集成学习是机器学习的重要研究方向之一,SVM集成近年来已经受到国内外很多从事机器学习、统计学习的研究者们的重视,并使得该领域成为了一个相当活跃的研究热点。对近年来SVM集成的研究与应用进行了综述,讨论了SVM集成需要解决的基本问题;讨论分析了构造差异性大的集成成员SVM的方法、有效的集成结论生成方法、SVM集成的典型应用;指出了目前存在的问题、以及几个重要的研究方向。
Ensemble learning is an important research direction in machine learning,in recent years,support vector machine(SVM) ensemble has attracted the attentions of researchers in the field of machine learning and statistical learning,and has become an active research hotspot.The research and application of SVM ensemble in recent years are surveyed in this paper and the fundamental problems in SVM ensemble are discussed.Methods of constructing SVM base classifiers with higher diversity,effective ways for combining base classifiers,and the typical applications of SVM ensemble are discussed and analyzed.The problems those exist in the research and application of SVM ensemble are pointed out and the important research directions are given.