概念漂移数据流分类研究综述

摘要

由于现有各种机器学习算法本质上都是基于一个静态学习环境，而以尽量保证学习系统泛化能力为目标的寻优过程，概念漂移数据流分类给机器学习带来了巨大挑战。从数据流与概念漂移、概念漂移数据流分类研究的发展与趋势、概念漂移数据流分类的主要研究领域、概念漂移数据流分类研究的新动态4个方面展开了文献综述，并分析了当前概念漂移数据流分类算法存在的问题。%Because the current machine learning algorithms all are an optimization procedure that aims to ensure the generalization ability based on a static learning environment in essence , classification of data streams with concept drift has brought severe challenges to machine learning .The survey consists of four aspects : introduction to data streams and concept drift, the development process and future trends ,the main research fields , and the new trends in the study field of the classification of data streams with concept drift .The existing problems are discussed at last .