BOYO, ASSTRID RIANGGI (2024) KOMPARASI ALGORITMA LSTM DAN ARIMA DALAM MEMPREDIKSI TINGGI MUKA AIR DI JAKARTA. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 COVER.pdf Download (526kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
02 BAB 1.pdf Restricted to Registered users only Download (157kB) |
|
|
Text (BAB II)
03 BAB 2.pdf Restricted to Registered users only Download (308kB) |
|
|
Text (BAB III)
04 BAB 3.pdf Restricted to Registered users only Download (225kB) |
|
|
Text (BAB IV)
05 BAB 4.pdf Restricted to Registered users only Download (762kB) |
|
|
Text (BAB V)
06 BAB 5.pdf Restricted to Registered users only Download (147kB) |
|
|
Text (DAFTAR PUSTAKA)
07 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (195kB) |
|
|
Text (LAMPIRAN)
08 LAMPIRAN.pdf Restricted to Registered users only Download (638kB) |
Abstract
This study aims to compare the performance of Long Short-Term Memory (LSTM) and Autoregressive Integrated Moving Average (ARIMA) algorithms in predicting water levels (TMA) in Jakarta. The variables examined are water levels from various locations in Jakarta, with data spanning from January to June 2024. The sample includes a total of 55,920 observations divided into various proportions (10:90 to 90:10). Data analysis methods involve normalizing data with MinMaxScaler for LSTM and testing stationarity and parameter search for ARIMA based on the Akaike Information Criterion (AIC). Evaluation results show that ARIMA generally performs better in terms of MSE, RMSE, and MAE, especially for short-term predictions with simple data. For instance, with a 90% training and 10% testing data split, ARIMA records an MSE of 354.89 and an RMSE of 18.84, while LSTM records an MSE of 354.45 and an RMSE of 18.83. Although LSTM shows a lower mean absolute error, it is more effective for data with high fluctuations and long-term predictions. In conclusion, ARIMA is more effective for simple time series data and short-term predictions, whereas LSTM is better suited for complex data with high variability. The choice of the best model depends on the prediction goals and data characteristics. Keywords: Long Short-Term Memory, Autoregressive Integrated Moving Average, water levels (TMA). Penelitian ini bertujuan untuk membandingkan kinerja algoritma Long Short-Term Memory (LSTM) dan Autoregressive Integrated Moving Average (ARIMA) dalam memprediksi tinggi muka air (TMA) di Jakarta. Variabel yang diteliti adalah tinggi muka air dari berbagai lokasi di Jakarta dengan periode data dari Januari hingga Juni 2024. Jumlah sampel mencakup total 55.920 pengamatan yang dibagi dalam berbagai proporsi (10:90 hingga 90:10). Metode analisis data melibatkan normalisasi data dengan MinMaxScaler untuk LSTM dan uji stasioneritas serta pencarian parameter ARIMA berdasarkan kriteria Akaike Information Criterion (AIC). Hasil evaluasi menunjukkan bahwa ARIMA umumnya lebih baik dalam hal MSE, RMSE, dan MAE, terutama untuk prediksi jangka pendek dengan data sederhana. Misalnya, pada pembagian 90% data latih dan 10% data uji, ARIMA mencatat MSE sebesar 354.89 dan RMSE sebesar 18.84, sementara LSTM mencatat MSE sebesar 354.45 dan RMSE sebesar 18.83. Sementara itu, meskipun LSTM menunjukkan kesalahan absolut rata-rata yang lebih rendah, ia lebih efektif untuk data dengan fluktuasi tinggi dan prediksi jangka panjang. Kesimpulannya, ARIMA lebih efektif untuk data time series sederhana dan prediksi jangka pendek, sementara LSTM lebih cocok untuk data kompleks dengan variabilitas tinggi. Pemilihan model terbaik tergantung pada tujuan prediksi dan karakteristik data. Kata Kunci: Long Short-Term Memory, Autoregressive Integrated Moving Average, tinggi muka air.
| Item Type: | Thesis (S1) |
|---|---|
| Call Number CD: | FIK/INFO. 24 213 |
| NIM/NIDN Creators: | 41520010127 |
| Uncontrolled Keywords: | Long Short-Term Memory, Autoregressive Integrated Moving Average, tinggi muka air. |
| Subjects: | 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika 500 Natural Science and Mathematics/Ilmu-ilmu Alam dan Matematika > 510 Mathematics/Matematika > 518 Numerical Analysis/Analisis Numerik, Analisa Numerik > 518.1 Algorithms/Algoritma |
| Divisions: | Fakultas Ilmu Komputer > Informatika |
| Depositing User: | Dede Muksin Lubis |
| Date Deposited: | 23 Sep 2024 04:25 |
| Last Modified: | 23 Sep 2024 04:25 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/91692 |
Actions (login required)
 |
View Item |
