<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-0601">
  <Title>Challenges for annotating images for sense disambiguation</Title>
  <Section position="6" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> This work draws attention to the need for considering natural language semantics in multi-modal settings. Annotating image senses adds increased complexity compared to word-sense annotation in text due to factors such as image properties, subjective perception, and annotator domainknowledge. Moreover, the concept of related senses as well as iconographic senses go beyond and diversify the notion of word sense. In the future, we would like to perform experimentation with human subjects to explore both similarity judgements for image pairs or groups, as well as issues in interannotator agreement for image disambiguation, and, finally, to better understand the role of iconography for semantic interpretation.</Paragraph>
  </Section>
class="xml-element"></Paper>