relation: https://eprints.untirta.ac.id/51559/
title: PREDIKSI HARGA SAHAM MENGGUNAKAN TEKNIK DEEP LEARNING DENGAN JARINGAN SARAF TIRUAN LONG SHORT-TERM MEMORY
creator: MULYA, SITI AMALIA
subject: T Technology (General)
description: Stocks play a crucial role in the global economy. Their inherently fluctuating nature is influenced by various factors, such as macroeconomic conditions, commodity prices, company performance, and public sentiment. Despite the availability of numerous analytical indicators to assist in analyzing stock movements, predicting stock price trends remains increasingly challenging. The inevitable advancement of technology, particularly in the field of Artificial Intelligence, offers new opportunities to improve the accuracy of stock price predictions. Therefore, this study develops a Deep Learning model using the Long Short-Term Memory (LSTM) algorithm to enhance the accuracy of daily stock price predictions.  In this study, the stock price prediction model is built by implementing the Long Short-Term Memory (LSTM) algorithm integrated into a Model Inference Pipeline using the REST API method, enabling real-time access in a production environment. The model is trained and validated using historical stock price data of PT. Bank Central Asia (BCA), achieving a Mean Squarred Error (MSE) 1033,00, Root Mean Squarred Error (RMSE) 32,14 and Mean Absolute Percentage Error (MAPE) 0,24%. As a form of further evaluation, the LSTM model was also compared with the Extreme Gradient Boosting (XGBoost) model. The comparison results show that the LSTM model outperforms XGBoost in terms of prediction accuracy. However, XGBoost has the advantage of faster inference time.  Based on the results, it can be concluded that the developed Long Short-Term Memory (LSTM) model is capable of providing accurate stock price predictions with minimal prediction errors.
date: 2025-07-14
type: Thesis
type: NonPeerReviewed
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/1/Siti%20Amalia%20Mulya_3337210039_Fulltext.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/21/Siti%20Amalia%20Mulya_3337210039_01.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/11/Siti%20Amalia%20Mulya_3337210039_02.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/4/Siti%20Amalia%20Mulya_3337210039_03.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/5/Siti%20Amalia%20Mulya_3337210039_04.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/6/Siti%20Amalia%20Mulya_3337210039_05.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/7/Siti%20Amalia%20Mulya_3337210039_Ref.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/8/Siti%20Amalia%20Mulya_3337210039_Lamp.pdf
format: text
language: id
identifier: https://eprints.untirta.ac.id/51559/9/Siti%20Amalia%20Mulya_3337210039_CP.pdf
identifier:   MULYA, SITI AMALIA  (2025) PREDIKSI HARGA SAHAM MENGGUNAKAN TEKNIK DEEP LEARNING DENGAN JARINGAN SARAF TIRUAN LONG SHORT-TERM MEMORY.  S1 thesis, Fakultas Teknik Universitas Sultan Ageng Tirtayasa.