R. De Caluwe

UFO: uncertainty and fuzziness in an object-oriented model

The authors present a method for coping with the uncertainty and fuzziness that cause problems in handling information. As these two concepts carry different semantic meanings, they distinguish between them by representing fuzzy information as conjunctive fuzzy sets and uncertainty by means of generalized fuzzy sets. An object-oriented methodology is chosen because it offers an excellent framework for handling uncertainty in a way completely transparent to the user. This treatment of uncertainty in a fuzzy object-oriented environment is well-suited to databases; hence examples from this field are selected. Genealogical databases, where uncertain information is very common, and where information is continuously updated, are used as a frame of reference. The proposed model also allows storage of as much information about uncertain objects as possible.<<ETX>>

A perspective from the Fuzzy Object Data Management Group (FODMG)

The Fuzzy Object Data Management Group has been formed as a joint international collaborative research effort among fuzzy database researchers in order to establish common terminology and concepts, to formalize and integrate the current body of fuzzy object model research, to provide a basis for future extensions, and to contribute to the commercial success of a fuzzy object data model. This paper presents the initial research efforts to use ODMG-93 object data model standard as the basis for defining a fuzzy object data model since it is becoming a defacto standard and several object-oriented database vendors are currently releasing commercial products in compliance with this standard. The syntactic extensions to the ODMG object model and the semantic issues related to these extensions in order to provide fuzzy set objects and fuzzy objects are presented. These extensions are an initial effort in accomplishing the challenging task of defining a fuzzy type and fuzzy hierarchy and their related semantics.

Some practical aspects of fuzzy database techniques: an example

Abstract Basically there are two approaches for dealing with incomplete or imprecise information in the framework of (relational) databases. Buckles and Petry introduced fuzzy similarity relations in order to estimate to what extent possible values of an attribute can be regarded as interchangeable. In the approach of Dubois. Prade and Testemale possibility distributions were used to represent all possible kinds of incompletely or fuzzily known values. This paper describes the application of a combination of both techniques. We focus on the very specific domain of criminal investigation, especially on criminal identification by means of a personal description. However, the developed method can be applied to a lot of other domains, where a similar sort of fuzziness and uncertainty shows up. For most attributes the fuzzy values are considered as primitive notions. Each domain is provided with a sup-W-transitive likeness relation in order to represent an existing kind of overlapping between the possible fuzzy values. Moreover, in order to restrict the number of attributes we allow multivalued ones: their values are represented as possibility distributions on the power class of the domain. The pattern-matching process uses a Prade-Testemale-like technique based on possibility distributions.

Assigning membership degrees to points of fuzzy boundaries

There has been a growing interest in the field of geographic information systems (GIS) to model vague and uncertain regions for which information about a given property is available (e.g. soil structure). Vague and uncertain mean in general that some knowledge is available about the position and the circumference of such a region, but that the information is mostly imprecise or lacks certainty. Either this lack of precise data is inherent to the property (e.g. the degree of soil pollution), or it is due to physical limitations in making accurate measurements (e.g. temperature). There has been some research in modeling such regions using fuzzy regions. Fuzzy regions are modeled by means of broad boundaries. A broad boundary is defined as the area enclosed between two non-intersecting crisp edges including these edges, one completely located within the other. Inside and on the inner edge lay the points for which the definition of the property is completely met; outside and on the outer edge lay the points for which the definition of the property is not at all met. Points of the broad boundary meet the definition of the property up to a certain extent. Up to now, mainly the applicability of the traditional set operators on fuzzy regions (intersection, inclusion and union) has been addressed in the literature. Furthermore, almost no contributions treat the degree to which the points in the fuzzy region come up to the given definition of a property. We present two methods to solve the latter problem, both from a theoretical as well as from a practical point of view.

Time management in fuzzy and uncertain object-oriented databases

The core of this chapter is the definition of a general object-oriented framework for the modelling of time and its use in the setting of fuzzy and uncertain object-oriented databases, in order to support the time concepts of transaction time, valid time and user-defined time and hence the traditional characteristics of temporal databases. The definition of this framework itself relies on a suitable, new theoretical time model that deals with the practical use of time indications in natural language.

Semantics and Modelling of Flexible Time Indications

This chapter presents an in-depth discussion of the semantics of time indications and proposes a formal time model suited to a variety of computer applications in which temporal databases play a role. The main characteristics of the model are the flexibility inherent to the ability of expressing approximate time indications in a way very close to natural language and the setting of a hierarchical framework, which allows to take into account the most appropriate level of granularity when expressing time indications.

Integrating Fuzziness in Database Models

In this paper we propose a way of integrating fuzziness and uncertainty in three database models: the relational database model, an extended entity-relationship model and an object-oriented database model. For this, we rely upon the notion of fuzzy truth value, defined as a fuzzy set on {true, false}, and on the corresponding fuzzy logic, of which simplicity is its strength.

A framework for flexible querying and mining of musical audio archives

In this paper the outline of a framework for the handling of flexible querying and mining of musical audio archives is presented. This framework has been designed within the scope of the DEKKMMA-project, which aims to build a digital musical audio archive for the ethnomusicological department of the Belgian Royal Museum for Central Africa. This museum holds one of the largest and world wide most important collections of music from Central Africa. Beside the digitizing and recording aspects, efficient access facilities for both specialized and nonspecialized users are required. In order to achieve this, the framework supports and integrates flexible querying, classification and object comparison mechanisms.

Fuzzy modeling of traffic noise annoyance

This paper presents a fuzzy rule-based model for the prediction of traffic noise annoyance. Several inference schemes are compared for their performance in prediction capabilities as well as in speed. It is shown that the fastest implementation does an equally good job, after optimization of certainty degrees attached to the rules. For this optimization, a genetic algorithm is applied.

Modelling of fuzzy and uncertain spatio-temporal information in databases : a constraint-based approach

A constraint-based generalized object-oriented database model is adapted to manage spatio-temporal information. The presented adaptation is based on the definition of a new data type, which is suited to handle both temporal and spatial information. Generalized constraints are used to describe spatio-temporal data, to enforce integrity rules on databases, to specify the semantics of a database scheme and to impose selection criteria in flexible database querying.