Evaluating the performance of bitcoin price forecasting using machine learning techniques on historical data

• Autorzy: Ahmed, Mamun , Suha, Sayma Alam , Mahi, Fahamida Hossain , Ahmed, Forhad Uddin

Warianty tytułu

PL

Ocena skuteczności prognozowania cen bitcoinów przy użyciu technik uczenia maszynowego na danych historycznych

• Języki publikacji: EN

Abstrakty

EN

Since entering the market in 2009, Bitcoin has had a price that is extremely erratic. Its price is influenced by factors such as adoption rates, regulatory changes, geopolitical occurrences, and macroeconomic developments. Experts believe that Bitcoin's price will rise in the long run due to limited supply and rising demand. Therefore, the aim of this study is to propose an ensemble feature selection and machine learning-based approach to predict bitcoin price. For this research purpose, the cryptocurrency-based dataset has been used, visualized, and preprocessed. Five different feature selection approaches (Pearson, RFE, Embedded Random Forest, Tree-based and Light GBM) are followed by ensemble methodology, with the maximum voting approach to extract the most significant features and generate a dataset with reduced attributes. Then the dataset with or without feature selection is used for bitcoin price prediction by applying ten different machine learning regressing models, which includes six traditional, four bagging and boosting ensemble techniques. The comparative result analysis through multiple performance parameters reveals that the decreased number of features improves the performance for each of the models and the ensemble models outperform other types of models. Therefore, Random Forest regression ensemble ML model can get the best prediction accuracy with 0.036018 RMSE, 0.029470 MAE and 0.934512 R2 employing the dataset with reduced features for estimating the value of bitcoin.

PL

Od momentu wejścia na rynek w 2009 roku, cena Bitcoina jest niezwykle nieregularna. Na jego cenę wpływają takie czynniki, jak wskaźniki popularności, zmiany regulacyjne, wydarzenia geopolityczne i zmiany makroekonomiczne. Eksperci uważają, że cena Bitcoina wzrośnie w dłuższej perspektywie ze względu na ograniczoną podaż i rosnący popyt. Dlatego też celem niniejszego badania jest zaproponowanie podejścia opartego na selekcji cech i uczeniu maszynowym do przewidywania ceny bitcoina. Do tego celu badawczego wykorzystano, zwizualizowano i wstępnie przetworzono zbiór danych oparty na kryptowalutach. Zastosowano pięć różnych podejść do wyboru cech (Pearson, RFE, Embedded Random Forest, Tree-based i Light GBM), a następnie metodologię ensemble, z podejściem maksymalnego głosowania w celu wyodrębnienia najważniejszych cech i wygenerowania zbioru danych ze zredukowanymi atrybutami. Następnie zbiór danych z lub bez selekcji cech jest wykorzystywany do przewidywania cen bitcoinów poprzez zastosowanie dziesięciu różnych modeli regresji uczenia maszynowego, w tym sześciu tradycyjnych, czterech technik baggingu i boostingu. Analiza porównawcza wyników za pomocą wielu parametrów wydajności pokazuje, że zmniejszona liczba cech poprawia wydajność każdego z modeli, a modele zespołowe przewyższają inne typy modeli. W związku z tym model Random Forest regression ensemble ML może uzyskać najlepszą dokładność przewidywania z 0,036018 RMSE, 0,029470 MAE i 0,934512 R2, wykorzystując zbiór danych ze zredukowanymi funkcjami do szacowania wartości bitcoinów.

Słowa kluczowe

PL

uczenie maszynowe; bitcoin; modele regresji; selekcja cech zespołu

EN

machine learning; bitcoin; regression models; ensemble feature selection

• Czasopismo: Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska
• Rocznik: 2024
• Tom: T. 14, nr 2
• Strony: 101--108
• Opis fizyczny: Bibliogr. 42 poz., tab., wykr.

Twórcy

autor

Ahmed, Mamun

• Bangladesh Army International University of Science & Technology, Computer Science & Engineering, Cumilla, Bangladesh,

autor

Suha, Sayma Alam

• Bangladesh University of Professionals, Department of Computer Science & Engineering, Dhaka, Bangladesh,

autor

Mahi, Fahamida Hossain

• Bangladesh Army International University of Science & Technology, Computer Science & Engineering, Cumilla, Bangladesh,

autor

Ahmed, Forhad Uddin

• Bangladesh Army International University of Science & Technology, Computer Science & Engineering, Cumilla, Bangladesh,

