PERBANDINGAN METODE PROYEKSI KENAIKAN MUKA AIR LAUT BERBASIS AUTOREGRESSIVE INTEGRATED MOVING AVERAGE & LONG SHORT-TERM MEMORY. STUDI KASUS (PESISIR UTARA BANTEN)

Muhammad Fikry Akbar, - (2023) PERBANDINGAN METODE PROYEKSI KENAIKAN MUKA AIR LAUT BERBASIS AUTOREGRESSIVE INTEGRATED MOVING AVERAGE & LONG SHORT-TERM MEMORY. STUDI KASUS (PESISIR UTARA BANTEN). S1 thesis, Universitas Pendidikan Indonesia.

	Text S_SIK_1904623_Title.pdf Download (1MB)
	Text S_SIK_1904623_Chapter1.pdf Download (175kB)
	Text S_SIK_1904623_Chapter2.pdf Restricted to Staf Perpustakaan Download (562kB)
	Text S_SIK_1904623_Chapter3.pdf Download (789kB)
	Text S_SIK_1904623_Chapter4.pdf Restricted to Staf Perpustakaan Download (1MB)
	Text S_SIK_1904623_Chapter5.pdf Download (117kB)
	Text S_SIK_1904623_Appendix.pdf Restricted to Staf Perpustakaan Download (205kB)

Abstract

ABSTRACT The rise in sea level is one of the significant impacts of climate change that can have broad effects on coastal ecosystems and human populations. In this context, this study aims to predict sea level rise on the North Coast of Banten using Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) methods to make predictions based on historical sea level rise data from 2015 to 2018. The analytical approach begins with descriptive statistical analysis to understand data characteristics and Augmented Dickey-Fuller tests to examine data stationarity. The analysis results reveal that the data exhibit high variability and specific trends. Therefore, the use of ARIMA and LSTM methods, which are sensitive to time patterns, is considered relevant. The research findings unveil that the LSTM model outperforms the ARIMA model significantly in predicting sea level rise. Evaluation is conducted using various prediction error metrics such as Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Symmetric Mean Absolute Percentage Error (SMAPE), and Mean Squared Logarithmic Error (MSLE). The LSTM model consistently yields lower error values, indicating more accurate predictions compared to the ARIMA model. Through the implementation of the LSTM model, predictions for sea level rise on the North Coast of Banten can be obtained for the upcoming seven days. These prediction results are presented in both normalized and denormalized formats. The model proves its capability to generate predictions that closely approximate actual values with a high level of accuracy. The outcomes of this study hold significant implications for modeling sea level rise phenomena and their climate change impacts. The more complex and adaptive nature of the LSTM model demonstrates its effectiveness in tackling prediction challenges within intricate time series data, such as sea level rise. Nonetheless, it's important to interpret these results within the context and limitations of the utilized model. Keywords: ARIMA, Sea Level Rise, Long Short-Term Memory, Banten North Coast, Forecasting

Item Type:	Thesis (S1)
Additional Information:	https://scholar.google.com/citations?view_op=new_articles&hl=en&imq=Muhammad+Fikry+Akbar# ID SINTA Dosen Pembimbing Ayang Armelita Rosalia : 6721849 Willdan Aprizal Arifin : 6745746
Uncontrolled Keywords:	ARIMA, Sea Level Rise, Long Short-Term Memory, Banten North Coast, Forecasting
Subjects:	L Education > L Education (General)
Divisions:	UPI Kampus Serang > S1 Sistem Informasi Kelautan
Depositing User:	- Muhammad Fikry Akbar
Date Deposited:	02 Oct 2023 02:45
Last Modified:	02 Oct 2023 02:45
URI:	http://repository.upi.edu/id/eprint/108781

Actions (login required)

View Item