Individualized learning for improving kernel Fisher discriminant analysis
作者：
Highlights：
• A novel concept “the individualized learning” is introduced for the first time.
• We propose the individualized KFDA (IKFDA) using the idea of the individualized learning.
• IKFDA is very suitable for dealing with the large-scale and high-dimensional data sets.
• IKFDA exploits multiple similarity measures to sufficiently learn the test samples.
• IKFDA can outperform many state-of-the-art classification methods.
摘要
•A novel concept “the individualized learning” is introduced for the first time.•We propose the individualized KFDA (IKFDA) using the idea of the individualized learning.•IKFDA is very suitable for dealing with the large-scale and high-dimensional data sets.•IKFDA exploits multiple similarity measures to sufficiently learn the test samples.•IKFDA can outperform many state-of-the-art classification methods.
论文关键词：Individualized learning,KFDA,Individualized KFDA (IKFDA),High-dimensional,Similarity measure
论文评审过程：Received 23 September 2012, Revised 19 March 2016, Accepted 23 March 2016, Available online 6 April 2016, Version of Record 26 May 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.03.029
