<?xml version="1.0" standalone="yes"?>
<Paper uid="P02-1021">
  <Title>Semi-Supervised Maximum Entropy Based Approach to Acronym and Abbreviation Normalization in Medical Texts</Title>
  <Section position="9" start_page="8" end_page="8" type="ackno">
    <SectionTitle>
6 Future Work
</SectionTitle>
    <Paragraph position="0"> In the future, I am planning to test the assumption that abbreviations and their expansions occur in similar contexts by testing on hand-labeled data. I also plan to vary the size of the window used for determining the local context from two words on each side of the expression in question as well as the cutoff used during ME training. It will also be necessary to extend this approach to other medical and possibly non-medical domains with larger data sets. Finally, I will experiment with combining the UMLS abbreviations table with the Mayo Clinic specific abbreviations.</Paragraph>
  </Section>
class="xml-element"></Paper>