Pattern Recognition with New Class Discovery

Abstract

Classification problems with non-mutually exclusive classes are discussed. A method, new class discovery (NCD), is proposed which can discover potentially unknown or untrained classes. It is formally proved that, in the non-mutually exclusive case, the error probability of the NCD method is less than the conventional classification method using Bayes's rule. Application of the method to real entomological data was successful: a species was discovered that had been independently verified by taxonomists.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Cite

Text

Chang and Loew. "Pattern Recognition with New Class Discovery." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1991. doi:10.1109/CVPR.1991.139730

Markdown

[Chang and Loew. "Pattern Recognition with New Class Discovery." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1991.](https://mlanthology.org/cvpr/1991/chang1991cvpr-pattern/) doi:10.1109/CVPR.1991.139730

BibTeX

@inproceedings{chang1991cvpr-pattern,
  title     = {{Pattern Recognition with New Class Discovery}},
  author    = {Chang, Inhao and Loew, Murray H.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {1991},
  pages     = {438-443},
  doi       = {10.1109/CVPR.1991.139730},
  url       = {https://mlanthology.org/cvpr/1991/chang1991cvpr-pattern/}
}