Data mining-based indoor air quality monitoring system in the era of big data

2018 International Conference on Air Pollution and Environmental Engineering

生态环境科学

Yuan, Lina^1 ; Chen, Huajun^1 ; Tian, Bo^1 ; Gong, Ging^2

College of Data Science, Tongren University, Tongren, China^1

Graduate School, Tongren University, Tongren, China^2

关键词: Data mining methods;    Indoor air quality;    Indoor air quality monitoring;    Indoor environment;    Indoor environment parameters;    Kernel principal component analyses (KPCA);    Nonlinear feature extraction;    Recognition accuracy;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/208/1/012047/pdf
DOI  :  10.1088/1755-1315/208/1/012047

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

With the ever-increasing enhancement of living standards, indoor air quality has been drawing more and more attention, due to the conditions of indoor environment with significant relevance to human growth development, health and work efficiency. The quality of indoor environment has a direct impact on the quality of people's life, and even concerns the issue of human survival. In consequence, the pervasive and exponentially increasing indoor air data presents imminent challenges to how to monitor and evaluate the indoor environment parameters, and create a comfortable, quiet and clean indoor environment, which is indispensable based on data mining methods in the context of big data. This paper focuses on two methods of nonlinear feature extraction: Kernel principal component analysis (KPCA) and PCA. Experimental results demonstrate that the recognition accuracy of feature extraction-based KPCA is superior to that of PCA.