Klasifikasi Tingkat Kekeringan Meteorologis di Provinsi Nusa Tenggara Timur dengan Remote Sensing Data Menggunakan Pendekatan Decision Trees-Based Machine Learning

Naufal, Achmad (2019) Klasifikasi Tingkat Kekeringan Meteorologis di Provinsi Nusa Tenggara Timur dengan Remote Sensing Data Menggunakan Pendekatan Decision Trees-Based Machine Learning. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 06211540000002-Undergraduate_Theses.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (5MB) | Request a copy

Abstract

Provinsi Nusa Tenggara Timur merupakan salah satu daerah di Indonesia yang sering terlanda bencana kekeringan. Selain merugikan masyarakat, sebagai pemerintah yang berkewajiban menjamin kesejahteraan rakyatnya ikut mengalami kerugian finansial dalam penanganan bencana kekeringan yang terjadi baik sebagai pemerintah pusat maupun pemerintah daerah. Klasifikasi tingkat kekekeringan berdasarkan SPI-3 bulanan di Nusa Tenggara Timur dilakukan menggunakan metode Classification and Regression Tree dan Random Forest serta oversampling menggunakan Synthetic Minority Oversampling Technique. Data SPI yang digunakan berasal dari dua sumber data yang berbeda yaitu TRMM dan MERRA-2. Metode random forest lebih baik dibandingkan dengan metode CART karena baik pada klasifikasi dengan oversampling maupun tanpa oversampling, random forest memiliki nilai rata-rata AUC pada seluruh grid di wilayah Provinsi Nusa Tenggara Timur lebih besar dibandingkan dengan nilai rata-rata AUC dari metode CART. ================================================================================================
Nusa Tenggara Timur Province is one of the regions in
Indonesia which is often hit by drought. In addition to harming the people, as a government that is obliged to ensure the welfare of its people, it also participates in financial losses in handling drought that occurs both as the central government and the regional government. Classification of the level of drought based on monthly SPI-3 in East Nusa Tenggara is carried out using
Classification and Regression Tree and Random Forest and also using oversampling method Synthetic Minority Oversampling Technique. The random forest method is better than the CART method because both oversampling and non-oversampling classifications, random forest has an average AUC value in the entire grid in the East Nusa Tenggara Province greater than the AUC average value of the CART method. The random forest method is better than the CART method because both oversampling and non oversampling classifications, random forest has an average AUC value in the entire grid in the East Nusa Tenggara Province greater than the AUC average value of the CART method.

Item Type:	Thesis (Undergraduate)
Additional Information:	RSSt 519.536 Nau k-1 2019
Uncontrolled Keywords:	CART, MERRA-2, Random Forest, Remote-sensing Data, TRMM
Subjects:	H Social Sciences > HA Statistics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Naufal Achmad
Date Deposited:	06 Oct 2021 18:46
Last Modified:	06 Oct 2021 18:46
URI:	http://repository.its.ac.id/id/eprint/61063

Actions (login required)

View Item