Spatial Temporal Density Based Spatial Cluster Of Application With Noise Untuk Mengelompokkan Curah Hujan Di Jawa Timur

Zenklinov, Amanatullah Pandu (2024) Spatial Temporal Density Based Spatial Cluster Of Application With Noise Untuk Mengelompokkan Curah Hujan Di Jawa Timur. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6003212009_Master Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 March 2026. Download (2MB) | Request a copy

Abstract

Zona musim (ZOM) adalah daerah yang pos hujan rata ratanya memiliki perbedaan yang jelas antara periode musim hujan dan musim kemarau. Provinsi Jawa Timur merupakan salah satu provinsi yang mengalami variasi curah hujan yang tinggi, yang dapat memengaruhi sektor pertanian, sumber daya air, dan lingkungan. Zona musim dilakukan pengelompokan agar setiap wilayahnya memiliki karakteristik yang sama. Metode analisis data yang digunakan adalah metode analisis cluster. Penelitian ini bertujuan untuk mengelompokkan curah hujan di Jawa Timur menggunakan algoritma Spatial Temporal DBSCAN (ST-DBSCAN) dengan jarak euclidean dan dynamic time warping (DTW), yang memungkinkan pemahaman yang lebih baik tentang pola curah hujan. Dilakukan perbandingan algoritma ST-DBSCAN dengan jarak euclidean dan ST-DBSCAN dengan jarak DTW guna mendapatkan hasil cluster terbaik. Hasil pengelompokan di evaluasi menggunakan silhouette index, davies-bouldin index, dan calinski-harabasz index. Hasil evaluasi cluster menunjukkan bahwa ST-DBSCAN dengan jarak DTW mendapatkan hasil yang lebih optimal dari pada ST-DBSCAN dengan jarak euclidean. Pola hujan di Jawa Timur pada musim penghujan yaitu bulan november hingga maret dan musim kemarau yaitu bulan april hingga oktober. Terdapat 12 cluster yang terbentuk dengan artian bahwa curah hujan di Jawa Timur sangat variatif.
====================================================================================================================================
Seasonal zone (SZ) is an area where the average rainfall differs significantly between the rainy season and the dry season periods. East Java Province is one of the regions experiencing high variations in rainfall, which can impact agriculture, water resources, and the environment. The zoning of seasons is conducted to ensure uniform characteristics within each region. The data analysis method employed is cluster analysis. This study aims to cluster rainfall in East Java using the Spatial Temporal DBSCAN (ST-DBSCAN) algorithm with both Euclidean distance and Dynamic Time Warping (DTW), facilitating a better understanding of rainfall patterns. A comparison between the ST-DBSCAN algorithm with Euclidean distance and ST-DBSCAN with DTW distance is conducted to obtain the best cluster results. The clustering results are evaluated using the silhouette index, Davies-Bouldin index, and Calinski-Harabasz index. The cluster evaluation results indicate that ST-DBSCAN with DTW distance yields more optimal results compared to ST-DBSCAN with Euclidean distance. Rainfall patterns in East Java occur during the rainy season from November to March and the dry season from April to October. Twelve clusters are formed, signifying the high variability of rainfall in East Java.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Cluster, Curah Hujan, DBSCAN, ST-DBSCAN, Cluster, DBSCAN, Rainfall, ST-DBSCAN
Subjects:	Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics) S Agriculture > S Agriculture (General) > S600.7.R35 Rain and rainfall
Divisions:	Faculty of Mathematics, Computation, and Data Science > Statistics > 49101-(S2) Master Thesis
Depositing User:	Amanatullah Pandu Zenklinov
Date Deposited:	19 Feb 2024 08:41
Last Modified:	19 Feb 2024 08:41
URI:	http://repository.its.ac.id/id/eprint/107600

Actions (login required)

View Item