Scholarlink

Air pollution, especially in enclosed spaces, poses serious health risks due to everyday activities like cooking and cleaning. Poor indoor air quality can lead to conditions such as Sick Building Syndrome (SBS), highlighting the need for advanced predictive models. Kolmogorov-Arnold Networks (KAN) provide an innovative solution for predicting pollutants such as CO2, TVOC, PM2.5, and PM10 using historical and real-time data. This study applies KANs to forecast pollution risk levels and demonstrates their potential for integration with IoT technologies to enable continuous, precise monitoring for safer indoor environments.

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

Sick Building Syndrome and Indoor Air Quality: Leveraging Kolmogorov-Arnold Networks for Predictive Pollutant Control

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Rodrigo Gabriel Linares Alzamora,
  • Paulo Nazareno Maia Sampaio,
  • Laura Margarita Rodríguez Peralta,
  • Ariel Isaac Posada Barrera,
  • Éldman de Oliveira Nunes

摘要

Air pollution, especially in enclosed spaces, poses serious health risks due to everyday activities like cooking and cleaning. Poor indoor air quality can lead to conditions such as Sick Building Syndrome (SBS), highlighting the need for advanced predictive models. Kolmogorov-Arnold Networks (KAN) provide an innovative solution for predicting pollutants such as CO2, TVOC, PM2.5, and PM10 using historical and real-time data. This study applies KANs to forecast pollution risk levels and demonstrates their potential for integration with IoT technologies to enable continuous, precise monitoring for safer indoor environments.