Methods of evaluating degrees of truth for linguistic summaries of data: a comparative analysis

  • Authors:
  • Adam Niewiadomski;Oskar Korczak

  • Affiliations:
  • Institute of Information Technology, Technical University of Łódź, Poland;Institute of Information Technology, Technical University of Łódź, Poland

  • Venue:
  • ICAISC'10 Proceedings of the 10th international conference on Artificial intelligence and soft computing: Part I
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The paper, like some of our previous publications, focuses on linguistic summaries of databases in the sense of Yager [1], with further modifications by Kacprzyk and Yager [2]. In particular, we are interested in some alternative methods of evaluating degrees of truth, DTs, for linguistic summaries. The considered methods, for instance, Sugeno integral, GD method, or MVCP, are alternatives for "traditional" DTs, by Zadeh [3]. Our original contribution is an analysis of these methods and their interpretation in terms of linguistic summarization of databases. Especially, different DTs for a summary are evaluated and the computational cost is assessed. Based on that, sensitivity of the methods to different parameters of linguistic summaries, e.g. increasing number of qualifiers or the first/the second form, single/composed summarizers, relative/absolute/coherent quantifiers, are examined.