Scholarlink

Air pollution, a problem on the agenda of many institutions such as the EU or WHO, seriously affects the quality of human life and the environment. Modeling the concentration of aerosol pollutants such as PM2.5 or PM10 is an actively explored field of study. However, the main efforts are focused on prediction models. There is a limited focus on the reduction in dimensionality of the available data. The main purpose of this paper is to present a modified PCA-based approach to reduce the dimensionality of air pollution data as a prerequisite step for further prediction models. The method has been tested on real data obtained from the Educational Antismog Network (ESA) in Poland.

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

Eigenmaps - Modified PCA-Based Approach for Dimensionality Reduction of Multivariate Time Series of Geospatial Data

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ewa Skubalska-Rafajłowicz,
  • Adam Krzyżak,
  • Michał Piórek

摘要

Air pollution, a problem on the agenda of many institutions such as the EU or WHO, seriously affects the quality of human life and the environment. Modeling the concentration of aerosol pollutants such as PM2.5 or PM10 is an actively explored field of study. However, the main efforts are focused on prediction models. There is a limited focus on the reduction in dimensionality of the available data. The main purpose of this paper is to present a modified PCA-based approach to reduce the dimensionality of air pollution data as a prerequisite step for further prediction models. The method has been tested on real data obtained from the Educational Antismog Network (ESA) in Poland.