Incremental Algorithms for Hierarchical Classification

Abstract

We study the problem of classifying data in a given taxonomy when classifications associated with multiple and/or partial paths are allowed. We introduce a new algorithm that incrementally learns a linear-threshold classifier for each node of the taxonomy. A hierarchical classification is obtained by evaluating the trained node classifiers in a top-down fashion. To evaluate classifiers in our multipath framework, we define a new hierarchical loss function, the H-loss, capturing the intuition that whenever a classification mistake is made on a node of the taxonomy, then no loss should be charged for any additional mistake occurring in the subtree of that node.

Cite

Text

Cesa-Bianchi et al. "Incremental Algorithms for Hierarchical Classification." Journal of Machine Learning Research, 2006.

Markdown

[Cesa-Bianchi et al. "Incremental Algorithms for Hierarchical Classification." Journal of Machine Learning Research, 2006.](https://mlanthology.org/jmlr/2006/cesabianchi2006jmlr-incremental/)

BibTeX

@article{cesabianchi2006jmlr-incremental,
  title     = {{Incremental Algorithms for Hierarchical Classification}},
  author    = {Cesa-Bianchi, Nicoló and Gentile, Claudio and Zaniboni, Luca},
  journal   = {Journal of Machine Learning Research},
  year      = {2006},
  pages     = {31-54},
  volume    = {7},
  url       = {https://mlanthology.org/jmlr/2006/cesabianchi2006jmlr-incremental/}
}