软件导刊2023，Vol.22Issue(12)：71-77,7.DOI:10.11907/rjdk.222480
利用集成OS-ELM的不平衡数据流分类与存储方法
Classification and Storage Method for Imbalanced Data Streams by Ensemble OS-ELM
摘要
Abstract
At present,most classification algorithms for imbalanced data streams require a large amount of historical data to be saved and re-peatedly scanned during the training process to improve classification accuracy,which is inconsistent with the single channel characteristics of data streams,and the infinite nature of data streams requires a lot of memory space consumption.To this end,a classification method for im-balanced data streams based on integrated undersampling and online sequence exceeding learning machine(EU-OS-ELM)is proposed.First-ly,the base classifier selects an online learning algorithm OS-ELM that is suitable for data streams;Then,using non replacement random un-dersampling to construct a training set to enhance the robustness of the algorithm;Finally,a fixed size matrix is used to store the feature infor-mation of historical data,which improves the accuracy of data stream classification while minimizing the required additional memory space.The comparative experimental results between EU-OS-ELM and mainstream algorithms on some datasets show that the proposed algorithm on-ly requires an additional memory space of 0.8906 KB on all datasets,proving the effectiveness of the algorithm.关键词
不平衡数据流/在线序列超限学习机/分类/存储/集成Key words
imbalanced data stream/OS-ELM/classification/storage/ensemble分类
信息技术与安全科学引用本文复制引用
汤程皓,梅颖,卢诚波..利用集成OS-ELM的不平衡数据流分类与存储方法[J].软件导刊,2023,22(12):71-77,7.基金项目
浙江省自然科学基金项目(LY18F030003) （LY18F030003）
