<?xml version="1.0" standalone="yes"?>
<Paper uid="W03-0415">
  <Title>Using LSA and Noun Coordination Information to Improve the Precision and Recall of Automatic Hyponymy Extraction</Title>
  <Section position="8" start_page="0" end_page="0" type="concl">
    <SectionTitle>
7 Conclusion and Further Work
</SectionTitle>
    <Paragraph position="0"> The results presented in this paper demonstrate that the application of linguistic information from automaticallylearned mathematical models can significantly enhance both the precision and the recall of pattern-based hyponymy extraction techniques. Using a graph model of noun similarity we were able to obtain an almost five-fold improvement in recall, though the precision of this technique is clearly affected by the correctness of the &amp;quot;seed-relationships&amp;quot; used. Using LSA filtering we eliminated spurious relations extracted by the original pattern method, reducing errors by 30%. Such filtering also eliminated spurious relations learned using the graph model that were the result of lexical ambiguity and of seed hyponymy relations inappropriate for the technique, reducing errors by 33%.</Paragraph>
    <Paragraph position="1"> This paper suggests many possibilities for future work.</Paragraph>
    <Paragraph position="2"> First of all, it would be interesting to apply LSA to a system for building an entire hypernym-labelled ontology in roughly the way described in (Caraballo, 1999), perhaps by using an LSA-weighted voting method to determine which hypernym would be used to label each node. We are considering how to extend our techniques to such a task.</Paragraph>
    <Paragraph position="3"> Also, systematic comparison of the lexicosyntactic patterns used for extraction to determine the relative productiveness and accuracy of each pattern might prove illuminating, as would comparison across different corpora to determine the impact of the topic area and medium/format of documents on the effectiveness of hyponymy extraction. Ultimately, the ability to predict a priori how well a knowledge-extraction system will work on a previously unseen corpus will be crucial to its usefulness. null Applying the techniques of this paper to a system that used mutual bootstrapping (Riloff and Jones, 1999) to find additional extraction patterns would also be interesting (such an approach is suggested in (Hearst, 1998)). And of course, further refinement of the mathematical models we use and our methods of learning them, including more sophisticated use of available tools for linguistic pre-processing, such as the identification and indexing of multiword expressions, could further improve the precision and recall of hyponymy extraction techniques.</Paragraph>
  </Section>
class="xml-element"></Paper>