Roy George

Fuzzy Database Modeling

Some recent fuzzy database modeling advances for the non-traditional applications are introduced in this book. The focus is on database models for modeling complex information and uncertainty at the conceptual, logical, physical design levels and from integrity constraints defined on the fuzzy relations. The database models addressed here are; the conceptual data models, including the ExIFO and ExIFO2 data models, the logical database models, including the extended NF2 database model, fuzzy object-oriented database model, and the fuzzy deductive object-oriented database model. Integrity constraints are defined on the fuzzy relations are also addressed. A continuing reason for the limited adoption of fuzzy database systems has been performance. There have been few efforts at defining physical structures that accomodate fuzzy information. A new access structure and data organization for fuzzy information is introduced in this book.

A variable-length genetic algorithm for clustering and classification

Pattern clustering and classification can be viewed as a search for, and labeling of a set of inherent clusters in any given data set. This approach can be divided broadly into two types namely supervised and unsupervised clustering. Motivated by human perception and Kohonens method, we present a novel method of supervised clustering and classification using genetic algorithms. Clusters in the pattern space can be approximated by ellipses or sets of ellipses in two dimensions and ellipsoids in general, and the search for clusters can be approximated as the search for ellipsoids or sets of ellipsoids. By assigning fuzzy membership values to points in the pattern space a fuzzy ellipsoid is obtained. The process of thresholding which follows can be thought of as warping the contour of the ellipse to include and exclude certain points in pattern space and in effect producing an arbitrarily shaped cluster. Here we examine the use of genetic algorithms in generating fuzzy ellipsoids for learning the separation of the classes. Our evaluation function drives the genetic search towards the smallest ellipsoid or set of ellipsoids, which maximizes the number of correctly classified examples, and minimizes the number of misclassified examples.

Uncertainty management issues in the object-oriented data model

This paper fully develops a previous approach by George et al. (1993) to modeling uncertainty in class hierarchies. The model utilizes fuzzy logic to generalize equality to similarity which permitted impreciseness in data to be represented by uncertainty in classification. In this paper, the data model is formally defined and a nonredundancy preserving primitive operator, the merge, is described. It is proven that nonredundancy is always preserved in the model. An object algebra is proposed, and transformations that preserve query equality are discussed.

Modelling class hierarchies in the fuzzy object-oriented data model

Abstract In this paper we describe a fuzzy logic based approach to modelling uncertainty in class hierarchies. It is shown that the traditional view of class hierarchies is subsumed in this model as a special case. The problem of multiple inheritance in class hierarchies is discussed and analyzed. The membership value derivations in the inheritance hierarchy reflects the degree of fuzziness existing in the data values and the semantics of the situation being modelled. Thus a more realistic modelling of the universe of discourse is possible through this approach. This model is compatible with existing object-oriented data models.

Design and implementation issues in the fuzzy object-oriented data model

Abstract Uncertainty modeling and data manipulation are desirable in object-oriented database systems to handle complex objects with imprecise properties. In this paper, we present an enhancement to the fuzzy object-oriented data model (FOOD) (George et al., Fuzzy Sets and Systems 60 (3) (1993) 259–272), that permits a truer representation of different types of uncertainty. Uncertainty is extended to the class. efinition level and is the basis for the determination of the membership of an object in a class. We describe a software architecture for the implementation of the model and discuss the significant details of a prototype implemented using the EXODUS storage manager (ESM).

Uncertainty Modeling In Object-Oriented Geographical Information Systems

The object-oriented data model has been utilized for geographical information applications, chiefly because of the rich set of modelling primitives it offers. Despite its representational capabilities, it still falls short in describing data typically seen in geographic information systems, i. e., imprecise, spatial or continuous valued data. In this paper, an extension to the object-oriented data model that permits the representation of imprecise data is discussed in the context of a soil information system. It is shown that this extension accommodates the querying and manipulation of spatial and continous valued data.

Computer security and ethics awareness in university environments: a challenge for management of information systems

There are great challenges for the people involved in the management of information systems to take special measures to secure their information systems. This is more apparent in a university setting, specifically when providing computing resources to the diverse student population. The current paper reports on a study of computer security and ethics awareness among students taking an introductory computer technology course at Historically Black Colleges and Universities (HBCU). Four hundred sixty five (465) student volunteers attending introductory computer technology courses participated in the survey. A concise questionnaire consisting of 21 questions was used. The questionnaire had two major parts. The first part measured the participants awareness of computer security. The second part measured their awareness of ethical computer use. In general, the majority of the participants had a satisfactory awareness of computer security and ethics. However, as the results of the survey reveal, a 20% to 52% lack of awareness of computer security is highly risky and a 14% to 24% violation of the code of ethics is not desirable. This suggests security and ethic awareness training is needed for university users. To help students at HBCUs increase their awareness of security and ethical issues, the authors provided several recommendations.

eWellness: Building a Smart Hospital by Leveraging RFID Networks

The application of radio frequency identification (RFID) sensor networks pervasively to the hospital environment has the potential to improve efficiencies in operational aspects through accurate capture of fine grained data. The sensor nodes generate large amounts of real or near real-time data. Understanding and managing these data in the context of the hospital pose challenges for the middleware which has to connect sensor networks with hospital applications. In this paper we propose an edgeware paradigm of organizing hospital RFID sensor data based on the notion of an event. The event is defined and its properties are discussed. A three layered event representation and reasoning model is developed to map raw sensory information to semantic events in the application domains. A prototype system, called eWellness is build for the test of the local hospital

Mining Weather Data Using Fuzzy Cluster Analysis

The need to analyze the vast quantities of weather data collected has led to the development of new data mining tools and techniques. Mining this data can produce new insights into weather, climatological and environmental trends that have both scientific and practical significance. This chapter discusses the challenges posed by weather databases and examines the use of fuzzy clustering for analyzing such data. It proposes the extension of the fuzzy K-Means clustering algorithm to account for the spatio-temporal nature of weather data. It introduces an unsupervised fuzzy clustering algorithm, based on the fuzzy KMeans and defines a cluster validity index which is used to determine an optimal number of clusters. These techniques are validated on weather data in the South Central US, and global climate data (sea level pressure). It is seen that the algorithm is able to identify and preserve interesting phenomena in the weather data.

Fuzzy Cluster Analysis of Spatio-Temporal Data

The quantities of earth science data collected have necessitated the development of new data mining tools and techniques. Mining this data can produce new insights into weather, climatological and environmental trends that have significance both scientifically and practically. This paper discusses the challenges posed by earth science databases and examines the use of fuzzy K-Means clustering for analyzing such data. It proposes the extension of the fuzzy K-Means clustering algorithm to account for the spatio-temporal nature of such data. The paper introduces an unsupervised fuzzy clustering algorithm, based on the fuzzy K-Means and defines a cluster validity index which is used to determine an optimal number of clusters. It is shown experimentally that the algorithms are able to identify and preserve regions of meteorological interest