The application of bivariate polar plots and k-means clustering to analysis air pollution in Taoyuan, Taiwan
Keywords:
Openair, Air Quality, Taoyuan, Bivariate polar plots, k-means clustering.Abstract
In this paper, we apply k-means clustering techniques directly to bivariate polar plots to identify and group
similar features. Bivariate polar plots method is one of the tools in open-air packagefor source detection and
characterisation. Bivariate polar plots provide an effective graphical means of discriminating different source types and
characteristics. Importantly, this paper links identified clusters to known emission characteristics to confirm the
inferences made in the analysis.The combination between k-means clustering and bivariate polar plots helps to avoid
making arbitrary decisions about how to extract and analyse different source features. Using this approach, we initially
understand and properly evaluate the level of pollution to improve air quality in Taoyuan city. This work not only
provides more cases for the approach to analysis of air pollution monitoring data but also use as a reference for further
research.
