficiency and performance.</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

Knowledge Application Context Mining and Knowledge Network Construction for Interdisciplinary Group Decision Product Design

  • 期刊论文

  • 发表日期：

    2025年2月12日

查看全文
  • Kexin Cheng,
  • Zuhua Jiang,
  • Lebao Wu

摘要

In the realm of interdisciplinary group decision-making for product design, identifying the core topics of knowledge exchange and visualizing the extent of interdisciplinarity are crucial for revealing focal points and potential conflicts within the decision-making process. This study addresses the challenge of structuring interdisciplinary knowledge in engineering product design by proposing a novel model that integrates domain-specific embedding and Attention mechanism for named entity recognition. The model proposed refines interdisciplinary engineering context elements and develops an embedding layer with domain-specific features to enhance the precision of knowledge entity extraction. The model quantifies the academic association among entities within the same context, utilizing weighted mutual information to build a knowledge network with contextual attributes. Applied to ship cabin design decision-making dialogues, the method extracts knowledge application contexts and assesses interdisciplinary degree, revealing core themes and potential conflicts. The approach effectively encapsulates engineering knowledge within disciplinary attributes, visualizes the exchange of disciplinary knowledge, and thereby enhances group decision-making efficiency and performance.