Scholarlink

<p>The growing frequency and severity of natural disasters underscore the need for advanced decision-support systems capable of generating adaptive, context-aware response recommendations. However, the effectiveness of such systems is often constrained by data heterogeneity and limited situational awareness, limiting first responders’ ability to assess evolving conditions and make data-driven decisions. These limitations can result in inefficiencies across critical operations, including search and rescue, medical assistance, evacuation planning, and resource allocation. In response to these challenges, this study introduces the First Responders System (FiReS), an agent-based reasoning framework for generating adaptive, context-aware recommendations in post-disaster scenarios. The system integrates semantically harmonized disaster data with case-based reasoning, rule-based inference, and probabilistic methods to support informed decision-making under uncertainty and evolving conditions. Recommendations are stored in the RDF repository, forming a structured feedback loop that iteratively enhances response strategies and strengthens the system’s adaptive learning capabilities. The system was validated through a machine learning-driven evaluation of its semantic integration and scenario-based assessment of its adaptive reasoning, confirming its effectiveness in generating context-aware disaster response recommendations.</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

FiReS: an agent based reasoning system for post-disaster response recommendations

  • 期刊论文

  • 发表日期：

    2026年4月13日

查看全文
  • Areti Bania,
  • Omiros Iatrellis,
  • Nicholas Samaras

摘要

The growing frequency and severity of natural disasters underscore the need for advanced decision-support systems capable of generating adaptive, context-aware response recommendations. However, the effectiveness of such systems is often constrained by data heterogeneity and limited situational awareness, limiting first responders’ ability to assess evolving conditions and make data-driven decisions. These limitations can result in inefficiencies across critical operations, including search and rescue, medical assistance, evacuation planning, and resource allocation. In response to these challenges, this study introduces the First Responders System (FiReS), an agent-based reasoning framework for generating adaptive, context-aware recommendations in post-disaster scenarios. The system integrates semantically harmonized disaster data with case-based reasoning, rule-based inference, and probabilistic methods to support informed decision-making under uncertainty and evolving conditions. Recommendations are stored in the RDF repository, forming a structured feedback loop that iteratively enhances response strategies and strengthens the system’s adaptive learning capabilities. The system was validated through a machine learning-driven evaluation of its semantic integration and scenario-based assessment of its adaptive reasoning, confirming its effectiveness in generating context-aware disaster response recommendations.