Prediksi Harga Bitcoin Menggunakan Model Hibrida Transformer Dan Long Short Term Memory
DOI:
https://doi.org/10.52436/1.jpti.1091Kata Kunci:
Deep Learning, LSTM, Model Hibrida, Peramalan Harga Bitcoin, TransformerAbstrak
Peramalan harga Bitcoin merupakan tantangan signifikan akibat fluktuasi pasar yang tinggi, di mana model deret waktu konvensional seringkali gagal menangkap dependensi data jangka pendek dan panjang secara bersamaan. Untuk mengatasi masalah tersebut, penelitian ini bertujuan mengembangkan dan mengevaluasi model deep learning hibrida guna meningkatkan presisi peramalan harga Bitcoin. Model yang diusulkan mengintegrasikan arsitektur Long Short-Term Memory (LSTM), yang andal dalam memproses data sekuensial, dengan mekanisme atensi pada Transformer yang mampu mengidentifikasi hubungan data yang relevan. Metode penelitian mencakup penggunaan data harga historis Bitcoin yang telah melalui tahap pra-pemrosesan dan normalisasi. Arsitektur model terdiri dari dua lapisan LSTM (30 unit), diikuti oleh lapisan Transformer dengan Multi-Head Attention (satu head, dimensi kunci empat), dan diakhiri dengan lapisan Global Average Pooling. Pelatihan model dilakukan menggunakan konfigurasi optimal selama 10 epoch dengan ukuran batch 64. Hasil evaluasi menunjukkan kinerja prediksi yang sangat akurat, dibuktikan dengan nilai Mean Absolute Percentage Error (MAPE) sebesar 0,000695 dan R-squared (R²) 0,99999. Temuan ini menegaskan bahwa pendekatan hibrida efektif dalam menangkap pola kompleks pada data harga Bitcoin, sehingga menawarkan alat yang lebih kuat untuk analisis keuangan di pasar cryptocurrency.
Unduhan
Referensi
M. Conti, K. E. Sandeep, C. Lal, and S. Ruj, “A survey on security and privacy issues of bitcoin,” IEEE Communications Surveys and Tutorials, vol. 20, no. 4, 2018, doi: 10.1109/COMST.2018.2842460.
L. Ø. Bergsli, A. F. Lind, P. Molnár, and M. Polasik, “Forecasting volatility of Bitcoin,” Res Int Bus Finance, vol. 59, 2022, doi: 10.1016/j.ribaf.2021.101540.
X. Song, “Predicting cryptocurrency investment suitability using machine learning techniques,” Applied and Computational Engineering, vol. 29, no. 1, 2023, doi: 10.54254/2755-2721/29/20230726.
L. Afuan and N. Hidayat, “Sentiment Analysis of the Kampus Merdeka Program on Twitter Using Support Vector Machine and a Feature Extraction Comparison: TF-IDF vs. FastText,” Journal of Applied Data Sciences, vol. 5, no. 4, pp. 1738–1753, Dec. 2024, doi: 10.47738/jads.v5i4.436.
X. Wu and S. Ding, “The impact of the Bitcoin price on carbon neutrality: Evidence from futures markets,” Financ Res Lett, vol. 56, 2023, doi: 10.1016/j.frl.2023.104128.
M. D. Angelo, I. Fadhiilrahman, and Y. Purnama, “Comparative Analysis of ARIMA and Prophet Algorithms in Bitcoin Price Forecasting,” in Procedia Computer Science, Elsevier B.V., 2023, pp. 490–499. doi: 10.1016/j.procs.2023.10.550.
Ardiansyah Risko Anwari and S. Sukirman, “RECOMMENDATION SYSTEM TO SELECT A MAJOR OF VOCATIONAL SCHOOL USING DECISION TREE,” Jurnal Teknik Informatika (Jutif), vol. 5, no. 2, pp. 589–598, Apr. 2024, doi: 10.52436/1.jutif.2024.5.2.1327.
S. Kashyap and M. Singh, “A Deep Learning Approach for Crypto Price Prediction,” International Research Journal of Engineering and Technology, vol. 3, no. 3, 2022.
A. S. Saud and S. Shakya, “Analysis of look back period for stock price prediction with RNN variants: A case study on banking sector of NEPSE,” in Procedia Computer Science, Elsevier B.V., 2020, pp. 788–798. doi: 10.1016/j.procs.2020.03.419.
S. Samsul, I. Tahyudin, and F. Setyo utomo, “Perbandingan Model CNN, LSTM, dan FNN dalam Klasifikasi Kulit Penderita Diabetes,” Jurnal Pendidikan dan Teknologi Indonesia, vol. 5, no. 3, pp. 571–583, Mar. 2025, doi: 10.52436/1.jpti.536.
N. Afrianto, D. H. Fudholi, and S. Rani, “Prediksi Harga Saham Menggunakan BiLSTM dengan Faktor Sentimen Publik,” Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 6, no. 1, pp. 41–46, Feb. 2022, doi: 10.29207/resti.v6i1.3676.
S. E. Parameswaran, V. Ramachandran, and S. Shukla, “Crypto Trend Prediction Based on Wavelet Transform and Deep Learning Algorithm,” in Procedia Computer Science, Elsevier B.V., 2024, pp. 1179–1189. doi: 10.1016/j.procs.2024.04.112.
D. Rothman, Transformers for Natural Language Processing Build innovative deep neural network architectures for NLP & Python, PyTorch, TensorFlow, BERT, RoBERTa & more. 2021.
D. Chicco, M. J. Warrens, and G. Jurman, “The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation,” PeerJ Comput Sci, vol. 7, pp. 1–24, 2021, doi: 10.7717/PEERJ-CS.623.
S. Mounika, “Crypto-Currency Price Prediction using CNN and LSTM Models,” Int J Res Appl Sci Eng Technol, vol. 9, no. 3, 2021, doi: 10.22214/ijraset.2021.33191.
S. Singh, A. Pise, and B. Yoon, “Prediction of bitcoin stock price using feature subset optimization,” Heliyon, vol. 10, no. 7, Apr. 2024, doi: 10.1016/j.heliyon.2024.e28415.
R. Sari, K. Kusrini, T. Hidayat, and T. Orphanoudakis, “Improved LSTM Method of Predicting Cryptocurrency Price Using Short-Term Data,” IJCCS (Indonesian Journal of Computing and Cybernetics Systems), vol. 17, no. 1, p. 33, Feb. 2023, doi: 10.22146/ijccs.80776.
D. Rothman, Transformers for natural language processing?: build, train, and fine-tuning deep neural network architectures for NLP with Python, PyTorch, TensorFlow, BERT, and GPT-3. 2022.
B. Zhong, “Deep learning integration optimization of electric energy load forecasting and market price based on the ANN–LSTM–transformer method,” Front Energy Res, vol. 11, 2023, doi: 10.3389/fenrg.2023.1292204.
E. M. Al-Ali et al., “Solar Energy Production Forecasting Based on a Hybrid CNN-LSTM-Transformer Model,” Mathematics, vol. 11, no. 3, 2023, doi: 10.3390/math11030676.
L. Dong et al., “Prediction of streamflow based on dynamic sliding window lstm,” Water (Switzerland), vol. 12, no. 11, 2020, doi: 10.3390/w12113032.
