<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-1113">
  <Title>Variants of tree similarity in a Question Answering task</Title>
  <Section position="9" start_page="106" end_page="106" type="concl">
    <SectionTitle>
7 Conclusion and Future Work
</SectionTitle>
    <Paragraph position="0"> For two different parsers, and two different question-answering tasks, we have shown that improved parse quality leads to better performance, and that a tree-distance measure out-performs a sequence distance measure. We have focussed on intrinsic, syntactic properties of parse-trees. It is not realistic to expect that exclusively using tree-distance measures in this rather pure way will give state-of-the-art question-answering performance, but the contribution of this paper is the (start of an) exporation of the syntactic parameters which effect the use of tree-distance in question answering. More work needs to be done in systematically varying the parsers, question-answering tasks, and parametrisations of tree-distance over all the possibilities. null There are many possibilities to be explored involving adapting cost functions to enriched node descriptions. Already mentioned above, is the possibility to involve semantic information in the cost functions. Another avenue is introducing weightings based on corpus-derived statistics, essentially making the distance comparision refer to extrinsic factors. One open question is whether analogously to idf, cost functions for (non-lexical) nodes should depend on tree-bank frequencies.</Paragraph>
    <Paragraph position="1"> Another question needing further exploration is the dependency-vs-constituency contrast. Interestingly Punyakonok(2004) themselves speculate: each node in a tree represents only a word in the sentence; we believe that appropriately combining nodes into meaningful phrases may allow our approach to perform better.</Paragraph>
    <Paragraph position="2"> We found working with constituency trees that it was the sub-traversal distance measure that performed best, and it needs to be seen whether this holds also for dependency trees. Also to be explored is the role of structural weighting in a system using dependency trees.</Paragraph>
    <Paragraph position="3"> A final speculation that it would be interesting to explore is whether one can use feed-back from performance on a QATD task as a driver in the machine-learning of probabilities for a parser, in an approach analogous to the use of the language-model in parser training.</Paragraph>
  </Section>
class="xml-element"></Paper>