Performance Analysis of a Probabilistic Inductive Learning System

Abstract

This paper presents a new method for acquiring classificatory knowledge by induction. Based on a probabilistic inference technique, this method allows inherent patterns in noisy training instances to be easily detected. A set of classification rules can then be constructed based on these detected patterns. The proposed method has been evaluated by testing it with some real-world data sets and the results show that it ont-performs some decision-tree based algorithms both in terms of computational efficiency and classification accuracy.

Cite

Text

Chan and Wong. "Performance Analysis of a Probabilistic Inductive Learning System." International Conference on Machine Learning, 1990. doi:10.1016/B978-1-55860-141-3.50006-7

Markdown

[Chan and Wong. "Performance Analysis of a Probabilistic Inductive Learning System." International Conference on Machine Learning, 1990.](https://mlanthology.org/icml/1990/chan1990icml-performance/) doi:10.1016/B978-1-55860-141-3.50006-7

BibTeX

@inproceedings{chan1990icml-performance,
  title     = {{Performance Analysis of a Probabilistic Inductive Learning System}},
  author    = {Chan, Keith C. C. and Wong, Andrew K. C.},
  booktitle = {International Conference on Machine Learning},
  year      = {1990},
  pages     = {16-23},
  doi       = {10.1016/B978-1-55860-141-3.50006-7},
  url       = {https://mlanthology.org/icml/1990/chan1990icml-performance/}
}