Çok Küçük Veri Seti Üzerinde Makine Öğrenmesi ile Bitcoin Fiyat Yönü Tahminlemesi

Yıl 2025, ERKEN GÖRÜNÜM, 1 - 1

Kağan Öktem , Adem Tekerek

https://doi.org/10.2339/politeknik.1667403

Cited By: 1

Öz

Günümüzde yatırım danışmanlığı, profesyonel olarak sunulan yaygın bir hizmettir. Bu hizmeti sağlamak için danışmanlık firmaları kurulur ve finans uzmanları istihdam edilir. Hizmetten faydalanmak isteyen bireyler aylık ücret ödeyerek danışmanlık alır. Finansal piyasalar uzmanlık gerektirir. Ancak yapay zekâ sistemlerinin gelişmesiyle, bu alanda da önemli dönüşümler yaşanmıştır. Özellikle LSTM ve GRU gibi derin öğrenme algoritmaları, doğrusal olmayan zaman serisi verilerle kısa, orta ve uzun vadeli fiyat tahminlerinde kullanılmaktadır. Ancak bu yöntemler büyük veri setleri gerektirir ve aşırı öğrenmeye (overfitting) yatkındır. Derin öğrenme ile pekiştirmeli öğrenmenin birlikte kullanımı başarılı sonuçlar verse de, bu modellerin entegrasyonu yoğun araştırma ve hesaplama gücü gerektirir. Bu çalışmada, Random Forest Regressor kullanılarak Bitcoin’in (BTC) günlük fiyat yönünü tahmin eden BTC-FYTR (Bitcoin Fiyat Yönü Tahminleme Robotu) modeli tanıtılmaktadır. Ensemble tabanlı bu makine öğrenimi modeli, büyük veri ihtiyacı duymadan, fiyatı etkileyen teknik indikatörleri başarıyla belirleyerek yüksek doğruluk sunmaktadır. Mart 2018’den günümüze kadar olan verilerle yapılan testlerde model, %99.20 başarı oranına ulaşmıştır. Google Colab v5e1 konfigürasyonunda yalnızca 22 saniyede sonuç üretmektedir. Ayrıca çalışmada 2017-2024 arası literatür incelenmiş, eksiklikler belirlenmiş ve bu çalışmanın alana katkısı ortaya konmuştur.

Anahtar Kelimeler

BTC , fiyat yönü tahminleme , random forest regressor , teknik indikatörler , makine öğrenimi , çok küçük veri seti

Bitcoin Price Direction Prediction Using Machine Learning on a Very Small Dataset

Öz

Investment advisory services are now commonly offered by consulting firms with financial experts, typically for a monthly fee. Financial markets require specialized knowledge, but advancements in artificial intelligence have revolutionized this field. Deep learning algorithms, especially Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), are widely used to predict asset price trends in nonlinear time-series data. However, they demand large datasets and are prone to overfitting. Recently, combining deep learning with reinforcement learning has shown promise, though it requires intensive research and computational resources. This study introduces the BTC-PDPR (Bitcoin Price Direction Prediction Robot) model, which predicts Bitcoin's daily price direction using the Random Forest Regressor. As an ensemble-based machine learning model, it works effectively with smaller datasets and identifies key technical indicators influencing price trends. The model achieved a 99.20% accuracy rate on data from March 2018 to the present. It runs efficiently in Google Colab (v5e1 configuration), producing results in just 22 seconds. This paper outlines the methodology, reviews relevant studies from 2017 to 2024, highlights gaps in the literature, and emphasizes the study’s contributions to the field.

Anahtar Kelimeler

BTC , price direction prediction , random forest regressor , technical indicators , machine learning , very small dataset

Etik Beyan

The author(s) of this article declare that the materials and methods used in this study do not require ethical committee permission and/or legal-special permission.

