Improvement of the State Merging Rule on Noisy Data in Probabilistic Grammatical Inference
Abstract
In this paper we study the influence of noise in probabilistic grammatical inference. We paradoxically bring out the idea that specialized automata deal better with noisy data than more general ones. We propose then to replace the statistical test of the Alergia algorithm by a more restrictive merging rule based on a test of proportion comparison. We experimentally show that this way to proceed allows us to produce larger automata that better treat noisy data, according to two different performance criteria (perplexity and distance to the target model).
Cite
Text
Habrard et al. "Improvement of the State Merging Rule on Noisy Data in Probabilistic Grammatical Inference." European Conference on Machine Learning, 2003. doi:10.1007/978-3-540-39857-8_17Markdown
[Habrard et al. "Improvement of the State Merging Rule on Noisy Data in Probabilistic Grammatical Inference." European Conference on Machine Learning, 2003.](https://mlanthology.org/ecmlpkdd/2003/habrard2003ecml-improvement/) doi:10.1007/978-3-540-39857-8_17BibTeX
@inproceedings{habrard2003ecml-improvement,
title = {{Improvement of the State Merging Rule on Noisy Data in Probabilistic Grammatical Inference}},
author = {Habrard, Amaury and Bernard, Marc and Sebban, Marc},
booktitle = {European Conference on Machine Learning},
year = {2003},
pages = {169-180},
doi = {10.1007/978-3-540-39857-8_17},
url = {https://mlanthology.org/ecmlpkdd/2003/habrard2003ecml-improvement/}
}