Makine Öğrenme Yöntemi ile Bitcoin Piyasasının Geleceğinin Araştırılması: Türkiye Örneğinde Uygulama

Year 2022, , 171 - 180, 05.10.2022

Merve Arslan , Özerk Yavuz , Serdar Kuzu , İsmail Erkan Çelik

https://doi.org/10.17753/sosekev.1140004

Abstract

Tarihsel süreç içerisinde değişen ihtiyaçlar ve teknolojik gelişmeler doğrultusunda paranın farklı şekiller aldığı görülmektedir. Son zamanlarda kripto para birimleri hayatımıza dahil olmuştur ve Bitcoin bu para birimlerinden birisidir. Bitcoin, kriptografik teknikler kullanarak ve merkezi bir otoritenin kontrolüne ihtiyaç duymadan çalışan dijital bir para birimidir. Teknolojik gelişmeler sonucunda yeni bir parasal araç olarak hayatımıza giren ve para birimlerine alternatif olacağı tahmin edilen Bitcoin'e ilginin arttığı görülmektedir. Bu makalede yöntem olarak yapay zekânın bir dalı olan makine öğrenmesi yöntemi kullanılmıştır. Türkiye örneğinde bitcoinin 2016 yılı günlük kapanış verileri kullanılmıştır. Bu çalışma ile Makine öğrenmesi yöntemi kullanılarak, Bitcoin piyasasının kapanış fiyatlarının tahmin edilmesi amaçlanmaktadır. Makine öğrenmesi yöntemi Bitcoin piyasasının nasıl şekilleneceğine dair bazı kurallar üretilmiş ve analiz kısmında bu sonuçlara yer verilmiştir.

Keywords

Para, Kripto Para, Bitcoin, Yapay Zekâ, Makine Öğrenmesi

RESEARCHING THE FUTURE OF BITCOIN MARKET WITH MACHINE LEARNING METHOD: ANAPPLICATION ON THE CASE OF TURKEY

Abstract

It is seen that money takes different forms in line with the changing needs and technological developments throughout the historical process. Recently, cryptocurrencies have been included in our lives with Bitcoin. Bitcoin is a digital currency that functions using cryptographic techniques and without the need for the control of a central authority. As a result of technological developments, it is seen that the interest in Bitcoin, which has entered our lives as a new monetary tool and is predicted to be an alternative to currencies, is increasing. In this article, the machine learning method, which is a branch of artificial intelligence, is used as a method. In the example of Turkey, the daily closing data of bitcoin for 2016 were used. The machine learning method is aimed to predict the closing prices of the bitcoin market.According to the analysis findings, it is seen that the closing prices realized in the 4th quarter are higher than the closing prices realized in the 1st quarter. If the volume USD is higher than 5517.34 then it is Q1. Some rules have been produced with the Machine Learning method.It is aimed to contribute to the literature by using themachine learning method for predicting Bitcoin closing prices.

Keywords

Money, Cryptocurrency, Bitcoin, Artificial Intelligence

References

• Afrin, F. & Nahar, I. (2015). Incremental learning-based intelligent job search system, [Doctoral Dissertation, Brac University].
• Ayodele, T. O. (2010). Machine learning overview. Intech, Open Access Publisher. https://www.intechopen.com/chapters/10683
• Brito, J. & Castillo, A. (2013). Bitcoin: A primer for policymakers. Mercatus Center at George Mason University. ttps://www.mercatus.org/system/files/Brito_BitcoinPrimer.pdf
• Böhme, R., Christin, N., Edelman, B., & Moore, T. (2015). Bitcoin: economics, technology, and governance. Journal of Economic Perspectives,29(2),213-238. http://dx.doi.org/10.1257/jep.29.2.213
• Ceylan, M. E. (2019). Bitcoin ekonomisi: Kripto para bitcoin'in finans sektörü içindeki yeri [Doctoral Dissertation, Batman University]. Yükseköğretim Kurulu Ulusal Tez Merkezi.
• Chao, W. L. (2011). Machine learning tutorial. DISP Lab, Graduate Institute of Communication Engineering, National Taiwan University.http://disp.ee.ntu.edu.tw/~pujols/Machine%20Learning%20Tutorial.pdf
• Sakız, B. & Gencer, A. H. (2018). Forecasting the bitcoin price via artificial neural networks. Internatıonal Conference on Eurasıan Economies. https://doi.org/10.36880/C10.02070
• Gentleman, R., Huber, W. & Carey, V. J., (2008). Supervised machine learning in bioconductor case studies. Springer.
• Franco, P. (2015). Understanding bitcoin: Cryptography, engineering, and economics. Wiley Finance Series.
• Ian H.Witten and Elbe Frank (2005). Datamining Practical machine learning tools and techniques. Morgan Kaufmann.
• Jang, H., Lee, J. (2017). An empirical study on modeling and prediction of bitcoin prices with bayesian neural networks based on blockchain information. IEEE Access, C. 6, 5427-5437. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8125674
• Kadiroğlu, Z., Akılotu, B. N. & Şengür, A. (2019). Mechanism of bitcoin and İnvestigation of the studies in the literature related to bitcoin. [Conference presentation abstract]. International Informatics and Software Engineering Conference, İEEE, Ankara. https://ieeexplore.ieee.org/xpl/conhome/8959514/proceeding
• D., Karahoca A., Karahoca A., & Yavuz Ö. (2013). An early warning system approach for theidentification of currency crises with data mining techniques. Neural Computing and Applications, 23(2471-2479). https://doi.org/10.1007/s00521-012-1206-9
• Kotsiantis, S. B., Zaharakis, I., & Pintelas, P. (2007). Supervised machine learning: a review of classification techniques. Springer.
• Nakamoto, S. (2019). Bitcoin: A peer-to-peer electronic cash system. https://bitcoin. org/bitcoin.pdf
• Pham, T. & Lee, S. (2017). Anomaly detection in bitcoin network using unsupervised learning methods. Balancing Method for High Dimensional Causal Inference. https://doi.org/10.48550/arXiv.1611.03941
• Schapire, R. E. (2003). The boosting approach to machine learning: an overview in nonlinear estimation and classification. Springer.
• Stern, S. (2017). Digital Currency: May be a bit player now but in the longer term a game changer for tax. Journal of Australian Taxation, 19 (1). https://www.jausttax.com.au/Articles_Free/JAT%20Volume%2019%20Issue%201%20-%20Stern.pdf