<?xml version="1.0" standalone="yes"?>
<Paper uid="P92-1011">
  <Title>COMPARING TWO GRAMMAR-BASED GENERATION A CASE STUDY</Title>
  <Section position="9" start_page="85" end_page="86" type="concl">
    <SectionTitle>
6. CONCLUSION
</SectionTitle>
    <Paragraph position="0"> In this paper we discussed several aspects of two natural language generation algorithms: SHDGA and EAA.</Paragraph>
    <Paragraph position="1"> Both algorithms operate under the same general set of conditions, that is, given a grammar, and a structured representation of meaning, they attempt to produce one or more corresponding surface strings, and do so with a minimal possible effort. We analyzed the performance of each algorithm in a few specific situations, and concluded that EAA is both more general and more efficient algorithm than SHDGA. Where EAA enforces the optimal traversal of the derivation tree by precomputing all possible orderings for nonterminal expansion, SHDGA can be guaranteed to display a compa- null rable performance only if its grammar is appropriately designed, and the semantic heads are carefully assigned (manually). With other grammars SHDGA will follow non-optimal generation paths which may lead to extreme inefficiency.</Paragraph>
    <Paragraph position="2"> In addition, EAA is a truly multi-directional algorithm, while SHDGA is not, which is a simple consequence of the restricted form of grammar that SHDGA can safely accept.</Paragraph>
    <Paragraph position="3"> This comparison can be broadened in several directions. For example, an interesting problem that remains to be worked out is a formal characterization of the grammars for which each of the two generation algorithms is guaranteed to produce a finite and/or optimal search tree. Moreover, while we showed that SHDGA will work properly only on a subset of EAA's grammars, there may be legitimate g~ that neither algorithm can handle.</Paragraph>
  </Section>
class="xml-element"></Paper>