Bayesian Information Extraction Network

TL;DR

This paper demonstrates how Dynamic Bayesian Networks can be effectively used for natural language information extraction by integrating various linguistic tools, resulting in improved performance over existing methods.

Contribution

It introduces a novel application of DBNs for information extraction, combining multiple linguistic features into a unified probabilistic model.

Findings

Outperforms previous results on a benchmark domain

Integrates shallow parsing, tagging, and recognition in a single model

Shows potential of DBNs for complex NLP tasks

Abstract

Dynamic Bayesian networks (DBNs) offer an elegant way to integrate various aspects of language in one model. Many existing algorithms developed for learning and inference in DBNs are applicable to probabilistic language modeling. To demonstrate the potential of DBNs for natural language processing, we employ a DBN in an information extraction task. We show how to assemble wealth of emerging linguistic instruments for shallow parsing, syntactic and semantic tagging, morphological decomposition, named entity recognition etc. in order to incrementally build a robust information extraction system. Our method outperforms previously published results on an established benchmark domain.