Bayesian Information Retrieval: Preliminary Evaluation
Abstract
Given a database of documents and a user’s query, how can we locate those documents that meet the user’s information needs? Because there is no precise definition of which documents in the database match the user’s query, uncertainty is inherent in the information retrieval process. Therefore, probability theory is a natural tool for formalizing the retrieval task. In this paper, we propose a Bayesian approach to one of the conventional probabilistic information retrieval models. We discuss the motivation for such a model, describe its implementation, and present some experimental results.
Cite
Text
Keim et al. "Bayesian Information Retrieval: Preliminary Evaluation." Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics, 1997.Markdown
[Keim et al. "Bayesian Information Retrieval: Preliminary Evaluation." Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics, 1997.](https://mlanthology.org/aistats/1997/keim1997aistats-bayesian/)BibTeX
@inproceedings{keim1997aistats-bayesian,
title = {{Bayesian Information Retrieval: Preliminary Evaluation}},
author = {Keim, Michelle and Lewis, David D. and Madigan, David},
booktitle = {Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics},
year = {1997},
pages = {303-318},
volume = {R1},
url = {https://mlanthology.org/aistats/1997/keim1997aistats-bayesian/}
}