| Home > Publications > Reports > Informatics (CW) |
CW 263
J. Ramon and M. Bruynooghe
A framework for defining distances between first-order logic objects
Abstract
Several learning systems, such as systems based on clustering and instance based learning, use a measure of distance between objects. Good measures of distance exist when objects are described by a fixed set of attributes as in attribute value learners. More recent learning systems however, use a first order logic representation. These systems represent objects as models or clauses. This paper develops a general framework for distances between such objects and reports a preliminary evaluation.
report.pdf / mailto: J. Ramon