<?xml version="1.0" standalone="yes"?>
<Paper uid="P92-1034">
  <Title>USING CLASSIFICATION TO GENERATE TEXT</Title>
  <Section position="5" start_page="269" end_page="270" type="ackno">
    <SectionTitle>
Acknowledgements
</SectionTitle>
    <Paragraph position="0"> The IDAS project is partially funded by UK SERC grant GR/F/36750 and UK DTI grant IED 4/1/1072, and we are grateful to SERC and DTI for their support of this work. We would also like  to thank the IDAS industrial collaborators -- Inference Europe, Ltd.; lgacal Instruments, Ltd.; and Racal Researdh Ltd. -- for all the help they have given us in performing this research.</Paragraph>
    <Paragraph position="1"> Appendix: A Comparison of Classification and Unification FUG is only one of a number of grammar formalisms based on feature logics. The logic underlying FUG is relatively simple, but much more expressive logics are now being implemented \[Emele and Zajac, 1990; D6rre and Seiffert, 1991; D/Srre and Eisele, 1991\]. Here we provide an initial formal characterisation of the relation between classification and unification, but abstracting away from the differences between the different unification systems.</Paragraph>
    <Paragraph position="2"> Crucial to all approaches in unification-based generation (or parsing) is the idea that at every level an input description (i.e. logical form or similar) 7 is combined with a set of axioms (type specifications, grammar functional descriptions, rules) and the resulting logical expression is then reduced to a normal form that can be used straightforwardly to construct the set of models for the combined axioms and description.</Paragraph>
    <Paragraph position="3"> Classification is an appropriate operation to use in normal form construction when the axioms take the form oq ~ fit, with ~ interpreted as logical implication, and where each ai and/~i is a term in a feature logic. If the input description is 'complete' with respect to the conditions of these axioms (that is, if 7 ^ ai ~ J- exactly when 7 _C ~i, where _ is subsumption), then it follows that for every model A4:</Paragraph>
    <Paragraph position="5"> (the relationship is more complex if the grammar is reeursive, though the same basic principle holds). The first step of the computation of the models of 7 and the axioms then just needs quick access to {fli17 _Coti}. The classification approach is to have the different ai ordered in a subsumption taxonomy. An input description 7 is placed in this taxonomy and the fll corresponding to its ancestors are collected.</Paragraph>
    <Paragraph position="6"> Input descriptions are 'complete' if every input description is fully specified as regards the conditions that will be tested on it. This implies a rigid distinction between 'input' and 'output' information which, in particular, means that classification will not be able to implement bidirectional grammars. If all the axioms are of the above form, input descriptions are complete and conjunctive, and the fli's are conjunctive (as is the case in IDAS) then there will always only be a single model.</Paragraph>
    <Paragraph position="7"> The above assumption about the form of axioms is clearly very restrictive compared to what is allowed in many modern unification formalisms.</Paragraph>
    <Paragraph position="8"> In IDAS, the notation is restricted even further by requiring the c~i and /~i to be purely conjunctive. In spite of these restrictions, the system is still in some respects more expressive than the simpler unification formalisms. In Definite Clause Grammars (DCGs) \[Pereira and Warren, 1980\], for instance, it is not possible to specify al --&amp;quot;/~1 and also c~z --*/~, whilst allowing that (al AOC/2) ~ (~1A~2) (unless aland as are related by subsumption) \[Mellish, 19911.</Paragraph>
    <Paragraph position="9"> The comparison between unification and classification is, unfortunately, made more complex when default inheritance is allowed in the classification system (as it is in IDAS). Partly, the use of defaults may be viewed formally as simply a mechanism to make it easier to specify 'complete' input descriptions. The extent to which defaults are used in an essential way in IDAS still remains to be investigated. Certainly for the grammar writer the ability to specify defaults is very valuable, and this has been widely acknowledged in grammar frameworks and implementations.</Paragraph>
  </Section>
class="xml-element"></Paper>