Online Machine Learning in Big Data Streams: Overview

András A. Benczúr, Levente Kocsis, Róbert Pálovics

Abstract: The area of online machine learning in big data streams covers algorithms that are distributed and work from data streams with only a limited possibility to store past data. The first requirement mostly concerns software architectures and efficient algorithms. The second one also imposes nontrivial theoretical restrictions on the modeling methods: In the data stream model, older data is no longer available to revise earlier suboptimal modeling decisions as the fresh data arrives. In this article, we provide an overview of the general requirements for online machine learning. We enumerate the distributed software architectures and libraries for online learning and show how various machine learning models are implemented in them. Some algorithms are described in more details in three more chapters of this handbook, including “Online Machine Learning Algorithms Over Data Streams”; “Reinforcement Learning, Unsupervised Methods, and Concept Drift in Stream Learning”; and “Recommender Systems Over Data Streams.” This article is a reference material and not a survey, with pointers to the most important resources and other surveys in the field.