Association rule mining is a very popular data mining technique. Rules in this technique are often used to identify and represent dependencies between attributes in databases. Specifically, fuzzy association rules are rules that use the concepts of fuzzy sets and can be considered as a special case of fuzzy predicates. Many quality measures have been defined for fuzzy association rules, but all consider a specific structure: antecedent and consequence. In the case of fuzzy predicates in the normal form (i.e., conjunctive or disjunctive), it is necessary to define different quality measures that do not consider the structure as an antecedent or a consequence. The only available measure for this scenario is the fuzzy predicate truth value (FPTV), which has serious limitations. The evaluation of fuzzy predicates in the normal form through appropriate quality measures has not yet been clearly defined in the literature. Thus, we propose several quality measures specifically for fuzzy predicates in the conjunctive (CNF) and disjunctive (DNF) normal forms. Experimental studies illustrate the use of the proposed measures and allow some general conclusions about each measure.Keywords: data mining, fuzzy predicate, quality measures, conjunctive and disjunctive normal forms. RESUMENLa extracción de las reglas de asociación es una técnica de minería de datos muy popular, las cuales son utilizadas a menudo para identificar y representar dependencias entre atributos en bases de datos. Específicamente, las reglas de asociación difusas utilizan conceptos de conjuntos difusos y pueden ser vistas como un caso especial de predicados difusos. Muchas medidas de calidad han sido definidas para reglas de asociación difusa, pero todas consideran la estructura específica de reglas: antecedente y consecuente.En el caso general de predicados difusos en forma normal (conjuntiva o disyuntiva), es necesario definir diferentes medidas de calidad que no estén en función de antecedente y consecuente, puesto que la única medida disponible para ello, es el valor de verdad para predicados difusos (FPTV) y tiene serias limitaciones. La evaluación de un predicado difuso en forma normal, a través de medidas adecuadas de calidad no ha sido todavía claramente definida por otros autores. Por esa razón, en este trabajo se proponen varias medidas de calidad para los predicados difusos, en formas normal conjuntiva o disyuntiva. Los experimentos demuestran el uso que se le puede dar a las métricas propuestas y permiten llegar a conclusiones generales de cada una de ellas.Palabras clave: minería de datos, predicados difusos, medidas de calidad, forma normal conjuntiva y disyuntiva.

The collection of methods known as "data mining" offers methodological and technical solutions to deal with the analysis of medical data and the construction of models. Medical data have a special status based upon their applicability to all people; their urgency (including life-or death); and a moral obligation to be used for beneficial purposes. Due to this reality, this article addresses the special features of data mining with medical data. Specifically, we will apply a recent data mining algorithm called FuzzyPred. It performs an unsupervised learning process to obtain a set of fuzzy predicates in a normal form, specifically conjunctive (CNF) and disjunctive normal form (DNF). Experimental studies in known medical datasets shows some examples of knowledge that can be obtained by using this method. Several kind of knowledge that was obtained by FuzzyPred in these databases cannot be obtained by other popular data mining techniques.

No abstract

No abstract

No abstract