Bibliografia

• [1] Andi H. K.: An Accurate Bitcoin Price Prediction Using Logistic Regression with LSTM Machine Learning Model. Journal of Soft Computing Paradigm 3(3), 2021, 205–217 [https://doi.org/10.36548/jscp.2021.3.006].
• [2] Arumalla G. S. et al.: Bitcoin price fluctuation analysis and prediction using machine learning. International Journal of Progressive Research in Engineering Management and Science – IJPREMS 03(03), 2022, 421–425.
• [3] Auti A. et al.: Bitcoin Price Prediction Using Svm. International Journal of Engineering Applied Sciences and Technology 6(11), 2022, 226–229.
• [4] Bhatt S. et al.: Machine Learning based Cryptocurrency Price Prediction using Historical Data and Social Media Sentiment. Computer Science & Information Technology – CS & IT 13, 2023, 1–11 [https://doi.org/10.5121/csit.2023.131001].
• [5] Bhattad S. et al.: Review of Machine Learning Techniques for Cryptocurrency Price Prediction. EasyChair 10190, 2023.
• [6] Chen J.: Analysis of Bitcoin Price Prediction Using Machine Learning. Journal of Risk and Financial Management 16(1), 2023, 51 [https://doi.org/10.3390/jrfm16010051].
• [7] Chen W. et al.: Machine Learning Model for Bitcoin Exchange Rate Prediction Using Economic and Technology Determinants. International Journal of Forecasting 37(1), 2021, 28–43 [https://doi.org/10.1016/j.ijforecast.2020.02.008].
• [8] Chen Z. et al.: Bitcoin Price Prediction Using Machine Learning: An Approach to Sample Dimension Engineering. Journal of Computational and Applied Mathematics 365, 2020, 112395 [https://doi.org/10.1016/j.cam.2019.112395].
• [9] Chowdhury R. et al.: An Approach to Predict and Forecast the Price of Constituents and Index of Cryptocurrency Using Machine Learning. Physica. A 551, 2020, 124569 [https://doi.org/10.1016/j.physa.2020.124569].
• [10] Dimitriadou A., Gregoriou A.: Predicting Bitcoin Prices Using Machine Learning. Entropy 25(5), 2023, 777 [https://doi.org/10.3390/e25050777].
• [11] Erfanian S. et al.: Predicting Bitcoin (BTC) Price in the Context of Economic Theories: A Machine Learning Approach. Entropy 24(10), 2022, 1487 [https://doi.org/10.3390/e24101487].
• [12] Gadey R. S. et al.: Price prediction of bitcoin using machine learning. International Journal of Engineering Applied Science and Technology 5(1), 2020, 502–506 [https://doi.org/10.33564/ijeast.2020.v05i01.089].
• [13] Iqbal M. et al.: Time-Series Prediction of Cryptocurrency Market Using Machine Learning Techniques. EAI Endorsed Transactions on Creative Technologies 8(28), 2021, 170286 [https://doi.org/10.4108/eai.7-7-2021.170286].
• [14] Islam M. R. et al.: Data-Driven Heart Disease Prediction by Ensemble Feature Selection and Machine Learning Techniques. 25th International Conference on Computer and Information Technology (ICCIT), 2022, 575–580 [https://doi.org/10.1109/iccit57492.2022.10054998].
• [15] Jaquart P. et al.: Short-term Bitcoin Market Prediction via Machine Learning. Journal of Finance and Data Science 7, 2021, 45–66 [https://doi.org/10.1016/j.jfds.2021.03.001].
• [16] Kavitha H. et al.: Performance Evaluation of Machine Learning Algorithms for Bitcoin Price Prediction. 2020 Fourth International Conference on Inventive Systems and Control (ICISC), 2020, [https://doi.org/10.1109/icisc47916.2020.9171147.
• [17] Kervanci, I. S., Akay F.: Review on Bitcoin Price Prediction Using Machine Learning and Statistical Methods. Sakarya University Journal of Computer and Information Sciences 3(3), 2020, 272–282 [https://doi.org/10.35377/saucis.03.03.774276].
• [18] Khedr A. M. et al.: Cryptocurrency Price Prediction Using Traditional Statistical and Machine‐learning Techniques: A Survey. International Journal of Intelligent Systems in Accounting, Finance & Management 28(1), 2021, 3–34 [https://doi.org/10.1002/isaf.1488].
• [19] Kiranashree B. K. et al.: Price Prediction of Bitcoins. 22 Mar. 2023, [https://journal.ijmdes.com/ijmdes/article/view/115].
• [20] Li Q.: Predicting Trends of Bitcoin Prices Based on Machine Learning Methods. 4th International Conference on Software and e-Business, 2020, 49–52 [https://doi.org/10.1145/3446569.3446588].
• [21] Loh E. C.: Emerging Trend of Transaction and Investment: Bitcoin Price Prediction Using Machine Learning. International Journal of Advanced Trends in Computer Science and Engineering 9(1.4), 2020, 100–104 [https://doi.org/10.30534/ijatcse/2020/1591.42020].
