ANALYZING THE EFFECTIVENESS OF WEKA PREDICTION ALGORITHMS FOR FORECASTING BITCOIN BEHAVIOR

Visualizações: 12

Authors

  • Daniel Pimenta Gonçalves da Fonte Federal University of Technology – Paraná https://orcid.org/0009-0009-9846-4906
  • Sandro Carvalho Izidoro Federal University of Itajubá https://orcid.org/0000-0001-5555-3321
  • José Donizetti de Lima Federal University of Technology – Paraná https://orcid.org/0000-0001-5260-9035
  • Sergio Luiz Pessa Ribas Federal University of Technology – Paraná
  • Matheus Henrique Dal Molin Ribeiro Federal University of Technology – Paraná
  • Érick Oliveira Rogrigues Federal University of Technology – Paraná

DOI:

https://doi.org/10.56579/rei.v8i2.2629

Keywords:

BTC, Algorithms, Time Series, WEKA, Cryptocurrency

Abstract

This study investigated the predictive performance of WEKA software in Bitcoin (BTC) time series forecasting, comparing its results with those obtained through R using historical data from May 2013 to December 2024 (4,254 observations). The analysis employed multiple machine learning algorithms evaluated through MAPE, RMSE, and MAE measures. Results revealed that while R generally achieved superior performance across most models, particularly in terms of error reduction, WEKA showed competitive and in some cases, superior performance in specific models: M5Rules reached MAPE of 2.25% and RMSE of $1,449.49, while SMOreg obtained MAPE of 2.40% and RMSE of $1,450.25 in the Forecast tab. Rule-based models, M5Rules and M5P, exhibited notably stronger performance in R, with MAPE of 3.79% and RMSE of $2,112.25, compared to modest results in WEKA where they showed correlation near zero in the Classify tab. The KPSS test confirmed series non-stationarity (p-value = 0.01), and Granger causality test identified significant temporal dependence at lag 5 (p-value = 8.7e-06). Findings indicate that while WEKA offers adequate functionality for specific models in sequential forecasting tasks, it exhibits instabilities and limitations in comprehensive temporal modeling, making R a more robust alternative for complex financial applications.

Downloads

Download data is not yet available.

Author Biographies

Daniel Pimenta Gonçalves da Fonte, Federal University of Technology – Paraná

Student of the Graduate Program (PPGEPS) at the Federal University of Technology – Paraná. Brazil, Paraná, Pato Branco.

Sandro Carvalho Izidoro, Federal University of Itajubá

Holds a degree in Data Processing Technology from Universidade José do Rosário Vellano (1993), a Bachelor’s degree in Computer Science from Universidade José do Rosário Vellano (1994), a Master’s degree in Electrical Engineering from the Federal University of Itajubá (2001), and a Ph.D. in Bioinformatics from the Federal University of Minas Gerais (2015), including a sandwich period at the Centre National de Séquençage Genoscope (CEA – France). Currently, he is an Associate Professor at the Federal University of Itajubá (Itabira Campus). He has experience in Structural Bioinformatics and Computer Science, with an emphasis on Artificial Intelligence, Genetic Algorithms, and Protein Function Prediction.

 

José Donizetti de Lima, Federal University of Technology – Paraná

Holds a degree in Data Processing Technology from Universidade José do Rosário Vellano (1993), a Bachelor’s degree in Computer Science from Universidade José do Rosário Vellano (1994), a Master’s degree in Electrical Engineering from the Federal University of Itajubá (2001), and a Ph.D. in Bioinformatics from the Federal University of Minas Gerais (2015), including a sandwich period at the Centre National de Séquençage Genoscope (CEA – France). Currently, he is an Associate Professor at the Federal University of Itajubá (Itabira Campus). He has experience in Structural Bioinformatics and Computer Science, with an emphasis on Artificial Intelligence, Genetic Algorithms, and Protein Function Prediction.

