Online Ensemble Methods for Binary Classification

• Arthur G. Soares UFMT
• Thiago P. da Silva UFMT

DOI: https://doi.org/10.5753/eri-mt.2024.245856

Abstract

In the context of Big Data, analyzing large volumes of data in real-time is crucial for quickly generating insights. Although data classification is challenging, online machine learning is more effective than traditional batch learning methods, especially for continuous changes. This study introduces four ensemble methods for the binary classification of online data streams, using diverse machine learning models. Experiments on eight datasets demonstrated the feasibility and strong potential of these approaches.

References

BASILI, V. R., CALDIERA, G., and RAMBACH, H. D. (1998). The goal question metric approach. Encyclopedia of software engineering., pages 528–532.

GOMES, H. M., BARDDAL, J. P., ENEMBREACK, F., and BIFET, A. (2017). A Survey on Ensemble Learning for Data Stream Classification. Association for Computing Machinery, pages 1–10.

KRAWCZYK, B., MINKU, L. L., GAMA, J., STEFANOWSKI, J., and WOZNIAK, M. (2017). Ensemble learning for data stream analysis: A survey. Information Fusion, pages 2–3, 133.

PEREIRA, L. F. and ROSSI, R. G. (2021). Estudo de técnicas de Ensemble para classificação binária. Universidade Federal de Mato Grosso do Sul, pages 1–2, 17.

SARKA, D., BALI, R., and SHARMA, T. (2018.). Practical Machine Learning with Python. ed. Apress.

SILVA, T. P., BATISTA, T. V., DELICATO, F. C., and PIRES, P. F. (2023). An Online Ensemble Method for Auto-Scaling NFV-based Applications in the Edge. Cluster Computing, pages 2–13.