<?xml version="1.0" standalone="yes"?>
<Paper uid="N06-2011">
  <Title>Spectral Clustering for Example Based Machine Translation</Title>
  <Section position="8" start_page="43" end_page="43" type="concl">
    <SectionTitle>
5 Conclusions and future work
</SectionTitle>
    <Paragraph position="0"> From the experimental results we see that spectral clustering leads to relatively purer and more intuitive clusters. These clusters result in an improved BLEU score in comparison with the clusters obtained through GAC. GAC can only collect clusters in convex regions in the term vector space, while spectral clustering is not limited in this regard. The ability of spectral clustering to represent non-convex shapes arises due to the projection onto the eigenvectors as described in (Ng. et. al., 2001).</Paragraph>
    <Paragraph position="1"> As future work, we would like to analyze the variation in performance as the amount of data increases. It is widely known that increasing the amount of training data in a generalized EBMT system eventually leads to saturation of performance, where all clustering methods perform about as well as baseline. Thus, all methods have an operating region where they are the most useful. We would like to locate and extend this region for spectral clustering. null Also, it would be interesting to compare the clusters obtained with spectral clustering and the Part of Speech tags of the words in the same cluster, especially for languages such as English where good taggers are available.</Paragraph>
    <Paragraph position="2"> Finally, an important direction of research is in automatically selecting the number of clusters for the clustering algorithm. To do this, we could use information from the eigenvalues or the distribution of points in the clusters.</Paragraph>
  </Section>
class="xml-element"></Paper>