PENGGUNAAN DEEP LEARNING UNTUK KLASIFIKASI JENIS GEMPA VULKANIK GUNUNG SINABUNG

AL GIBRAN RAYA ALIEFIO SANTOSO, . (2026) PENGGUNAAN DEEP LEARNING UNTUK KLASIFIKASI JENIS GEMPA VULKANIK GUNUNG SINABUNG. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text File Cover.pdf Download (892kB)
	Text Bab 1.pdf Download (260kB)
	Text Bab 2.pdf Restricted to Registered users only Download (723kB) | Request a copy
	Text Bab 3.pdf Restricted to Registered users only Download (579kB) | Request a copy
	Text Bab 4.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text Bab 5.pdf Restricted to Registered users only Download (258kB) | Request a copy
	Text Daftar Pustaka.pdf Download (224kB)
	Text Lampiran dan Riwayat Hidup.pdf Restricted to Registered users only Download (260kB) | Request a copy

Abstract

AL GIBRAN RAYA ALIEFIO SANTOSO. PENGGUNAAN DEEP LEARNING UNTUK KLASIFIKASI JENIS GEMPA VULKANIK GUNUNG SINABUNG. Di bawah bimbingan Dr. Bambang Heru Iswanto, M.Si. dan Dr. Mohammad Hasib, M.Sc. Aktivitas vulkanik Gunung Sinabung menghasilkan data seismik dalam jumlah besar sehingga diperlukan metode analisis otomatis yang mampu mengidentifikasi karakteristik sinyal seismik dari berbagai event gempa vulkanik sinyal secara efisien. Penelitian ini bertujuan menganalisis karakteristik sinyal seismik Gunung Sinabung, mengekstraksi fitur menggunakan CNN AutoEncoder, serta mengevaluasi kualitas pengelompokan event gempa menggunakan K-Means Clustering. Data penelitian berupa rekaman sinyal seismik yang dipraproses melalui deteksi dan segmentasi event menggunakan metode Short-Term Average/Long-Term Average (STA/LTA). Setiap event direpresentasikan dalam empat jenis fitur, yaitu waveform, spectrogram, Mel-Frequency Cepstral Coefficients (MFCC), dan Fast Fourier Transform (FFT) spectrum yang digabungkan menjadi dataset multikanal sebagai masukan CNN AutoEncoder. Fitur laten yang dihasilkan kemudian dikelompokkan menggunakan K-Means dan dievaluasi menggunakan Silhouette Score, Davies-Bouldin Index (DBI), dan Calinski-Harabasz Index (CHI). Hasil penelitian menunjukkan bahwa representasi fitur laten yang dihasilkan CNN AutoEncoder mampu menghasilkan kualitas cluster yang lebih baik dibandingkan penggunaan fitur mentah, sehingga lebih efektif dalam membedakan karakteristik berbagai jenis event gempa vulkanik. Pendekatan ini berpotensi mendukung pengembangan sistem pemantauan aktivitas gunung api secara otomatis dan efisien. Kata kunci: gempa vulkanik, Gunung Sinabung, CNN AutoEncoder, K-Means Clustering, deep learning, seismik. ***** AL GIBRAN RAYA ALIEFIO SANTOSO. DEEP LEARNING-BASED CLASSIFICATION OF VOLCANIC EARTHQUAKE TYPES AT MOUNT SINABUNG. Supervised by Dr. Bambang Heru Iswanto, M.Si. and Dr. Mohammad Hasib, M.Sc. The large volume of seismic data generated by volcanic activity at Mount Sinabung requires automated analysis methods to support efficient monitoring. This study aims to analyze the characteristics of seismic signals associated with different volcanic earthquake events, extract latent features using a Convolutional Neural Network (CNN) AutoEncoder, and evaluate the quality of earthquake event clustering using the K-Means algorithm. The seismic data were preprocessed through event detection and segmentation using the Short-Term Average/Long-Term Average (STA/LTA) method. Each event was represented by four signal features, namely waveform, spectrogram, Mel-Frequency Cepstral Coefficients (MFCC), and Fast Fourier Transform (FFT) spectrum, which were combined into a multi-channel dataset as the input to the CNN AutoEncoder. The extracted latent features were then clustered using K-Means and evaluated using the Silhouette Score, Davies-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI). The results indicate that the latent feature representation learned by the CNN AutoEncoder produces higher-quality clusters than raw features, enabling better discrimination of different volcanic earthquake events. These findings demonstrate that the combination of CNN AutoEncoder and K-Means clustering is an effective approach for automated seismic data analysis and has the potential to support volcano monitoring systems. Keywords: volcanic earthquakes, Mount Sinabung, CNN AutoEncoder, K-Means clustering, deep learning, seismic data.

Item Type:	Thesis (Sarjana)
Additional Information:	1) Dr. Bambang Heru Iswanto, M.Si. , 2) Dr. Mohammad Hasib, M.Sc.
Subjects:	Sains > Fisika
Divisions:	FMIPA > S1 Fisika
Depositing User:	Users 36202 not found.
Date Deposited:	13 Aug 2026 01:33
Last Modified:	14 Aug 2026 07:40
URI:	http://repository.unj.ac.id/id/eprint/70294

Actions (login required)

View Item