Posters

Principles of Data Mining and Knowledge Discovery

Volume 1510 of the series Lecture Notes in Computer Science pp 459-467

Date:

Clasitex+: A tool for knowledge discovery from texts

  • José Francisco Martínez TrinidadAffiliated withCentro de Investigación en Computación, I. P. N.
  • , Beatriz Beltrán MartínezAffiliated withFacultad de Ciencias de la Computación, Benemérita Universidad Autónoma de Puebla
  • , Adolfo Guzmán ArenasAffiliated withCentro de Investigación en Computación, I. P. N.
  • , José Ruiz SchulcloperAffiliated withCentro de Investigación en Computación, I. P. N.Instituto de Cibernética, Matemática y Física, CITMA

* Final gross prices may vary according to local VAT.

Get Access

Abstract

In this work the CLASITEX+ system for discovers the most important themes treated in a text written in Spanish or English is presented. This system works on the basis of trees of concepts and find: a) the most frequent concepts in the text, b) the relation between these concepts computing the co-ocurrence into the sentences that conform the text. Also CLASITEX+ can give us a distribution map of the most frequent concepts in the text An important characteristic of the system is the amount of concepts in Spanish and English handled by the system, also the execution time in the document analysis is very acceptable.