摘要
为了解决复杂工业过程的概念漂移问题,提高集成学习模型的泛化性能,基于分组遗传算法,提出一种用于提升基学习器间多样性的建模方法.该方法以在线极限学习机作为基学习器,根据基学习器在滑动窗口上的性能对其进行分组,并执行进化操作,同时引入基因流概念,增加了基学习器间的多样性,提高了集成算法在处理概念漂移数据流时的预测性能.最后使用合成数据集和真实数据集验证了所提算法的合理性与有效性.
To solve the problem of concept drift in complex industrial process and improve the generalization performance of the ensemble learning model,a modeling method was proposed to enhance diversity among base learners based on grouping genetic algorithm.The online sequential extreme learning machine(OS_ELM)was used as base learner.The base learners were grouped according to their performance on the sliding window,and evolution operations were performed.At the same time,the concept of gene flow was introduced,which increased the diversity among base learners and improved the prediction performance of the ensemble algorithm in dealing with the concept drift data streams.Finally,the rationality and effectiveness of the proposed algorithm were verified by using the synthetic data sets and real-world data sets.
作者
陈双叶
赵荣
符寒光
CHEN Shuangye;ZHAO Rong;FU Hanguang(Faculty of Information Technology,Beijing University of Technology,Beijing 100124,China;College of Materials Science and Engineering,Beijing University of Technology,Beijing 100124,China)
出处
《北京工业大学学报》
CAS
CSCD
北大核心
2021年第8期886-894,共9页
Journal of Beijing University of Technology
基金
国家重点研发计划资助项目(2017YFB0306404)。
关键词
分组遗传算法
基因流
集成学习
在线极限学习机
准确性
多样性
group genetic algorithm
gene flow
ensemble learning
online sequential extreme learning machine(OS_ELM)
accuracy
diversity
