Ayundari, Iftitah (2019) Penentuan Zona Musim di Mojokerto Menurut Karakteristik Curah Hujan Dengan Metode Time Series Based Clustering. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211540000039-Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Zona musim adalah daerah yang pos hujan rata ratanya memiliki perbedaan yang jelas antara periode musim hujan dan musim kemarau. Metode analisis data yang diterapkan dalam pengelompokkan ZOM ini adalah metode analisis cluster, yaitu mengelompokkan suatu objek yang memiliki kesamaan sifat tanpa menghilangkan struktur alami objek. Oleh karena kelompok yang dihasilkan memiliki makna, seperti pola atau klasifikasi. Data curah hujan merupakan data series sehingga dalam melakukan analisis cluster menggunakan jarak basis time series, yaitu pada penelitian ini menggunkan jarak Dynamic Time Warping (DTW) dan autocorrelation function (ACF). Tujuannya adalah mengetahui karateristik zona musim di seluruh wilayah Mojokerto dan memetakan pola zona musim di Mojokerto berdasarkan data curah hujan menggunakan time series based clustering. Pengelompokan dilakukan dengan membentuk 2 hingga 5 kelompok, dan selanjutnya akan dilakukan pemilihan kelompok optimum dengan pseudo-f statistics dan kriteria silhouette. Data curah hujan yang digunakan adalah curah hujan bulanan. Evaluasi kinerja zona musim dilakukan dengan rasio simpangan baku dimana metode yang dibandingkan adalah pengelompokkan dengan jarak ACF, DTW, dan hasil pengelompokkan oleh BMKG. Berdasarkan hasil evaluasi kinerja zona musim dengan rasio simpangan baku metode terbaik yang dipilih adalah pengelompokkan dengan jarak DTW
=================================================================================================================================
Seasonal zones are areas where the average rainfall post has a clear difference between the period of the rainy season and the dry season. The data analysis method applied in ZOM grouping is cluster analysis method, which is grouping objects that have similar properties without removing the natural structure of the object. Because the groups produced have meaning, such as patterns or classifications. Rainfall data is a data series so that in conducting cluster analysis using time series base distance, that is in this study using the Dynamic Time Warping (DTW) distance and autocorrelation function (ACF). The aim is to find out the characteristics of the season zones in all Mojokerto regions and to map the pattern of the season zones in Mojokerto based on rainfall data using time series based clustering. Grouping is done by forming 2 to 5 groups, and then the optimum group selection will be carried out with pseudo-f statistics and silhouette criteria. The rainfall data used is monthly rainfall. Evaluation of the season zone performance is carried out with a standard deviation ratio where the method compared is grouping with distance ACF, DTW, and the results of grouping by BMKG. Based on the results of the evaluation of the season zone performance with the standard deviation ratio the best method chosen is grouping with distance DTW
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.55 Ayu p-1 2019 |
| Uncontrolled Keywords: | Curah Hujan, Clustering, Mojokerto, Zona Musim |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA30.3 Time-series analysis Q Science > QA Mathematics > QA278.55 Cluster analysis |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Ayundari Iftitah |
| Date Deposited: | 21 Feb 2024 08:10 |
| Last Modified: | 21 Feb 2024 08:10 |
| URI: | http://repository.its.ac.id/id/eprint/64199 |
Actions (login required)
 |
View Item |