Corpus-based and Knowledge-based Measures of Text Semantic Similarity

Description:

This article discusses corpus-based and knowledge-based measures of text semantic similarity.

Creator(s):
Creation Date: July 2006
Partner(s):
UNT College of Engineering
Collection(s):
UNT Scholarly Works
Usage:
Total Uses: 951
Past 30 days: 21
Yesterday: 0
Creator (Author):
Mihalcea, Rada, 1974-

University of North Texas

Creator (Author):
Corley, Courtney

University of North Texas

Creator (Author):
Strapparava, Carlo, 1962-

ITC-irst, Istituto per la Ricerca Scientifica e Tecnologica

Date(s):
  • Creation: July 2006
Description:

This article discusses corpus-based and knowledge-based measures of text semantic similarity.

Degree:
Note:

Copyright 2006 American Association for Artificial Intelligence (AAAI). All rights reserved. http://www.aaai.org

Note:

Abstract: This paper presents a method for measuring the semantic similarity of texts, using corpus-based and knowledge-based measures of similarity. Previous work on this problem has focused mainly on either large documents (e.g. text classification, information retrieval) or individual words (e.g. synonymy tests). Given that a large fraction of the information available today, on the Web and elsewhere, consists of short text snippets (e.g. abstracts of scientific documents, image captions, product descriptions), in this paper the authors focus on measuring the semantic similarity of short texts. Through experiments performed on a paraphrase data set, the authors show that the semantic similarity method out-performs methods based on simple lexical matching, resulting in up to 13% error rate reduction with respect to the traditional vector-based similarity metric.

Physical Description:

6 p.

Language(s):
Subject(s):
Keyword(s): lexical resources | texts | semantic similarities | specificity
Source: American Association for Artificial Intelligence (AAAI) Conference, 2006, Boston, Massachusetts, United States
Contributor(s):
Partner:
UNT College of Engineering
Collection:
UNT Scholarly Works
Identifier:
  • ARK: ark:/67531/metadc30981
Resource Type: Paper
Format: Text
Rights:
Access: Public