A comparative study of information extraction strategies

Ronen Feldman, Yonatan Aumann, Michal Finkelstein-Landau, Eyal Hurvitz, Yizhar Regev, Ariel Yaroshevich

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

23 Scopus citations

Abstract

The availability of online text documents exposes readers to a vast amount of potentially valuable knowledge buried therein. The sheer scale of material has created the pressing need for automated methods of discovering relevant information without having to read it all. Hence the growing interest in recent years in Text Mining. A common approach to Text Mining is Information Extraction (IE), extracting specific types (or templates) of information from a document collection. Although many works on IE have been published, researchers have not paid much attention to evaluate the contribution of syntactic and semantic analysis using Natural Language Processing (NLP) techniques to the quality of IE results. In this work we try to quantify the contribution of NLP techniques, by comparing three strategies for IE: naïve co-occurrence, ordered co-occurrence, and the structure-driven method - a rule-based strategy that relies on syntactic analysis followed by the extraction of suitable semantic templates. We use the three strategies for the extraction of two templates from financial news stories. We show that the structure-driven strategy provides significantly better precision results than the two other strategies (80-90% for the structure-driven compared with about only 60% for the co-occurrence and ordered co-occurrence). These results indicate that a syntactical and semantic analysis is necessary if one wishes to obtain high accuracy.

Original languageAmerican English
Title of host publicationComputational Linguistics and Intelligent Text Processing - 3rd International Conference, CICLing 2002, Proceedings
EditorsAlexander Gelbukh
PublisherSpringer Verlag
Pages349-359
Number of pages11
ISBN (Print)3540432191, 9783540457152
DOIs
StatePublished - 2002
Externally publishedYes
Event3rd Annual Conference on Intelligent Text Processing and Computational Linguistics, CICLing 2002 - Mexico City, Mexico
Duration: 17 Feb 200223 Feb 2002

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2276
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference3rd Annual Conference on Intelligent Text Processing and Computational Linguistics, CICLing 2002
Country/TerritoryMexico
CityMexico City
Period17/02/0223/02/02

Bibliographical note

Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 2002.

Keywords

  • Information extraction
  • Natural language processing

Fingerprint

Dive into the research topics of 'A comparative study of information extraction strategies'. Together they form a unique fingerprint.

Cite this