摘要
随着科学技术的快速发展，计算机技术也得到了广泛的应用。在过去的十几年里，机器学习领域中主要以集成学习为主，并且集成学习成为当时一个备受关注的课题。现阶段，Bagging和Booting是最为流行的集成学习方法。神经网络或决策树构造方法不是很稳定，虽然被作为集成学习算法。由于这种算法不是很稳定，需要对原始的Booting算法进行改进才能够满足现实情况的需要，改进后的Boot算法称为LAdaBoot算法。该文主要论述了原始的Boot算法，并详细的论述了LAdaBoot算法，同时与其他的算法进行了相关比较，进而为相关人员提供参考依据。
with the rapid development of science and technology, computer technology has been widely used. In the past 10 years, in the field of machine learning mainly integrated learning based, and ensemble learning became a topic of concern. At this stage, Bagging and Booting are the most popular ensemble learning method. Neural network or a decision tree construction method is not very stable, although as the ensemble learning algorithm. As a result of this algorithm is not very stable, the need for the original Booting algorithm can meet the needs of the reality of the situation is improved, the improved Boot algorithm called LAdaBoot al- gorithm. This paper mainly discusses the original Boot algorithm, and discusses in details the LAdaBoot algorithm, at the same time were related to compare with other algorithms, and provide reference for the related personnel.
作者
王超
WANG Chao （Guangzhou HOLLEY Science and; Technology Career Technical College, Guangzhou 511325, China）
出处
《电脑知识与技术》
2015年第3期286-289,共4页
Computer Knowledge and Technology
