Skip to main content

1997 | Buch

Uncertainty Management in Information Systems

From Needs to Solutions

herausgegeben von: Amihai Motro, Philippe Smets

Verlag: Springer US

insite
SUCHEN

Über dieses Buch

As its title suggests, "Uncertainty Management in Information Systems" is a book about how information systems can be made to manage information permeated with uncertainty. This subject is at the intersection of two areas of knowledge: information systems is an area that concentrates on the design of practical systems that can store and retrieve information; uncertainty modeling is an area in artificial intelligence concerned with accurate representation of uncertain information and with inference and decision-making under conditions infused with uncertainty. New applications of information systems require stronger capabilities in the area of uncertainty management. Our hope is that lasting interaction between these two areas would facilitate a new generation of information systems that will be capable of servicing these applications. Although there are researchers in information systems who have addressed themselves to issues of uncertainty, as well as researchers in uncertainty modeling who have considered the pragmatic demands and constraints of information systems, to a large extent there has been only limited interaction between these two areas. As the subtitle, "From Needs to Solutions," indicates, this book presents view­ points of information systems experts on the needs that challenge the uncer­ tainty capabilities of present information systems, and it provides a forum to researchers in uncertainty modeling to describe models and systems that can address these needs.

Inhaltsverzeichnis

Frontmatter
1. Introduction
Abstract
As its title suggests, “Uncertainty Management in Information Systems” is a book about how information systems can be made to manage information permeated with uncertainty. This subject is at the intersection of two areas of knowledge: information systems is an area that concentrates on the design of practical systems that can store and retrieve information; uncertainty modeling is an area in artificial intelligence concerned with accurate representation of uncertain information and with inference and decision-making under conditions infused with uncertainty.
Amihai Motro
2. Sources of Uncertainty, Imprecision, and Inconsistency in Information Systems
Abstract
An information system is a computer model of the real world. Like any other model, it captures an abstracted version of the real world, using a level of abstraction that is implied by the expected applications. As with any other model, the most important consideration is the integrity of the model; i.e., the accuracy of the representation. Unfortunately, our knowledge of the real world is often imperfect, thus challenging our ability to create and maintain information systems of integrity.
Amihai Motro
3. Imperfect Information in Relational Databases
Abstract
Databases model the real world at two levels: the database schema specifies the structure of the database, and the database extension represents a set of facts or events of the real world. The Appropriate Scheme Assumption [10] expresses that this two-level description is a valid view of the world. Such a distinction between schema and extension is often blurred in artificial intelligence.
Esteban Zimányi, Alain Pirotte
4. Uncertainty in Intelligent Databases
Abstract
By an intelligent database we mean a traditional relational database with additional functionalities to represent either (1) general rules, as in deductive databases, or (2) some kind of incomplete information, like marked null values or disjunctive facts, or (3) additional meta-information, like information validity, uncertainty factors, or some kind of modality.
Robert Demolombe
5. Uncertain, Incomplete, and Inconsistent Data in Scientific and Statistical Databases
Abstract
This chapter is a survey of several issues and applications in uncertain, inconsistent, and incomplete data in scientific and statistical databases (SSDBs).
Stephen Kwan, Frank Olken, Doron Rotem
6. Knowledge Discovery and Acquisition from Imperfect Information
Abstract
This chapter discusses the issues of imperfect information in the fields of knowledge discovery in databases (KDD) and knowledge acquisition for expert systems (KA) . My perspective on these issues is more of a practitioner motivated by pressing application needs and less of the researcher motivated by the desire to push the frontiers of science.
Gregory Piatetsky-Shapiro
7. Uncertainty in Information Retrieval Systems
Abstract
Information retrieval (IR) is concerned with identifying documents in a collection that a user in need of information will judge to be useful or relevant. We are generally provided with a description of the user’s information need, or query. We then select documents likely to be relevant by comparing the query with prestored document representations, and may, optionally, revise the query representation for subsequent retrieval. The retrieval process, then, involves query acquisition, document selection, and query revision. A document collection may include normal text documents (e.g., journal articles), but it may also include nontext materials (photographs, museum pieces, software modules, and so on). Information storage and retrieval is a central element of systems that support such functions as office automation, library automation, legal research, or software engineering. Research in IR spans many subdisciplines of computer and information science.
Howard R. Turtle, W. Bruce Croft
8. Imperfect Information: Imprecision and Uncertainty
Abstract
Imperfection, be it imprecision or uncertainty, pervades real-world scenarios and must therefore be incorporated into every information system that attempts to provide a complete and accurate model of the real world. Yet, this is hardly achieved in today’s information systems. A major reason might be the inherent difficulty of understanding the various aspects of imprecision and uncertainty.
