ANALISIS PERBANDINGAN ALGORITMA XGBOOST DAN SUPPORT VECTOR MACHINE DALAM KLASIFIKASI AKTIVITAS GEMPA VULKANIK BERDASARKAN DATA SEISMIK (STUDI KASUS BPPTKG)

Fansuri, Mochammad Farhan (2025) ANALISIS PERBANDINGAN ALGORITMA XGBOOST DAN SUPPORT VECTOR MACHINE DALAM KLASIFIKASI AKTIVITAS GEMPA VULKANIK BERDASARKAN DATA SEISMIK (STUDI KASUS BPPTKG). Skripsi thesis, UPN "Veteran" Yogyakarta.

	Text COVER.pdf Download (283kB)
	Text ABSTRAK.pdf Download (171kB)
	Text DAFTAR ISI.pdf Download (487kB)
	Text DAFTAR PUSTAKA.pdf Download (370kB)
	Text HALAMAN PENGESAHAN PEMBIMBING.pdf Download (719kB)
	Text NASKAH TA LENGKAP.pdf Restricted to Repository staff only Download (5MB)

Official URL: upnyk.ac.id

Abstract

The identification and classification of volcanic earthquake activity require advanced
and accurate methods. Machine learning is one of the approaches that can be utilized to classify
types of volcanic earthquake activity based on seismic data. However, selecting an appropriate
machine learning model is crucial to obtain optimal results. This study aims to compare the
performance of the XGBoost and Support Vector Machine (SVM) algorithms in classifying
volcanic earthquake activity using seismic data from BPPTKG, in order to determine the most
effective model.
The research methodology includes problem analysis, data collection, data processing,
model development, model evaluation, and presentation of results. The data used are in
MiniSEED format, and data balancing is carried out using the hybrid sampling method
SMOTE-RUS. The results show that the XGBoost model achieved an accuracy of 74%, which
is higher than the linear kernel SVM with 69% accuracy. Additionally, SVM with RBF and
polynomial kernels each achieved an accuracy of 68%.
The implementation of SMOTE-RUS proved to have a significant impact on model
accuracy and training time, where models applying SMOTE-RUS achieved higher accuracy
although with relatively slower training time. Based on these results, it can be concluded that
the combination of the XGBoost algorithm and the SMOTE-RUS hybrid sampling technique is
effective in classifying volcanic earthquake activity based on seismic data in the BPPTKG case
study.
Keywords: SVM; XGBoost; BPPTKG, seismic data, volcanic eathquake

Item Type:	Tugas Akhir (Skripsi)
Additional Information:	Mochammad Farhan Fansuri (Penulis - 123200127); Awang Hendrianto Pratomo (Pembimbing)
Uncontrolled Keywords:	SVM; XGBoost; BPPTKG, seismic data, volcanic eathquake
Subjek:	T Technology > T Technology (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Fakultas Teknik Industri > (S1) Informatika
Depositing User:	UPA Perpustakaan
Date Deposited:	14 Oct 2025 01:54
Last Modified:	14 Oct 2025 01:56
URI:	http://eprints.upnyk.ac.id/id/eprint/44304

Actions (login required)

View Item