Application of Soft-Clustering Analysis Using Expectation Maximization Algorithms on Gaussian Mixture Model

Research on soft-clustering has not been explored much compared to hard-clustering. Soft-clustering algorithms are important in solving complex clustering problems. One of the methods is Gaussian Mixture Model (GMM). GMM a method classify data points into different clusters based distribution. This study aims determine number formed by using method. The used this synthetic water quality indicators obtained from Kaggle website. stages are: imputing Not Available (NA) value (if there an NA value), checking distribution, conducting normality test, and standardizing data. next step estimate parameters with Expectation Maximization (EM) algorithm. best biggest Bayesian Information Creation (BIC). results showed that was 3 clusters. Cluster 1 consisted 1110 observations low-quality category, cluster 2 499 medium 1667 high-quality category or acceptable. recommend can be grouped correctly when variables generally normally distributed. applied real data, both which distributed have mixture non-Gaussian.