Sergio Luiz Pessa Ribas, Federal University of Technology – Paraná

Mechanical Engineer [UNISINOS–1988], Master’s degree in Engineering and Materials Science from Universidade Estadual de Ponta Grossa (2005), Ph.D. in Production Engineering from UFRGS (2010). Professor at the Federal University of Technology – Paraná in the following programs: Undergraduate: Industrial Maintenance Technology and Mechanical Engineering; Graduate Program (Stricto Sensu): Production and Systems Engineering – PPGEPS – UTFPR/PB. Research and professional activities in the areas of: Mechanical Engineering, including Industrial Maintenance (Management and Evaluation); Maintenance and Fleet Evaluation; Development of Composite Materials (Functional and Structural); Ergonomics and Occupational Safety; Ageing and ICT; Ergonomic Assessment Methods and Occupational Ergonomics; Management of Ergonomics and Occupational Safety Systems; Risk Management in Production Systems.

Matheus Henrique Dal Molin Ribeiro, Federal University of Technology – Paraná

Bachelor’s degree in Mathematics from the Federal University of Technology – Paraná (UTFPR), Master’s degree in Biostatistics from the State University of Maringá, and Ph.D. in Production and Systems Engineering from the Pontifical Catholic University of Paraná, for which he received an Honorable Mention in the CAPES Thesis Award 2022 in the Engineering III area. Associate Professor in Higher Education at UTFPR, Pato Branco campus, affiliated with the Department of Mathematics (DAMAT). Permanent Professor in the Graduate Program in Production and Systems Engineering (PPGEPS) at UTFPR, Pato Branco campus. He has experience in the areas of mathematics, probability and statistics, and data science, working mainly on the following topics: Data Science, artificial intelligence, machine learning, time series analysis and forecasting, and optimization methods.

Érick Oliveira Rogrigues, Federal University of Technology – Paraná

Associate Professor of Computer Science at the Federal University of Technology – Paraná (UTFPR) and permanent professor in the Graduate Program in Production and Systems Engineering (PPGEPS) at the Pato Branco and Curitiba campuses. He holds a Ph.D. in Computer Science from the Federal Fluminense University, including a 6-month period at the Petroleum Institute at Khalifa University in Abu Dhabi (focus: visual computing and machine learning), where he worked as a research assistant. He received an Honorable Mention from CAPES for his doctoral thesis (one of the top 3 theses in Computer Science in Brazil). He also received the Best Doctoral Thesis Excellence Award in Science and Technology from UFF. He holds a Master’s degree in Computing from the Federal Fluminense University and a Bachelor’s degree in Computer Systems from UFF. On that occasion, he also received the award for best Master’s thesis in the exact and technological sciences from UFF. He has also received other awards related to scientific papers and teaching honors (patron and honored professor). He has publications in top international journals in his field. He has worked with several programming languages over more than 20 years of experience in the field. He has led and contributed to the creation of several projects, including games, image libraries, artificial intelligence software, medical image processing, among others. His main areas of interest include: image processing, pattern recognition, knowledge discovery, machine learning, data mining, general applications of artificial intelligence, optimization (programming and processes), parallel programming on GPUs, applications of computing in healthcare (medical informatics, diagnosis, support systems, etc.), simulators, optimization, games and engines, and multiplatform applications (including mobile, web, and desktop). His main programming languages include Java, C/C++, CUDA, JavaScript, PHP, CSS, GPU shaders, among others.

References

ALI, A. et al. Network intrusion detection leveraging machine learning and feature selection. In: IEEE. 17th International Conference on Smart Communities: Improving Quality of Life Using ICT, IoT and AI (HONET), 2020, [S.l.]. Anais [...]. Piscataway: IEEE, 2020. p. 49–53. DOI: https://doi.org/10.1109/HONET50430.2020.9322813

BEHNOOD, A. et al. Prediction of the compressive strength of normal and high-performance concretes using M5P model tree algorithm. Construction and Building Materials, v. 142, p. 199–207, 2017. DOI: https://doi.org/10.1016/j.conbuildmat.2017.03.061

BREIMAN, L. Bagging predictors. Machine Learning, v. 24, p. 123–140, 1996. DOI: https://doi.org/10.1023/A:1018054314350

BREIMAN, L. Random forests. Machine Learning, v. 45, p. 5–32, 2001. DOI: https://doi.org/10.1023/A:1010933404324

CARUANA, R. et al. Ensemble selection from libraries of models. In: Proceedings of the Twenty-First International Conference on Machine Learning, 2004, [S.l.]. Anais [...]. [S.l.]: [s.n.], 2004. p. 18. DOI: https://doi.org/10.1145/1015330.1015432

