---
title: "FASTUS: A Finite-state Processor for Information Extraction from Real-world Text"
authors:
  - "Douglas E. Appelt"
  - "Jerry R. Hobbs"
  - "John Bear"
  - "David J. Israel"
  - "Mabry Tyson"
doi: ""
url: "https://ijcai.org/proceedings/1993/162"
pdf: "https://ijcai.org/proceedings/1993/0162.pdf"
venue: "Proceedings of the Thirteenth International Joint Conference on Artificial Intelligence"
year: 1993
pages: "1172-1177"
track: "Natural Language"
keywords:
---

# FASTUS: A Finite-state Processor for Information Extraction from Real-world Text

Douglas E. Appelt, Jerry R. Hobbs, John Bear, David J. Israel, Mabry Tyson

*Proceedings of the Thirteenth International Joint Conference on Artificial Intelligence*, pages 1172-1177.

## Abstract

Approaches to text. processing that. rely on parsing the t.ext with a wht.ext.-free gralllma.r tend to be slow and error-prone bPcause of t.he massive ambiguity of long sent.<~nces . ln contrast, FASTUS employs a noll(l<'t<'rlllillistic finite-state language modd that . produces a phrasal decomposition of a. S<~nt.enc<' into nouu groups, verb groups and particles . Anot.lwr finite-state machine recognizes domain-spP-<·.ifir phrases based 011 combinations of the heads of the constituents found in t.h< • first. pass . F'ASTUS has been evaltnl.ted 011 several blind t.<•st.s that. demonstrate that st.a.t.e-of-t.lw-art. l><'rformanre 011 information-ext .rac.t.ion t.asks is obtainable with surprisingly lit.tlf ~ conrput.at.iona.l t>ffort..