Many researchers from different sciences focused their attention on quantifying the degree to which an antecedent in a rule supports a conclusion. This long-standing problem results to be particularly interesting in the case of fuzzy association rules between a fuzzy antecedent and a fuzzy consequence: in fact, rules become much more flexible in describing information hidden in the data and new interestingness measures can be defined in order to assess their relevance. This implies, in particular, a new definition of support and of confidence of the association rule. In this framework, we focus on fuzzy confirmation measures defined in terms of confidence. In this way, it is possible to propose new fuzzy confirmation measures in a setting that allows their comparison with reference to some potential properties.

On Fuzzy Confirmation Measures of Fuzzy Association Rules

Celotto Emilio;Ellero Andrea
;
Ferretti Paola
2021-01-01

Abstract

Many researchers from different sciences focused their attention on quantifying the degree to which an antecedent in a rule supports a conclusion. This long-standing problem results to be particularly interesting in the case of fuzzy association rules between a fuzzy antecedent and a fuzzy consequence: in fact, rules become much more flexible in describing information hidden in the data and new interestingness measures can be defined in order to assess their relevance. This implies, in particular, a new definition of support and of confidence of the association rule. In this framework, we focus on fuzzy confirmation measures defined in terms of confidence. In this way, it is possible to propose new fuzzy confirmation measures in a setting that allows their comparison with reference to some potential properties.
2021
Progresses in Artificial Intelligence and Neural Systems. Smart Innovation, Systems and Technologies
File in questo prodotto:
File Dimensione Formato  
preprint Celotto_Ellero_Ferretti On Fuzzy Confirmation Measures of Fuzzy Association Rules.pdf

non disponibili

Tipologia: Documento in Pre-print
Licenza: Accesso chiuso-personale
Dimensione 257.41 kB
Formato Adobe PDF
257.41 kB Adobe PDF   Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/3728329
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact