A Framework to Spatially Cluster Air Quality Monitoring Stations in Peninsular Malaysia using the Hybrid Clustering Method
Multiple variables must be analyzed in order to assess air quality trends. It turns into a multidimensional issue that calls for dynamic methods. In order to provide an improved spatial cluster distribution with distinct validation, this study set out to illustrate the hybrid cluster method in air q...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universiti Teknologi Malaysia
2023
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/43267/3/A%20Framework%20-%20Copy.pdf http://ir.unimas.my/id/eprint/43267/ https://mjfas.utm.my/index.php/mjfas/article/view/2620 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Multiple variables must be analyzed in order to assess air quality trends. It turns into a multidimensional issue that calls for dynamic methods. In order to provide an improved spatial cluster distribution with distinct validation, this study set out to illustrate the hybrid cluster method in air quality monitoring stations in Peninsular Malaysia. The Department of Environment, Malaysia (DOE), provided the data set, which covered the two-year period from 2018 to 2019. This study included six air quality pollutants: PM10, PM2.5, SO2, NO2, O3, and CO. Principal component analysis (PCA), a multivariate technique, was used to condense the information found in enormous data tables in order to better comprehend the variables (to reduce dimensionality) prior to grouping the data. The PCA factor scores were then used to produce the AHC. The clusters were validated using discriminant analysis (DA). 36 of 47 stations required additional analysis using AHC, according to the PCA factor scores. Low Polluted Region (LPR = seven stations), Moderate Polluted Region (MPR = 20 stations), and High Polluted Region (HPR = nine stations) were created from AHC and share the same characteristics. The DA results showed 84 % correct classification rate for the clusters. With regard to identifying and categorizing stations according to air quality characteristics, the framework presented here offers an improved method. This illustrates that the hybrid cluster method utilized in this work can produce a new method of pollutant distributions that is helpful in air pollution investigations. |
|---|