Research and Publications

Introduction

This 2024 KDD publication provides a practical, accessible overview of machine learning techniques designed for real‑time, evolving data streams. It introduces the core challenges of streaming data—such as concept drift, limited labels, and continuous adaptation—and demonstrates how modern online learning algorithms and tools can be applied effectively in real‑world environments.

Key outcomes

• Presents a practitioner‑focused introduction to streaming and online machine learning.

• Explains how to build models that adapt to concept drift and shifting data distributions.

• Highlights modern frameworks and tools for deploying streaming ML systems.

• Bridges theory and practice through examples relevant to large‑scale, dynamic data applications.

Funding

This work was supported by TAIAO – Time‑Evolving Data Science / Artificial Intelligence for Advanced Open Environmental Science, funded by the New Zealand Ministry of Business, Innovation and Employment (MBIE).

How to cite this article

APA 7th Edition: Gomes, Heitor Murilo, et al. “Adaptive Random Forests for Evolving Data Stream Classification.” Machine Learning, vol. 106, no. 9–10, 2017, pp. 1469–1495. https://doi.org/10.1145/3637528.3671442. 

MLA 9th Edition: Gomes, Heitor M., and Albert Bifet. “Practical Machine Learning for Streaming Data.” Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’24), 2024, https://doi.org/10.1145/3637528.3671442.

Chicago (Author-Date): Gomes, Heitor M., and Albert Bifet. 2024. “Practical Machine Learning for Streaming Data.” In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’24). https://doi.org/10.1145/3637528.3671442