• [22] Mangla N. et al.: Bitcoin price prediction using machine learning. International Journal of Information and Computing Science 6(5), 2019, 318–320.
• [23] Mujlid H.: A Survey on Machine Learning Approaches in Cryptocurrency: Challenges and Opportunities. 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), IEEE, 2023.
• [24] Nagamani P. et al.: Bitcoin Price Prediction Using Machine Learning Algorithms. Advances in engineering research/Advances in Engineering Research, 2023, 389–396 [https://doi.org/10.2991/978-94-6463-252-1_43].
• [25] Pabuçcu H. et al.: Forecasting the Movements of Bitcoin Prices: An Application of Machine Learning Algorithms. Quantitative Finance and Economics 4(4), 2020, 679–692 [https://doi.org/10.3934/qfe.2020031].
• [26] Parvez S. J. et al.: Bitcoin price prediction using Random Forest Regression. Journal of Positive School Psychology, 2022, 4352–4358.
• [27] Poongodi M. et al.: Bitcoin Price Prediction Using ARIMA Model. International Journal of Internet Technology and Secured Transactions 10(4), 2020, 396 [https://doi.org/10.1504/ijitst.2020.108130].
• [28] Pour E. S. et al.: Cryptocurrency Price Prediction with Neural Networks of LSTM and Bayesian Optimization. European Journal of Business and Management Research 7(2), 2022, 20–27 [https://doi.org/10.24018/ejbmr.2022.7.2.1307].
• [29] Pragadareddy K. T. et al.: Price prediction model of bitcoin using decision tree classification. International Journal of Food and Nutritional Sciences (IJFANS) 11(1), 2022.
• [30] Ranjan S. et al.: Bitcoin Price Prediction: A Machine Learning Sample Dimension Approach. Computational Economics 61(4), 2022, 1617–1636 [https://doi.org/10.1007/s10614-022-10262-6].
• [31] Reddy K. R. et al.: Bitcoin Price Prediction and Forecasting. International Research Journal of Engineering and Technology (IRJET) 9(04), 2022, 2395–0056.
• [32] Ren Y.-S. et al.: Past, Present, and Future of the Application of Machine Learning in Cryptocurrency Research. Research in International Business and Finance 63, 2022, 101799 [https://doi.org/10.1016/j.ribaf.2022.101799].
• [33] Roh Y. et al.: A Survey on Data Collection for Machine Learning: A Big Data – AI Integration Perspective. IEEE Transactions on Knowledge and Data Engineering 33(4), 2021, 1328–1347 [https://doi.org/10.1109/tkde.2019.2946162].
• [34] Sahi G. et al.: Predicting Cryptocurrency Price Using Machine Learning. European Economic Letters (EEL) 13(1), 2023, 11–16.
• [35] Samaddar M. et al.: A Comparative Study of Different Machine Learning Algorithms on Bitcoin Value Prediction. International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT), 2021 [https://doi.org/10.1109/icaect49130.2021.9392629].
• [36] Senthilkumar S., Nivedha B.: Bitcoin Price Prediction Using Ml. Social Science Research Network, 2022 [https://doi.org/10.2139/ssrn.4128261].
• [37] Shahbazi Z., Byun Y.-C.: Knowledge Discovery on Cryptocurrency Exchange Rate Prediction Using Machine Learning Pipelines. Sensors 22(5), 2022, 1740 [https://doi.org/10.3390/s22051740].
• [38] Shakri I. H.: Time Series Prediction Using Machine Learning: A Case of Bitcoin Returns. Studies in Economics and Finance 39(3), 2021, 458–470 [https://doi.org/10.1108/sef-06-2021-0217].
• [39] Shankhdhar A. et al.: Bitcoin Price Alert and Prediction System Using Various Models. IOP Conference Series. Materials Science and Engineering 1131(1), 2021, 012009 [https://doi.org/10.1088/1757-899x/1131/1/012009].
• [40] Squarepants S.: Bitcoin: A Peer-to-Peer Electronic Cash System. Social Science Research Network, 2008 [https://doi.org/10.2139/ssrn.3977007].
• [41] Suha S. A., Sanam T. F.: A Machine Learning Approach for Predicting Patient’s Length of Hospital Stay With Random Forest Regression. IEEE Region 10 Symposium (TENSYMP), 2022 [https://doi.org/10.1109/tensymp54529.2022.9864447].
• [42] Yan K., Wang Y.: Prediction of Bitcoin prices’ trends with ensemble learning models. Fifth International Conference on Computer Information Science and Artificial Intelligence (CISAI 2022), 2023, 900–905 [https://doi.org/10.1117/12.2667793].

Identyfikatory

DOI

10.35784/iapgos.5657