CHOWDHURY, R. et al. An approach to predict and forecast the price of constituents and index of cryptocurrency using machine learning. Physica A: Statistical Mechanics and its Applications, v. 551, p. 124569, 2020. DOI: https://doi.org/10.1016/j.physa.2020.124569

CONTRERAS, R. C. et al. Genetic algorithm for feature selection applied to financial time series monotonicity prediction: Experimental cases in cryptocurrencies and Brazilian assets. Entropy, v. 26, n. 3, p. 177, 2024. DOI: https://doi.org/10.3390/e26030177

DERBENTSEV, V. et al. Forecasting cryptocurrency prices time series using machine learning approach. In: SHS WEB OF CONFERENCES, 2019, [S.l.]. Anais [...]. [S.l.]: EDP Sciences, 2019. v. 65, p. 02001. DOI: https://doi.org/10.1051/shsconf/20196502001

DERBENTSEV, V.; MATVIYCHUK, A.; SOLOVIEV, V. N. Forecasting of cryptocurrency prices using machine learning. In: Advanced Studies of Financial Technologies and Cryptocurrency Markets. [S.l.]: Springer, 2020. p. 211–231. DOI: https://doi.org/10.1007/978-981-15-4498-9_12

DERBENTSEV, V. et al. Forecasting cryptocurrency prices using ensembles-based machine learning approach. In: IEEE. International Conference on Problems of Infocommunications. Science and Technology (PIC S&T), 2020, [S.l.]. Anais [...]. Piscataway: IEEE, 2020. p. 707–712. DOI: https://doi.org/10.1109/PICST51311.2020.9468090

FANG, Q. et al. Prediction of blast-induced ground vibration in open-pit mines using a new technique based on imperialist competitive algorithm and M5Rules. Natural Resources Research, v. 29, n. 2, p. 791–806, 2020. DOI: https://doi.org/10.1007/s11053-019-09577-3

FRANK, E.; MAYO, M.; KRAMER, S. Alternating model trees. In: Proceedings of the 30th Annual ACM Symposium on Applied Computing, 2015, [S.l.]. Anais [...]. [S.l.]: [s.n.], 2015. p. 871–878. DOI: https://doi.org/10.1145/2695664.2695848

GNANAMBAL, S. et al. Classification algorithms with attribute selection: an evaluation study using WEKA. International Journal of Advanced Networking and Applications, v. 9, n. 6, p. 3640–3644, 2018.

HARYADI, D. et al. Implementation of support vector regression for polkadot cryptocurrency price prediction. JOIV: International Journal on Informatics Visualization, v. 6, n. 1-2, p. 201–207, 2022. DOI: https://doi.org/10.30630/joiv.6.1-2.945

HOROWITZ, J.; KLEMELÄ, J.; MAMMEN, E. Optimal estimation in additive regression models. Bernoulli, v. 12, n. 2, p. 271–298, 2006. DOI: https://doi.org/10.3150/bj/1145993975

KALEKAR, P. S. Time series forecasting using Holt-Winters exponential smoothing. Kanwal Rekhi School of Information Technology, v. 4329008, n. 13, p. 1–13, 2004.

KARGAR, K.; SAFARI, M. J. S.; KHOSRAVI, K. Weighted instances handler wrapper and rotation forest-based hybrid algorithms for sediment transport modeling. Journal of Hydrology, v. 598, p. 126452, 2021. DOI: https://doi.org/10.1016/j.jhydrol.2021.126452

KURNIAWAN, K.; MADELAN, S. Forecasting using time series analysis method in cryptocurrency period 2015–2022. International Journal of Innovative Science and Research Technology, v. 7, p. 1454–1459, 2022.

LV, L.-T.; JI, N.; ZHANG, J.-L. A RBF neural network model for anti-money laundering. In: International Conference on Wavelet Analysis and Pattern Recognition, 2008, [S.l.]. Anais [...]. Piscataway: IEEE, 2008. v. 1, p. 209–215. DOI: https://doi.org/10.1109/ICWAPR.2008.4635778

NEWSON, R. Parameters behind “nonparametric” statistics: Kendall’s tau, Somers’ D and median differences. The Stata Journal, v. 2, n. 1, p. 45–64, 2002. DOI: https://doi.org/10.1177/1536867X0200200103

