Millani, Alief Indy (2026) Implementasi Metode Support Vector Regression (SVR) pada Prediksi Harga Altcoin Dengan Firefly Optimization. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
Cover.pdf Download (1MB) |
|
Text (Bab 1)
Bab 1.pdf Download (247kB) |
|
Text (Bab 2)
Bab 2.pdf Restricted to Repository staff only until 8 January 2029. Download (488kB) | Request a copy |
|
Text (Bab 3)
Bab 3.pdf Restricted to Repository staff only until 8 January 2029. Download (877kB) | Request a copy |
|
Text (Bab 4)
Bab 4.pdf Restricted to Repository staff only until 8 January 2029. Download (4MB) | Request a copy |
|
Text (Bab 5)
Bab 5.pdf Download (239kB) |
|
Text (Daftar Pustaka)
Daftar Pustaka.pdf Download (179kB) |
|
Text (Lampiran)
Lampiran.pdf Restricted to Repository staff only until 8 January 2029. Download (663kB) | Request a copy |
Abstract
The digital development in Indonesia post-pandemic has driven an increase in investments, one of which is in digital assets like Cryptocurrency. Cryptocurrency, supported by cryptographic and blockchain technology, offers significant profit potential but also comes with High market volatility, increasing risks for investors. Therefore, accurate price prediction is crucial in optimizing profits and minimizing potential losses.This study focuses on predicting altcoin prices, which refer to all cryptocurrencies except Bitcoin, using a combination of Support Vector Regression (SVR) and Firefly Optimization. Firefly Optimization is employed to optimize SVR parameters such as C, Gamma, and Epsilon, thereby improving prediction accuracy. The altcoins analyzed in this research are Ethereum, Solana, and Litecoin. The results show that Firefly Optimization successfully enhances SVR’s prediction performance by reducing error values such as MAPE, MAE, MSE, and RMSE, while increasing R² values. For the Solana dataset, the MAPE value decreased from 7.62% to 3.42%, accompanied by an improvement in R² from 0.80 to 0.95. In the Litecoin dataset, MAPE decreased from 9.21% to 3.49%, while R² increased from 0.83 to 0.94. For Ethereum, the Firefly-SVR model produced similar improvements, reducing MAPE from 9.06% to 3.03% and increasing R² from 0.40 to 0.93. Although this method improves the accuracy of altcoin price predictions, the implementation of Firefly Optimization requires Higher computational time For Solana, the computation time increased from 11.07 seconds to 613.11 seconds. For Litecoin, it rose from 10.67 seconds to 501.70 seconds, and for Ethereum, from 9.69 seconds to 521.64 seconds. Thus, this study contributes to the development of a new analytical method that can be applied to other financial markets, considering the trade-off between prediction accuracy and computational efficiency.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T58.6-58.62 Management Information Systems |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Alief Indy Millani | ||||||||||||
| Date Deposited: | 09 Jan 2026 06:59 | ||||||||||||
| Last Modified: | 09 Jan 2026 06:59 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/48606 |
Actions (login required)
 |
View Item |