A Study on the Performance of Large Bayes Classifier

Abstract

Large Bayes (LB) is a recently introduced classifier built from frequent and interesting itemsets. LB uses itemsets to create context-specific probabilistic models of the data and estimate the conditional probability P(c _i| A) of each class c _i given a case A . In this paper we use chi-square tests to address several drawbacks of the originally proposed interestingness metric, namely: (i) the inability to capture certain really interesting patterns, (ii) the need for a user-defined and data dependent interestingness threshold, and (iii) the need to set a minimum support threshold. We also introduce some pruning criteria which allow for a trade-off between complexity and speed on one side and classification accuracy on the other. Our experimental results show that the modified LB outperforms the original LB, Naïve Bayes, C4.5 and TAN.

Cite

Text

Meretakis et al. "A Study on the Performance of Large Bayes Classifier." European Conference on Machine Learning, 2000. doi:10.1007/3-540-45164-1_29

Markdown

[Meretakis et al. "A Study on the Performance of Large Bayes Classifier." European Conference on Machine Learning, 2000.](https://mlanthology.org/ecmlpkdd/2000/meretakis2000ecml-study/) doi:10.1007/3-540-45164-1_29

BibTeX

@inproceedings{meretakis2000ecml-study,
  title     = {{A Study on the Performance of Large Bayes Classifier}},
  author    = {Meretakis, Dimitris and Lu, Hongjun and Wüthrich, Beat},
  booktitle = {European Conference on Machine Learning},
  year      = {2000},
  pages     = {271-279},
  doi       = {10.1007/3-540-45164-1_29},
  url       = {https://mlanthology.org/ecmlpkdd/2000/meretakis2000ecml-study/}
}