NIRANJAN, A. et al. ERCR TV: Ensemble of random committee and random tree for efficient anomaly classification using voting. In: Proceedings of the 2018 3rd International Conference for Convergence in Technology (I2CT), 2018, [S.l.]. Anais [...]. Piscataway: IEEE, 2018. p. 1–5. DOI: https://doi.org/10.1109/I2CT.2018.8529797

PAUDEL, N.; BHATTA, J. Mushroom classification using random forest and REP tree classifiers. Nepal Journal of Mathematical Sciences, v. 3, n. 1, p. 111–116, 2022. DOI: https://doi.org/10.3126/njmathsci.v3i1.44130

POYRAZ, İ.; GÜRHANLI, A. Demand forecasting with time series analysis using drug sales data. International Journey of Engineering Research and Applications, v. 10, n. 7, p. 51-54, 2020.

RAMCHOUN, H. et al. Multilayer perceptron: Architecture optimization and training. International Journal of Interactive Multimedia and Artificial Intelligence, v. 4, n. 1, p. 26-30, 2016. DOI: https://doi.org/10.9781/ijimai.2016.415

REDDY, B. H. M.; REDDY, S. V.; SAROJAMMA, B. Data Mining Techniques for estimation of wind speed using WEKA. International Journal of Computer Sciences and Engineering (IJCSE), v. 9, n. 9, p. 49–53, 2021. DOI: https://doi.org/10.26438/ijcse/v9i9.4851

SABANCI, K.; ÜNLERŞEN, M. F.; POLAT, K. Classification of different forest types with machine learning algorithms. Research in Rural Development, v. 1, p. 254–260, 2016.

SAMSON, T. K. Comparative analysis of machine learning algorithms for daily cryptocurrency price prediction. Information Dynamics and Applications, v. 3, n. 1, p. 64–76, 2024. DOI: https://doi.org/10.56578/ida030105

SCHULZ, E.; SPEEKENBRINK, M.; KRAUSE, A. A tutorial on Gaussian process regression: Modelling, exploring, and exploiting functions. Journal of Mathematical Psychology, v. 85, p. 1–16, 2018. DOI: https://doi.org/10.1016/j.jmp.2018.03.001

SHI, L. et al. Signal prediction based on boosting and decision stump. International Journal of Computational Science and Engineering, v. 16, n. 2, p. 117–122, 2018. DOI: https://doi.org/10.1504/IJCSE.2018.090450

SINGH, P.; AGRAWAL, S. Node localization in wireless sensor networks using the M5P tree and SMOreg algorithms. In: 5th International Conference on Computational Intelligence and Communication Networks, 2013, [S.l.]. Anais [...]. Piscataway: IEEE, 2013. p. 104–104. DOI: https://doi.org/10.1109/CICN.2013.32

SKOWRON, A.; SURAJ, Z. A parallel algorithm for real-time decision making: a rough set approach. Journal of Intelligent Information Systems, v. 7, p. 5–28, 1996. DOI: https://doi.org/10.1007/BF00125520

SU, X.; YAN, X.; TSAI, C.-L. Linear regression. Wiley Interdisciplinary Reviews: Computational Statistics, v. 4, n. 3, p. 275–294, 2012. DOI: https://doi.org/10.1002/wics.1198

SUN, G. Cryptocurrency price prediction based on Xgboost, LightGBM and BNN. Applied and Computational Engineering, v. 49, p. 273–279, 2024. DOI: https://doi.org/10.54254/2755-2721/49/20241414

Downloads

Published

2026-04-21

How to Cite

Fonte, D. P. G. da, Izidoro, S. C., Lima, J. D. de, Ribas, S. L. P., Ribeiro, M. H. D. M., & Rogrigues, Érick O. (2026). ANALYZING THE EFFECTIVENESS OF WEKA PREDICTION ALGORITHMS FOR FORECASTING BITCOIN BEHAVIOR. Interdisciplinary Studies Journal, 8(2), 01–25. https://doi.org/10.56579/rei.v8i2.2629

Issue

Vol. 8 No. 2 (2026): JOURNAL OF INTERDISCIPLINARY STUDIES

Section

ARTICLES

License

Copyright (c) 2026 Interdisciplinary Studies Journal

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

The Journal of Interdisciplinary Studies adopts the Creative Commons Attribution 4.0 International License (CC BY 4.0), which allows for sharing and adapting the work, including for commercial purposes, provided proper attribution is given and the original publication in this journal is acknowledged.

Metrics