Philippe Smets
9. Probabilistic and Bayesian Representations of Uncertainty in Information Systems: A Pragmatic Introduction
Abstract
A great deal has been written about the underlying principles of alternative methods of representing uncertainty—not the least about probability and Bayesian methods. While we cannot entirely resist discussing basic principles, we will focus on the pragmatic issues, which too often get lost under the mass of philosophy and mathematics. We will address such questions as: How can we use probability to represent the various types of uncertainty? How can we quantify these uncertainties? How much effort is necessary to do so? How can we obtain the greatest benefits from representing uncertainty while minimizing the effort? There are a variety of reasons to represent uncertainty and a variety of probabilistic and Bayesian ways to do so, requiring varying amounts of effort. We discuss an approach to resolve these issues, so that the costs will be commensurate with the benefits.
Max Henrion, Henri J. Suermondt, David E. Heckerman
10. An Introduction to the Fuzzy Set and Possibility Theory-Based Treatment of Flexible Queries and Uncertain or Imprecise Databases
Abstract
The information to be stored in databases is not always precise and certain, and, occasionally, some information might be missing altogether.1 When the available information is imperfect, it is often desirable to try to represent it in the database nonetheless, so that it can be used to answer queries of interest as much as possible. A related issue is the handling of imperfect or flexible queries. For example, a natural query language may use a word or a phrase whose meaning is vague or even entirely unclear. As another example, a query may reflect a user’s uncertainty about what he is looking for. In addition, one may want to use vague predicates in a query to express pbl]References among the admissible answers.
Patric Bosc, Henri Prade
11. Logical Handling of Inconsistent and Default Information
Abstract
The subjects of this chapter are two important and related kinds of uncertainty in information systems: inconsistent information and default (defeasible) information. In many information system applications, there is a need to represent and reason with inconsistent data. For example, in a tax collection agency, database records on individual taxpayers should be allowed to have inconsistent information, as such information could be used to direct enquiries by tax inspectors. Default information , such as rules that are usually true but are allowed to have exceptions, tends to reduce the size of databases significantly, yet without significant loss of utility for many applications. For example, a market research agency could use default information in its consumer profiles: for its kind of business such a level of accuracy could be deemed sufficient.
Philippe Besnard, Luis Fariñas del Cerro, Dov Gabbay, Anthony Hunter
12. The Transferable Belief Model for Belief Representation
Abstract
As shown in Chapter 8, there are different forms of imperfect data, be they uncertain or imprecise. Models have been proposed for each form, but modeling combined forms of imperfect data has hardly been achieved. It would nevertheless seem useful to have a single model that could represent several forms of uncertainty. Possibly, this could be achieved by simulating what might be the human approach to such problems.
Philippe Smets, Rudolf Kruse
13. Approximate Reasoning Systems: Handling Uncertainty and Imprecision in Information Systems
Abstract
In this chapter we provide a personal perspective of the progress made in the field of approximate reasoning systems and their relevance and applicability to information systems. Within the scope of this chapter, the concept of information systems covers the entire range of Databases (DB), Information Retrieval (IR), and Knowledge-Based Systems (KBS).
Piero P. Bonissone
14. On the Classification of Uncertainty Techniques in Relation to the Application Needs
Abstract
Computerized information systems are subject to a variety of imperfections and uncertainty; either in the data themselves or in the query designed to access items of information from the system. Meanwhile, a great deal of research work is available on techniques for handling uncertainty. It is clear that these techniques can be exploited for the management of uncertainty in information systems. However, the plethora of available techniques poses a challenge to the designers of information systems as to which technique is best suited to the problem to be managed. This dilemma is familiar to the designers of artificial intelligence (AI) systems. Many system builders have expressed a wish to see research results that would provide a comprehensible classification matrix showing the suitability of each uncertainty techniques to some abstract features of the problem needing uncertainty management. This chapter argues that such a classification is not possible given the present state of research in uncertainty and indeed it is not meaningful to attempt to produce a matrix of the type suggested.
E. H. Mamdani
15. A Bibliography on Uncertainty Management in Information Systems
Abstract
This is an evolving bibliography of documents on uncertainty and imprecision in information systems. By uncertainty and imprecision, we mean the representation of and query support for information that is fuzzy, unknown, partially known, vague, uncertain, probabilistic, indefinite, disjunctive, possible, maybe, incomplete, approximate, erroneous, or imprecise. Currently, the bibliography concentrates almost exclusively on database and knowledge-base systems, with few bl]References on other kinds of information systems.
Curtis E. Dyreson
Backmatter
Metadaten
Titel
Uncertainty Management in Information Systems
herausgegeben von
Amihai Motro
Philippe Smets
Copyright-Jahr
1997
Verlag
Springer US
Electronic ISBN
978-1-4615-6245-0
Print ISBN
978-1-4613-7865-5
DOI
https://doi.org/10.1007/978-1-4615-6245-0