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.
Autori: | Ferretti Paola (Corresponding) | |
Data di pubblicazione: | 2021 | |
Titolo: | On Fuzzy Confirmation Measures of Fuzzy Association Rules | |
Titolo del libro: | Progresses in Artificial Intelligence and Neural Systems. Smart Innovation, Systems and Technologies | |
Digital Object Identifier (DOI): | http://dx.doi.org/10.1007/978-981-15-5093-5_30 | |
Appare nelle tipologie: | 3.1 Articolo su libro |
File in questo prodotto:
File | Descrizione | Tipologia | Licenza | |
---|---|---|---|---|
preprint Celotto_Ellero_Ferretti On Fuzzy Confirmation Measures of Fuzzy Association Rules.pdf | Documento in Pre-print | Accesso chiuso-personale | Riservato |