Modelling the competence of case-bases

Lecture notes in computer sciencePublished 1 January 1998

Barry Smyth, Elizabeth McKenna

DOI10.1007/bfb0056334

Source URL

Citations129

SJR quartileQ2

SJR score0.35

SNIP0.55

TL;DR

It is argued that this model has an important role to play in areas such as the evaluation and benchmarking of case-based techniques, and a novel application of the model is demonstrated as a guide to case-base designers during the case authoring process.

Abstract

The competence of a case-based system (the range of problems it can solve) depends critically on the cases in the case-base. However, the precise relationship between cases and overall competence is a complex one. For example, some cases can be critical to competence, while others may be largely redundant. In this paper we present, and evaluate, a new model of case competence. We argue that this model has an important role to play in areas such as the evaluation and benchmarking of case-based techniques, and we demonstrate a novel application of the model as a guide to case-base designers during the case authoring process.

Keywords

Computer Science

Lecture notes in computer scienceRefining conversational case libraries

142 Citations1997David W. Aha, Leonard A. Breslow

This work describes an approach for revising case libraries according to design guidelines, its implementation in Clire, and empirical results showing that, under some conditions, this approach can improve conversational CBR performance.

Lecture notes in computer scienceMaintaining unstructured case bases

91 Citations1997Kirsti Racine, Qiang Yang

This paper focuses on two difficult problems in case-base maintenance: redundancy and inconsistency detection, and presents methods and a system for maintaining large and unstructured case bases.

Lecture notes in computer scienceThe utility problem analysed

87 Citations1996Barry Smyth, Pádraig Cunningham

The utility problem is discussed from a case-based reasoning perspective, examining its root causes in an experimental CBR system and suggesting ways to improve coverage and quality and improve system efficiency.

Expert SystemsTowards a methodology for evaluating expert systems

85 Citations1990Alun Preece

The paper concludes that any evaluation methodology should incorporate both types of method, and presents prescriptions and an outline for such a methodology, which combines the methods for maximum effectiveness.

Lecture notes in computer scienceApplying case retrieval nets to diagnostic tasks in technical domains

40 Citations1996Mario Lenz, Hans-Dieter Burkhard +1 more

Objectdirected Case Retrieval Nets is presented, a memory model developed for an application of Case-Based Reasoning to the task of technical diagnosis to store cases and enhance this network with an object model encoding knowledge about the devices in the application domain.

Expert Systems with ApplicationsVerification and validation of case-based systems

38 Citations1993Daniel E. O’Leary

New statistical and structural approaches are presented designed to exploit unique aspects of case-based reasoning for verification purposes and develop new approaches to generating comparative solutions for validation purposes.

The Utility Problem in Case-Based Reasoning

32 Citations1993Anthony Francis, Ashwin Ram

The objective of the survey was to determine the presenting diagnoses of the patients who presented to the Emergency Department of four tertiary care teaching hospitals in Karachi i.e. the Aga Khan University Hospital, Liaquat National Hospital, Jinnah Postgraduate Medical Centre and Civil Hospital Karachi.

Lecture notes in computer scienceA criterion of comparison between two case bases

14 Citations1995Jean Lieber

The definition of case-based reasoning given in this paper stresses the fact that the goal of the similarity measure is to give an o priori estimation of the performance of the adaptation process.