Prediksi Harga Cryptocurrency Menggunakan Metode Bi-LSTM

Prakosa, Muhammad Abi (2024) Prediksi Harga Cryptocurrency Menggunakan Metode Bi-LSTM. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) skripsi selesai 2-1-25.pdf Download (874kB)
	Text (BAB 1) skripsi selesai 2-27-30.pdf Download (84kB)
	Text (BAB 2) skripsi selesai 2-31-52.pdf Restricted to Repository staff only until 20 September 2026. Download (495kB)
	Text (BAB 3) skripsi selesai 2-53-66.pdf Restricted to Repository staff only until 20 September 2026. Download (215kB)
	Text (BAB 4) skripsi selesai 2-67-106.pdf Restricted to Repository staff only until 20 September 2026. Download (1MB)
	Text (BAB 5) skripsi selesai 2-107-108.pdf Download (10kB)
	Text (Daftar Pustaka) skripsi selesai 2-109-114.pdf Download (153kB)

Abstract

Today's rapid technological advances have made it easier for various human activities, including in the financial sector which is now inseparable from technology. Every investment always involves risk, including in cryptocurrencies like Quant (QNT). Unlike conventional currencies that are supervised and controlled by certain institutions, QNT is decentralized, so its price movements are not under the supervision or control of any party. As a result, the QNT exchange rate has become unstable and inconsistent. With prediction methods such as Bi-LSTM that can manage complex time-series data, these methods can be used to predict crypto market volatility. This study aims to assist QNT investors in determining relevant methods to predict cryptocurrency prices as well as considering the results of this research such as comparing the RMSE yield between the Bi-LSTM and LSTM methods as well as the results of the prediction graph. In the test results, a satisfactory score was obtained between the two methods with epoch parameters of 100, batch size 64, data split 90:10, learning rate 0.01 and hidden layer as many as 2 (16, 32). Bi-LSTM is superior by obtaining an RMSE of 0.01039086 and an LSTM of 0.01317647, Bi-LSTM has a smaller RMSE value while this larger LSTM indicates that the prediction of using Bi-LSTM is better than that of LSTM.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Puspita, Sari Anggraini	NIDN0716088605	anggraini.puspita.if@upnjatim.ac.id
Thesis advisor	Sihananto, Andreas Nugroho	NIDN0012049005	andreas.nugroho.jarkom@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Muhammad Abi Prakosa

Date Deposited:

20 Sep 2024 03:54

Last Modified:

20 Sep 2024 03:54

URI:

https://repository.upnjatim.ac.id/id/eprint/29628