Mathematical formulation and application of kernel tensor decomposition based unsupervised feature extraction

作者：

Highlights：

• Developed kernelized tensor decomposition method for unsupervised feature extraction.

• Method can realize efficient tensor computation in high dimensional feature space.

• Applied method in various biological applications to demonstrate effectivity.

• Proposed method can capture different underlying structures in data sets.

摘要

•Developed kernelized tensor decomposition method for unsupervised feature extraction.•Method can realize efficient tensor computation in high dimensional feature space.•Applied method in various biological applications to demonstrate effectivity.•Proposed method can capture different underlying structures in data sets.

论文关键词：Kernel trick,Tensor decomposition,Unsupervised learning,Feature extraction,Applications in biology and medicine

论文评审过程：Received 9 October 2020, Revised 27 January 2021, Accepted 30 January 2021, Available online 6 February 2021, Version of Record 11 February 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.106834