Conditional Relative Frequency Distributions with Undefined Observations and Generalized Fuzzy Orthopartitions

Abstract

Conditional relative frequency distributions are tools extensively employed in statistics and machine learning for analyzing connections of two or more categorical variables, examining patterns, and comparing data. As a first goal, we introduce the so-called conditional relative frequency distributions with undefined observations for representing frequencies characterized by uncertainty. After that, we show that conditional relative frequency distributions with undefined observations can be identified with particular generalized fuzzy orthopartitions, which are mathematical models describing vague partitions where the membership of elements to classes is only partially known.

Cite

Text

Boffa and Ciucci. "Conditional Relative Frequency Distributions with Undefined Observations and Generalized Fuzzy Orthopartitions." Journal of Artificial Intelligence Research, 2025. doi:10.1613/JAIR.1.18020

Markdown

[Boffa and Ciucci. "Conditional Relative Frequency Distributions with Undefined Observations and Generalized Fuzzy Orthopartitions." Journal of Artificial Intelligence Research, 2025.](https://mlanthology.org/jair/2025/boffa2025jair-conditional/) doi:10.1613/JAIR.1.18020

BibTeX

@article{boffa2025jair-conditional,
  title     = {{Conditional Relative Frequency Distributions with Undefined Observations and Generalized Fuzzy Orthopartitions}},
  author    = {Boffa, Stefania and Ciucci, Davide},
  journal   = {Journal of Artificial Intelligence Research},
  year      = {2025},
  doi       = {10.1613/JAIR.1.18020},
  volume    = {83},
  url       = {https://mlanthology.org/jair/2025/boffa2025jair-conditional/}
}