Given two-data databases, record linkage algorithms try to establish which records of these files contain information on the same individual. Standard record linkage algorithms assume that both files are described using the same attributes. In this article, we study the nonstandard case when
the attributes are not the same. We apply aggregation operators for extracting relevant information for this purpose. We restrict to the case of numerical databases.
This article gives a new approach to aggregating assuming that there is an indistinguishability operator or similarity defined on the universe of discourse. The very simple idea is that when we want to aggregate two values a and b we are looking for a value l that is as similar to a as to b or, in a more logical language, the degrees of equivalence of l with a and b must coincide. Interesting aggregation operators on the unit interval are obtained from natural indistinguishability operators associated to t-norms that are ordinal sums.
This article gives a new approach to aggregating assuming that there is an indistinguishability operator or similarity defined on the universe of discourse. The very simple idea is that when we want to aggregate two values a and b we are looking for a value l that is as similar to a as to b or, in a more logical language, the degrees of equivalence of l with a and b must coincide. Interesting aggregation operators on the unit interval are obtained from natural indistinguishability
operators associated to t-norms that are ordinal sums.
Moreno, A.; Cortes, U.; Sales, T. International journal of intelligent systems Vol. 15, num. 3, p. 197-215 DOI: 10.1002/(SICI)1098-111X(200003)15:3<197::AID-INT4>3.0.CO;2-L Data de publicació: 2000-03 Article en revista
Doxastic logics have been widely used to model the reasoning processes that a rational agent may perform on its beliefs. The possible worlds model and the Kripke semantics provide an intuitive semantics to doxastic formulas, but they may only be used to model ideal agents. A different model is proposed in this article. In this new approach, the consistent and complete standard possible worlds are replaced by conceivable situations, that are those scenarios that the modelled agent is capable of considering, regardless of their partiality or inconsistency. A general class of non-ideal reasoners, the rational inquirers, is defined, and it is shown how the evolution of the beliefs of this kind of agents, caused by a multidimensional dynamic analysis, may be modelled in the framework of conceivable situations.
Lascio, L.; Gisolfi, A.; Cortes, U. International journal of intelligent systems Vol. 14, num. 10, p. 981-993 DOI: 10.1002/(SICI)1098-111X(199910)14:10<981::AID-INT3>3.0.CO;2-B Data de publicació: 1999-10 Article en revista
In this paper a modification of the generalized modus ponens is presented, namely, rule: if X is bB then Y is cC; fact: X is aB, conclusion: Y is dC where a, b, c, e, and d are linguistic hedges, and B, C are fuzzy sets. The procedure that allows one to evaluate the modifier d is very simple and gives results given in Refs. 15, 18, 26, and 27. Our approach is algebraic-based and realizes Zadeh's calculus on words by means of Chang's MY algebra.
The dimension of a fuzzy equivalence relation is the minimum number of fuzzy sets needed to generate it. A general theorem is proved that characterizes unidimensional fuzzy equivalence relations. The multidimensional case is also studied under some restrictive conditions (regular fuzzy equivalence relations).
The paper discusses the suitability of T-transitivity for modelling approximate equality. It explains how T-transitivity, a property suitable for modelling vagueness, can still be used to handle the imprecision inherent in approximate equality. Further remarks on resemblance relations are made to clarify how far they improve traditional fuzzy equivalence relations.
In this article a new approach to the formalization of inductive inference in terms of non-monotonic inference is proposed. Induction is characterized as closed-world reasoning from the available data, followed by an inductive jump, which consists in assuming that valid conclusions in the database (assuming closed-world) hold also in the rest of the world. This conception of induction results is adequate to characterize those inference processes that could be formalized, that is, those based in analytical procedures of pattern-matching or regularity detection in the available data. The proposed characterization formally describes the implicit deductive processes of induction and its non-monotonic nature, and could he used as an abstract model of the mental process that leads to obtaining inductive hypotheses. This proposal reduces the problem of induction automatization to that of deduction automatization. Also, it constitutes a formal framework that covers several inductive inference methods used in machine learning. Besides it formalizes inductive definitions, which are very common in science and computer science.