Clustering rainfall data using Piccolo method

Abstract

Clustering is an approach to data analysis to organize data into similar groups. Model-based clustering is one of the methods that can be used in time series data clustering. This study aims to cluster rainfall data using Piccolo distance and modeling for individual and cluster levels. The parameters of the model were used to calculate the distance between time series data. The distance between time series data was calculated by the Piccolo distance, then the Ward method was used for clustering. The number of clusters determined by the average of Silhouette value. Rainfall data for the individual and the cluster level was modeled with the ARIMA model. The RMSE of the individual and cluster models was compared. In this study, the clustering of rainfall data was carried out in the West Java region. The results obtained there are eight clusters of rainfall data. The RMSE of the cluster model was not different from the individual model. The model with clustering is more effective and it can be used to represent individual models in the cluster.