<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-2011">
  <Title>Combining labelled and unlabelled data: a case study on Fisher kernels and transductive inference for biological entity recognition</Title>
  <Section position="9" start_page="5" end_page="5" type="concl">
    <SectionTitle>
8 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper, we presented a comparison between two state-of-the-art methods to combine labelled and unlabelled data: Fisher kernels and transductive inference. Our experimental results suggest that both method are able to yield a sizeable improvement in performance. For example transductive learning yields performance similar to inductive learning with only about a quarter of the data. These results are very encouraging for tasks where annotation is costly while unannotated data is easy to obtain, like our task of biological entity recognition. In addition, it provides a way to benet from the availability of large electronic databases in order to automatically extract knowledge.</Paragraph>
  </Section>
class="xml-element"></Paper>