Wouter Gelade

  1. Learning Deterministic Regular Expressions for the Inference of Schemas from XML Data.

    Authors: Geert Jan Bex, Wouter Gelade, Frank Neven, Stijn Vansummeren
    Subjects: Databases
    Abstract

    Inferring an appropriate DTD or XML Schema Definition (XSD) for a given
    collection of XML documents essentially reduces to learning deterministic
    regular expressions from sets of positive example words. Unfortunately, there
    is no algorithm capable of learning the complete class of deterministic regular
    expressions from positive examples only, as we will show. The regular
    expressions occurring in practical DTDs and XSDs, however, are such that every
    alphabet symbol occurs only a small number of times.

Syndicate content