Muhammad Ilham Ramadhan, . (2025) PERBANDINGAN LONG SHORT TERM MEMORY DENGAN GATED RECURRENT UNITS UNTUK MEMPREDIKSI INDEKS HARGA SAHAM GABUNGAN PADA SEKTOR PERTAMBANGAN. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (176kB) |
|
Text
AWAL.pdf Download (1MB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (373kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (867kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (438kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 5.pdf Download (288kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (251kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (111kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (8MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (11MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (1MB) |
Abstract
Stock investment decision making can be estimated based on the inflation of the JCI (composite stock price index). The movement of JCI is often difficult to predict due to many related factors. A commonly used solution to overcome this problem is prediction using deep learning algorithms. Therefore, researchers want to find out which is superior to the LSTM (Long Short Term Memory) and GRU (Gated Recurrent Units) algorithms in predicting the value of the JCI in the mining sector whose stock performance has been very positive lately. The stages of this research consist of problem identification, literature study, data collection, data preprocessing, data mining, model visualization, and conclusion. The dataset is taken from yahoo finance site coded ADRO.JK and DSSA.JK. The model is assessed based on three evaluation metrics, namely Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and R squared (R^2). The results showed that: In ADRO.JK the lowest RMSE of LSTM test is 0.022, while the GRU test is 0.015; the lowest MAPE of LSTM test is 16.9%, while the GRU test is 17.1%; the highest R^2 of LSTM test is 0.9739, while the GRU test is 0.95546; In DSSA.The lowest JK RMSE of the LSTM test was 0.042, while the GRU test was 0.014; the lowest MAPE of the LSTM test was 109.5%, while the GRU test was 113.6%; and the highest R^2 of the LSTM test was 0.9784, while the GRU test was 0.99260. GRU was found to provide lower RMSE and higher R^2 values, while LSTM produced lower MAPE values. It can be concluded that the GRU algorithm has a better performance in predicting the JCI value in the mining industry.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2110511078] [Pembimbing 1: Iin Ernawati] [Pembimbing 2: Radinal Setyadinsa] [Penguji 1: Musthofa Galih Pradana] [Penguji 2: Muhammad Panji Muslim[ |
| Uncontrolled Keywords: | JCI, Mining Industry, Long Short Term Memory, Gated Recurrent Units |
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | MUHAMMAD ILHAM RAMADHAN |
| Date Deposited: | 12 Aug 2025 03:38 |
| Last Modified: | 12 Aug 2025 03:38 |
| URI: | http://repository.upnvj.ac.id/id/eprint/37344 |
Actions (login required)
 |
View Item |