Online Machine Learning in Big Data Streams.

摘要: The area of online machine learning in big data streams covers algorithms that are (1) distributed and (2) work from with only a limited possibility to store past data. first requirement mostly concerns software architectures efficient algorithms. second one also imposes nontrivial theoretical restrictions on the modeling methods: In stream model, older is no longer available revise earlier suboptimal decisions as fresh arrives. In this article, we provide an overview libraries well models for learning. We highlight most important ideas classification, regression, recommendation, unsupervised streaming data, show how they implemented various processing systems. This article reference material not survey. do attempt be comprehensive describing all existing methods solutions; rather, give pointers resources field. All related sub-fields, algorithms, learning, hugely dominant current research development conceptually new results components emerging at time writing. refer several survey results, both Compared surveys, our different because discuss recommender systems extended detail.