CBR techniques can be used for flexible database querying purposes. More specifically, CBR techniques can be used for instance-based prediction with which unknown data values can be approximated. Hereby, four main processes can be identified: case description, case comparison, prediction, and revision.
Published in Chapter:
Flexible Querying Techniques Based on CBR
Guy De Tré (Ghent University, Belgium), Marysa Demoor (Ghent University, Belgium), Bert Callens (Ghent University, Belgium), and Lise Gosseye (Ghent University, Belgium)
Copyright: © 2008
|Pages: 24
DOI: 10.4018/978-1-59904-853-6.ch007
Abstract
In case-based reasoning (CBR), a new untreated case is compared to cases that have been treated earlier, after which data from the similar cases (if found) are used to predict the corresponding unknown data values for the new case. Because case comparisons will seldom result in an exact-similarity matching of cases and the conventional CBR approaches do not efficiently deal with such imperfections, more advanced approaches that adequately cope with these imperfections can help to enhance CBR. Moreover, CBR in its turn can be used to enhance flexible querying. In this chapter, we describe how fuzzy set theory can be used to model a gradation in similarity of the cases and how the inevitable uncertainty that occurs when predictions are made can be handled using possibility theory resulting in what we call flexible CBR. Furthermore, we present how and under which conditions flexible CBR can be used to enhance flexible querying of regular databases.