Catálogo de publicaciones - libros

The article focusses on propositional logics with semantics based on rough sets. Many approaches to rough sets (including generalizations) have come to the fore since the inception of the theory, and resulted in different “rough logics” as well. The essential idea behind these logics is, quite naturally, to interpret well-formed formulae as rough sets in (generalized) approximation spaces. The syntax, in most cases, consists of modal operators along with the standard Boolean connectives, in order to reflect the concepts of lower and upper approximations. Non-Boolean operators make appearances in some cases too.

Information systems (“complete” and “incomplete”) have always been the “practical” source for approximation spaces. Characterization theorems have established that a rough set semantics based on these “induced” spaces, is no different from the one mentioned above. We also outline some other logics related to rough sets, e.g. logics of information systems – which, in particular, feature expressions corresponding to attributes in their language. These systems address various issues, such as the temporal aspect of information, multiagent systems, rough relations.

An attempt is made here to place this gamut of work, spread over the last 20 years, in one platform. We present the various relationships that emerge and indicate questions that surface.

We introduce the intuitionistic rough set and intuitionistic rough relational and object oriented database models. The intuitionistic rough set database models draw benefits from both the rough set and intuitionistic techniques, providing greater management of uncertainty for databases applications in a less than certain world. We provide the foundation for the integration of intuitionistic rough sets into modeling of uncertainty in databases. This builds upon some of our previous research [2,3] with integrating fuzzy and rough set techniques for uncertainty management in databases.

In this paper we present results of experiments conducted to compare three types of missing attribute values: lost values, ”do not care” conditions and attribute-concept values. For our experiments we selected six well known data sets. For every data set we created 30 new data sets replacing specified values by three different types of missing attribute values, starting from 10%, ending with 100%, with increment of 10%. For all concepts of every data set concept lower and upper approximations were computed. Error rates were evaluated using ten-fold cross validation. Overall, interpreting missing attribute values as lost provides the best result for most incomplete data sets.

This paper reviews, rather non-technically, Pawlak’s approach to vagueness through rough sets and looks for a foundation of rough sets in an early work of Obtułowicz. An extension of Obtułowicz’s proposal is suggested that in turn, hints at a unified approach to rough sets and fuzzy sets.

In this article, we compare mappings of Pawlak’s lower and upper approximations of concepts with those proposed by Skowron and Stepaniuk. It is known that both approaches coincide for the standard rough inclusion, so we consider the case of an arbitrary rough inclusion function. Even if the approximation space investigated is based on an arbitrary non-empty binary relation, the lower approximation mappings are equal in both approaches. Nevertheless, the upper approximation mappings are different in general.

Data preparation is a very important but also a time consuming part of a Data Mining process. In this paper we describe a hierarchical method of text classification based on regular expressions. We use the presented method in our data mining system during a pre-processing stage to transform Latin free-text medical reports into a decision table. Such decision tables are used as an input for rough sets based rule induction subsystem. In this study we also compare accuracy and scalability of our method with a standard approach based on dictionary phrases.

The problem considered in this article is how does one go about discovering and designing intelligent systems. The solution to this problem is considered in the context of what is known as wisdom technology (wistech), an important computing and reasoning paradigm for intelligent systems. A rough-granular approach to wistech is proposed for developing one of its possible foundations. The proposed approach is, in a sense, the result of the evolution of computation models developed in the Rasiowa–Pawlak school. We also present a long-term program for implementation of what is known as a wisdom engine. The program is defined in the framework of cooperation of many Research & Development (R & D) institutions and is based on a wistech network (WN) organization.

This research study presents rough set-based decision systems applications to the acoustical domain. Two areas are reviewed for this purpose, namely music information classification and retrieval and noise control. The main aim of this paper is to show results of both measurements of the acoustic climate and a survey on noise threat, conducted in schools and students’ music clubs. The measurements of the acoustic climate employ multimedia noise monitoring system engineered at the Multimedia Systems Department of the Gdansk University of Technology. Physiological effects of noise exposure are measured using pure tone audiometry and otoacoustic emission tests. All data are gathered in decision tables in order to explore the significance of attributes related to hearing loss occurence and subjective factors that attribute to the noise annoyance. Future direction of experiments are shortly outlined in Summary.

Manually evaluating important and interesting rules generated from data is generally infeasible due to the large number of rules extracted. Different approaches such as rule interestingness measures and rule quality measures have been proposed and explored previously to extract interesting and high quality association rules and classification rules. Rough sets theory was originally presented as an approach to approximate concepts under uncertainty. In this paper, we explore rough sets based rule evaluation approaches in knowledge discovery. We demonstrate rule evaluation approaches through a real-world geriatric care data set from Dalhousie Medical School. Rough set based rule evaluation approaches can be used in a straightforward way to rank the importance of the rules. One interesting system developed along these lies in HYRIS (HYbrid Rough sets Intelligent System). We introduce HYRIS through a case study on survival analysis using the geriatric care data set.

This article is dedicated to the creative genius Zdzislaw Pawlak for his contribution to the theoretical development of science and technology in China. His distinguished discovery of Rough Set Theory is a formal theory which is well suited for uncertainty computing to analyze imprecise, uncertain or incomplete information of data. Inspired by his work scientists and engineers in China has developed many theories and applications in various science and technology fields. For instance, J.H.Dai studied the theories of Rough Algebras and Axiom Problem of Rough 3-Valued Algebras [1, 2]. G.L.Liu studied the Rough Sets over Fuzzy Lattices [3, 4]. D.W.Pei studied the Generalized Model of Fuzzy Rough Sets [5]. W.Z.Wu Studied the On Random Rough Sets [6]. D.Q.Miao studied the Rough Group and Their Properties [7]. These are part of their recent research results related to rough set theory. As a matter of fact, there are still many researchers working in the field of rough sets in China, who have proposed many creative results for last few years. These results are not listed one by one in this short commemorative article. We will try to review all the “Rough Set” researchers and their research results in the appeared next article.