Using k-Means and Self Organizing Maps in Clustering Air Pollution Distribution in Makassar City, Indonesia

Abstract

Air pollution is an important environmental problem for specific areas, including Makassar City, Indonesia. The increase should be monitored and evaluated, especially in urban areas that are dense with vehicles and factories. This is a challenge for local governments in urban planning and policy-making to fulfill the information about the impact of air pollution. The clustering of starting points for the distribution areas can ease the government to determine policies and prevent the impact. The k-Means initial clustering method was used while the Self-Organizing Maps (SOM) visualized the clustering results. Furthermore, the Geographic Information System (GIS) visualized the results of regional clustering on a map of Makassar City. The air quality parameters used are Suspended Particles (TSP), Sulfur Dioxide (SO2), Nitrogen Dioxide (NO2), Carbon Monoxide (CO), Surface Ozone (O3), and Lead (Pb) which are measured during the day and at night. The results showed that the air contains more CO, and at night, the levels are reduced in some areas. Therefore, the density of traffic, industry and construction work contributes significantly to the spread of CO. Air conditions vary, such as high CO levels during the day and TSP at night. Also, there is a phenomenon at night that a group does not have SO2 and O3 simultaneously. The results also show that the integration of k-Means and SOM for regional clustering can be appropriately mapped through GIS visualization.

Jurnal Varian Full Text

“…This algorithm is the result of the development of the k-means algorithm (Mau and Huynh, 2021) (Ahmad and Dey, 2011) to handle clustering on data with mixed numeric and categorical type attributes (Dinh et al, 2021). The development carried out by Huang maintains the efficiency of the k-means algorithm in dealing with large data and can be applied to numerical and categorical data (Annas et al, 2022). The basic development of the k-prototypes algorithm is in measuring the similarity between the object and its centroid prototype (Pham et al, 2011).…”

“…Another challenge encountered is the type of variable that characterizes the objects (Li et al, 2019). Characteristics of objects consisting of numerical variables are measured by Euclid distance as in the k-means algorithm (Annas et al, 2022). Furthermore, the characteristics of objects consisting of categorical variables can be measured using the mode, the smaller the value of the mode, the more similar objects are and vice versa.…”

Clustering Couples of Childbearing Age to Get Family Planning Counseling Using K-Means Method

“…Table 6 shows the comparison between optimized K-Means with the KKZ algorithm and regular K-Means. It shows that the ideal cluster of K-Means with optimization in Madiun City with 4 clusters and 0.42 average silhouette coefficient value perform better than the improved K-Means without optimization for Home Industry dataset with 3 cluster and average silhouette coefficient of 0.36 in Bangka Belitung Province [35], and traditional K-Means for Air Pollution with 4 clusters and 0.28 average silhouette coefficient value in Makassar City [36].…”

Pengelompokan Jenis Penerimaan Pajak di Kota Makassar Menggunakan Fuzzy Clustering

“…Clustering adalah sebuah proses untuk mengelompokan data ke dalam beberapa cluster atau kelompok sehingga data dalam satu cluster memiliki tingkat kemiripan yang maksimum dan data antar cluster memiliki kemiripan yang minimum [1,2,3,4,5]. Kemiripan yang dimaksud merupakan pengukuran secara numerik terhadap dua buah objek.…”