Skip to Main content Skip to Navigation
Conference papers

Fuzzy Cognitive Quantification

Anne Laurent 1 Charles Tijus Bernadette Bouchon-Meunier 2
1 TATOO - Fouille de données environnementales
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
2 MALIRE - Machine Learning and Information Retrieval
LIP6 - Laboratoire d'Informatique de Paris 6
Résumé : This article presents the results of a two stage investigation about how linguistic quantifiers are used to summarize expressions of quantity. In the first step subjects were asked to give verbal descriptions of arrays of percentages. Then a second group of subjects was asked to reproduce the original array from the verbal description. The second group produced quantities extremely similar to the original percentages even though the verbal descriptions they used did not describe all categories within the arrays. We shall show that quantifiers have implied meaning (e.g., between a large number and most) and that similar linguistic constructions may refer to amounts that are noticeably different (e.g., the principal vs. principally). Finally we highlight the importance of the implicit, topic-related meaning in the choice of spoken complements by showing how the concept of fuzzy quantifiers can be applied not only to their modeling, but also to their use in multi-dimensional data searches.
Keywords : Fuzzy logic Quantifier
Document type :
Conference papers
Complete list of metadata

Cited literature [14 references]  Display  Hide  Download
Contributor : Christine Carvalho de Matos <>
Submitted on : Monday, October 16, 2006 - 8:29:55 AM
Last modification on : Monday, February 22, 2021 - 4:34:01 PM
Long-term archiving on: : Thursday, September 20, 2012 - 11:55:45 AM



  • HAL Id : lirmm-00106502, version 1


Anne Laurent, Charles Tijus, Bernadette Bouchon-Meunier. Fuzzy Cognitive Quantification. 26th Annual Meeting of the Cognitive Science Society (CogSci), Aug 2004, Chicago, United States. ⟨lirmm-00106502⟩



Record views


Files downloads