Antonio Gisolfi

A fuzzy-based methodology for the analysis of diabetic neuropathy

A new model for the fuzzy-based analysis of diabetic neuropathy is illustrated, whose pathogenesis so far is not well known. The underlying algebraic structure is a commutative l-monoid, whose support is a set of classifications based on the concept of linguistic variable introduced by Zadeh. The analysis is carried out by means of patients anagraphical and clinical data, e.g. age, sex, duration of the disease, insulinic needs, severity of diabetes, possible presence of complications. The results obtained by us are identical with medical diagnoses. Moreover, analyzing suitable relevance factors one gets reasonable information about the etiology of the disease, our results agree with most credited clinical hypotheses.

Medical Differential Diagnosis through Type-2 Fuzzy Sets

In this paper we present a model of differential medical diagnosis, based on type 2 fuzzy sets. Given the patient history case and a set of suspected-inherent pathologies, the method creates a set of linguistically labelled clusters of the pathologies. The pathologies present in each cluster are ordered through a similarity function. In the case of lack of information, our method indicates the elements needed to have a more precise diagnosis and it can control its same accuracy

A new model for linguistic modifiers

Abstract This paper presents a new model for the representation of linguistic modifiers. Each modifiers is characterized by three parameters and is identified by a single positive real number n. The number n fully characterizes the generic term of a boolean linguistic variable. The representation of the terms maintains the order relation with which they are used in natural language. Lastly, it is shown how this new model allows the fuzzy set of each boolean expression to be evaluated through suitably simple fuzzy functions.

A complete, flexible fuzzy-based approach to the classification problem

Abstract We present an algebraic structure as a complete methodology of classification. Utilizing this structure, we apply an algebraic approximation to the problem of generating appropriate clusters of objects characterized by fuzzy attributes. More precisely, the values of the attributes are expressed in terms of linguistic labels, and thus are handled as fuzzy numbers. This opens new possibilities in all those fields for which the need to describe the population under analysis by means of more natural terms becomes crucial. In fact, in these cases, the application of resolution strategies based on the adoption of “standard” methods, such as a distance matrix, appears as a brutal effort to adapt quantitative methods to qualitative problems.

Linguistic hedges and the generalized modus ponens

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 Zadehs calculus on words by means of Changs MV algebra. ©1999 John Wiley & Sons, Inc.

A Fuzzy-Based Approach to Stereotype Selection in Hypermedia

This paper presents a stereotype-based user model for adaptive hypermedia systems. We use a suitable algebraic fuzzy structure which can adequately reflect some features of the user in the model, and apply this model to adapt the navigation and the content of hypermedia nodes to the users needs. The model includes temporal representations of the user and approximates every real user by a set of stereotypes, from which the one realizing the best approximation can always be extracted. The set of stereotypes is the support set of the structure and the operations defined therein may be supplied with adequate semantics which allow for the selection of the stereotype.

An algebraic approximation to the classification with fuzzy attributes

Abstract This paper proposes an algebraic approximation to the problem of generating appropriate classifications in a universe of discourse in which the characteristics of the elements are defined through attributes whose values are linguistic labels. Classifications are generated by means of operations between ordered strings induced by the fuzzy attributes. Moreover, a heuristic measure is used, based on the concept of statistical variance, in order to recognize the potential relevance of each attribute with respect to the generated classifications.

An algebraic fuzzy structure for approximate reasoning

Abstract The concept of C-set and of the related C-calculus was introduced to study some standard problems of signal processing. Now, it is shown how the theory of fuzzy sets can lead to the C-fsets and to the C-fcalculus. Afterwards, we focus our attention on the theory of approximate reasoning and we show how C-fcalculus can be regarded, in some cases, as a possible alternative to the theory of approximate reasoning.

Uncertainty processing in user-modeling activity

In this paper we investigate the role of a new algebraic fuzzy structure as dynamic model to represent users states during navigation in a multimedia tutoring system. The structure can be profitably utilized to monitor the evolution of mental states in order to suit the resources and the didactic strategies to the users needs. The problem of building a user model is favourably affected by our approach: in fact the huge quantities of data present in overlay and bug catalogue models are no longer required. The fuzzy structure allows to monitor the cognitive variables that are necessary to design a user model that can operate in a way independent from the particular topic to be taught.

Classifying through an algebraic fuzzy structure: The relevance of the attributes

An appropriate fuzzy structure was previously defined to generate appropriate classifications in a universe of discourse in which the characteristics of the elements are defined through attributes whose values are linguistic labels. the way the structure operates depends upon suitable operations on fuzzy attributes represented by fuzzy numbers. However, this classification method suffers from a main disadvantage, i.e., the inherent rigidity deriving from the circumstance that all the attributes share the same weight and, consequently, are taken into account completely disregarding the fact that usually some attributes are more relevant than others. In this article we analyze a suitable weighting rule allowing us to treat in a different way attributes whose relevance is different. the validity of the approach was checked by means of a case study concerning diets for breast‐feeding women.