<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-3206">
  <Title>Scaling Web-based Acquisition of Entailment Relations</Title>
  <Section position="12" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> We have described a scalable Web-based approach for entailment relation acquisition which requires only a standard phrasal lexicon as input. This minimal level of input is much simpler than required by earlier web-based approaches, while succeeding to maintain good performance. This result shows that it is possible to identify useful anchor sets in a fully unsupervised manner. The acquired templates demonstrate a broad range of semantic relations varying from synonymy to more complicated entailment. These templates go beyond trivial paraphrases, demonstrating the generality and viability of the presented approach.</Paragraph>
    <Paragraph position="1"> From our current experiments we can expect to learn about 5 relations per lexicon entry, at least for the more frequent entries. Moreover, looking at the extended test, we can extrapolate a notably larger yield by broadening the search space. Together with the fact that we expect to find entailment relations for about 85% of a lexicon, it is a significant step towards scalability, indicating that we will be able to extract a large scale KB for a large scale lexicon.</Paragraph>
    <Paragraph position="2"> In future work we aim to improve the yield by increasing the size of the sample-corpus in a qualitative way, as well as precision, using statistical methods such as supervised learning for better anchor set identification and cross-correlation between different pivots. We also plan to support noun phrases as input, in addition to verb phrases. Finally, we would like to extend the learning task to discover the correct entailment direction between acquired templates, completing the knowledge required by practical applications.</Paragraph>
    <Paragraph position="3"> Like (Lin and Pantel, 2001), learning the context for which entailment relations are valid is beyond the scope of this paper. As stated, we learn entailment relations holding for some, but not necessarily all, contexts. In future work we also plan to find the valid contexts for entailment relations.</Paragraph>
  </Section>
class="xml-element"></Paper>