<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-0710">
  <Title>Sentence extraction as a classification task</Title>
  <Section position="8" start_page="291" end_page="291" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> We have rephcated Kuplec et al's experiment for automatic sentence extraction using several rodependent heurmtlcs and superwsed learning The summaries for our documents were not written by professional abstractors, but by the authors themselves As a result, our data demonstrated conmderably lower overlap between sentences m the summary and sentences m the mare text We used an alternative evaluation that mL~ed ahgned sentences with other good can&amp;dates for extraction, as identified by a human judge We obtained a 68 4% recall and preclmon on our text material, compared to a 28 0% baseline and a best mdlvldual method of 55 2% Combimng m&amp;vldually weaker methods results m an increase of around 20% of the best method, m line with Kupmc et al's results Thin shows the ~mefulness. of Kuplec et al's methodology for a different type of data and evaluation strategy We found that there was no difference m performance between our evaluation strategies (alignment or human judgement), apart from external constraints on the task hke the compression rate We also show that increased trmmng did not slgmficantly improve the sentence extraction results, and conclude that there m more room for improvement m the extraction methods themselves With respect to our ultimate goal of generatmg of higher quahty abstracts (more coherent, more flexible variable-length abstracts), we argue that the use of human-selected extraction can&amp;dates m ad* Vantageous to the task Our favounte heurmhc includes meta-lmgmstic cue phrases, because they can be used to detect rhetorical structure m the document, and because they provide a rhetoncal context for each extracted sentence m addition to its propomhonal content</Paragraph>
  </Section>
class="xml-element"></Paper>