<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-3815">
  <Title>Context Comparison as a Minimum Cost Flow Problem</Title>
  <Section position="8" start_page="103" end_page="103" type="concl">
    <SectionTitle>
7 Conclusions
</SectionTitle>
    <Paragraph position="0"> We have given an overview of our network flow formalism which seamlessly combines distributional and ontological information. Given a suitable ontology, a context vector of word frequencies can be transformed into a context profile--a frequency distribution over the concepts in the ontology. In contrast to traditional non-graphical approaches to measuring only the distributional distance between context vectors, we provide a graphical formalism which incorporates both the semantic distance of the component nodes as well as the distributional differences between the context profiles. By taking advantage of the graphical structure of an ontology, our method allows a systematic and meaningful way of abstracting over words in a context, and by extension, a meaningful way of comparing contexts.</Paragraph>
    <Paragraph position="1"> One concern with our method in its pre-transformation form is its inability to incorporate sophisticated concept-to-concept semantic distances efficiently. To remedy this, we propose a novel technique that mimics the structure of the more computationally intensive network. Our preliminary evaluation shows that the transformation does not hamper the method's ability to make fine-grained semantic distinctions, and the computational complexity is drastically reduced as well. Generally, our network flow method presents a highly competitive alternative to a purely distributional and non-graphical approach. null In our on-going work, we are further exploring how the choice of junction influences the performance of different types of concept-to-concept semantic distances. For example, would a bottom-up junction selection approach (from the profile nodes instead of from the root level) result in better performance? In addition, we intend to examine the graphical properties of the individual profiles as well as the routes between the concepts across profiles selected by our network flow methods. Such analyses will help us gain insight into the strengths (and weaknesses) of taking advantage of a graphical representation of contexts as well as treating an ontology as a metric space for context comparisons.</Paragraph>
  </Section>
class="xml-element"></Paper>