A fast learning method for streaming and randomly ordered multi-class data chunks by using one-pass-throw-away class-wise learning concept
作者：
Highlights：
• We propose the 1-pass-throw-away learning method to classify a large or stream data.
• Parameter update for versatile elliptic function in chunk data case is presented.
• Both incremental and batch methods are used to compare with the proposed method.
• The method gives high accuracy in classification on 5-fold cross-validation test.
• The proposed method takes the fast learning time and less complexity on large data.
摘要
•We propose the 1-pass-throw-away learning method to classify a large or stream data.•Parameter update for versatile elliptic function in chunk data case is presented.•Both incremental and batch methods are used to compare with the proposed method.•The method gives high accuracy in classification on 5-fold cross-validation test.•The proposed method takes the fast learning time and less complexity on large data.
论文关键词：Classification,Versatile elliptic basis function,Incremental learning algorithm
论文评审过程：Received 5 June 2015, Revised 30 June 2016, Accepted 1 July 2016, Available online 5 July 2016, Version of Record 15 July 2016.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.07.002
