Evaluation of knowledge bases by means of multi-dimensional OWA operators

  • Authors:

  • Isabel Aguiló;Javier Martín;Gaspar Mayor;Jaume Suòer

  • Affiliations:

  • Department of Mathematics and Computer Science, University of the Balearic Islands, 07122 Palma de Mallorca, Illes Balears, Spain, {isabel.aguilo, javier.martin, gmayor, jaume.sunyer}@uib.es;Department of Mathematics and Computer Science, University of the Balearic Islands, 07122 Palma de Mallorca, Illes Balears, Spain, {isabel.aguilo, javier.martin, gmayor, jaume.sunyer}@uib.es;Department of Mathematics and Computer Science, University of the Balearic Islands, 07122 Palma de Mallorca, Illes Balears, Spain, {isabel.aguilo, javier.martin, gmayor, jaume.sunyer}@uib.es;Department of Mathematics and Computer Science, University of the Balearic Islands, 07122 Palma de Mallorca, Illes Balears, Spain, {isabel.aguilo, javier.martin, gmayor, jaume.sunyer}@uib.es

  • Venue:
  • Proceedings of the 2005 conference on Artificial Intelligence Research and Development
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper we present the evaluation of the quality of knowledge bases by means of multi-dimensional aggregation operators. The knowledge bases to be evaluated are rule bases of the type antecedent-consequent, where we assign to each base a list of the relevance factors corresponding to the rules. This relevance factor is obtained by measuring certain considered qualitative or quantitative characteristics. The quality coefficient of the base is then obtained by aggregation of the list of relevance factors. Since each knowledge base has a different number of rules, these lists of relevance factors have different lengths. Thus, the use of multi-dimensional aggregation functions to obtain the quality coefficient allows us to compare the rule bases. The analysis of the properties that the aggregation function should satisfy leads to the use of multi-dimensional OWA operators.