Semi-Supervised Feature Selection with Adaptive Discriminant Analysis

Abstract

In this paper, we propose a novel Adaptive Discriminant Analysis for semi-supervised feature selection, namely SADA. Instead of computing fixed similarities before performing feature selection, SADA simultaneously learns an adaptive similarity matrix S and a projection matrix W with an iterative method. In each iteration, S is computed from the projected distance with the learned W and W is computed with the learned S. Therefore, SADA can learn better projection matrix W by weakening the effect of noise features with the adaptive similarity matrix. Experimental results on 4 data sets show the superiority of SADA compared to 5 semisupervised feature selection methods.

Cite

Text

Zhong et al. "Semi-Supervised Feature Selection with Adaptive Discriminant Analysis." AAAI Conference on Artificial Intelligence, 2019. doi:10.1609/AAAI.V33I01.330110083

Markdown

[Zhong et al. "Semi-Supervised Feature Selection with Adaptive Discriminant Analysis." AAAI Conference on Artificial Intelligence, 2019.](https://mlanthology.org/aaai/2019/zhong2019aaai-semi/) doi:10.1609/AAAI.V33I01.330110083

BibTeX

@inproceedings{zhong2019aaai-semi,
  title     = {{Semi-Supervised Feature Selection with Adaptive Discriminant Analysis}},
  author    = {Zhong, Weichan and Chen, Xiaojun and Yuan, Guowen and Li, Yiqin and Nie, Feiping},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2019},
  pages     = {10083-10084},
  doi       = {10.1609/AAAI.V33I01.330110083},
  url       = {https://mlanthology.org/aaai/2019/zhong2019aaai-semi/}
}