Destekleyen Kurum

Gazi University, Faculty of Technology, Computer Engineering Department

Kaynakça

• [1] PhamToan D., VoThiHang N., PhamThi B., “Improving forecasting model for fuzzy time series using the self-updating clustering and bi-directional long short term memory algorithm”, Expert Systems With Applications, 241: 122767, (2024).
• [2] Mardjo A., Choksuchat C., "HyBiLSTM: Multivariate Bitcoin Price Forecasting Using Hybrid Time-Series Models With Bidirectional LSTM", IEEE Access, 12: 50792-50808, (2024).
• [3] Serafini G., Yi P., Zhang Q., Brambilla M., Wang J., Hu Y., Li B., "Sentiment-Driven Price Prediction of the Bitcoin based on Statistical and Deep Learning Approaches", Proceedings of the International Joint Conference on Neural Networks (IJCNN), Glasgow, UK, 1-8, (2020).
• [4] Taslim D. G., Murwantara I. M., "A Comparative Study of ARIMA and LSTM in Forecasting Time Series Data", Proceedings of the 2022 International Conference on Information Technology, Computer, and Electrical Engineering (ICITACEE), Semarang, Indonesia, 231-235 (2022).
• [5] Makridakis S., Spiliotis E., Assimakopoulos V., "Statistical and Machine Learning Forecasting Methods: Concerns and Ways Forward", PLoS ONE, 13: e0194889, (2018).
• [6] Kumari S. E., Ramesh B., "Prediction of Stock Prices using Machine Learning Regression and Classification Algorithms", Proceedings of the 2020 IEEE International Conference for Emerging Technology (INCET), Belgaum, India, 1-5, (2020).
• [7] Ramya N., Sanjay Roshan R., Vishal Srinivas R., Krishna Prasad D., "Crypto-Currency Price Prediction using Machine Learning", Proceedings of the Sixth International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India, 1455-1458, (2022).
• [8] Bouktif S., Fiaz A., Awad M., "Augmented Textual Features-Based Stock Market Prediction", IEEE Access, 8: 40269-40282, (2020).
• [9] Ahmed F. U., Ahmed M., F. Mahi H., Abdullah S. H., Suha S. A., "A Comparative Performance Evaluation of Bitcoin Price Prediction Using Machine Learning Techniques", International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD), Dhaka, Bangladesh, 194-198, (2023).
• [10] Basha M. S. A., Oveis P. M., Prabavathi C., Lakshmi M. B., Sucharitha M. M., "An Efficient Machine Learning Approach: Analysis of Supervised Machine Learning Methods to Forecast the Diamond Price", 2023 International Conference for Advancement in Technology (ICONAT), Goa, India, 1-6, (2023).
• [11] Gudelek M. U., Boluk S. A., Ozbayoglu A. M., "A Deep Learning based Stock Trading Model with 2-D CNN Trend Detection", Proceedings of the 2017 IEEE Symposium Series on Computational Intelligence (SSCI), Honolulu, HI, USA, 1-8, (2017).
• [12] Behura J. P., Pande S. D., Ramesh J. V. N., "Stock Price Prediction using Multi-Layered Sequential LSTM", EAI Endorsed Transactions on Scalable Information Systems, 1: 1-8, (2023).
• [13] Akter M. S., Shahriar H., Rahman M. A., Rahman M., Cuzzocrea A., "Early Prediction of Cryptocurrency Price Decline: A Deep Learning Approach", Proc. 26th International Conference on Computer and Information Technology (ICCIT), Cox’s Bazar, Bangladesh, 1-6, (2023).
• [14] Altuner A.B., Kilimci Z.H., "A Novel Deep Reinforcement Learning Based Stock Price Prediction Using Knowledge Graph and Community Aware Sentiments", Turkish Journal of Electrical Engineering and Computer Sciences, 30: No. 4, (2023).
• [15] Zouaghia Z., Kodia Aouina Z., Ben Said L., "Hybrid Machine Learning Model for Predicting NASDAQ Composite Index", Proceedings of the 2023 International Symposium on Networks, Computers and Communications (ISNCC), Doha Qatar, 1-6, (2023).
• [16] Lin W., Xie L., Xu H., "Deep-Reinforcement-Learning-Based Dynamic Ensemble Model for Stock Prediction", Electronics, 12: 4483 (2023).
• [17] Haarnoja T., Zhou A., Abbeel P., Levine S., "Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor", 35th International Conference on Machine Learning (ICML 2018), Stockholm, Sweden, https://arxiv.org/abs/1801.01290, (2018).
• [18] Kabbani T., Duman E., "Deep Reinforcement Learning Approach for Trading Automation in the Stock Market", IEEE Access, 10: 93564-93574, (2022).
• [19] Liu R., Guo W., Jin K., Ding Y., Chen Z., Ma S., Li Z., “Computer Intelligent Investment Strategy Based on Deep Reinforcement Learning and Multi-Layer LSTM Network”, Proceedings of the 2022 IEEE 2nd International Conference on Data Science and Computer Application (ICDSCA), Dalian, China, 1006-1015, (2022).
• [20] https://www.cryptodatadownload.com/data/premium/, “BTC - Bitcoin_technical_analysis (2018 - 2025)”, (2025).
• [21] https://taapi.io/indicators/balance-of-power, “Technical Analysis Indicators API (Updated Daily BOP Indicator Values)”, (2025).
• [22] https://finance.yahoo.com/quote/BTC-USD, “BTC Daily Closing Price”, (2025).
• [23] Breiman L., "Random Forests", Machine Learning, 45: 5-32, (2001).
• [24] Livshin I., "Balance of Power", Technical Analysis of Stocks & Commodities, 19: 26-30, (2001).