<?xml version="1.0" standalone="yes"?>
<Paper uid="P91-1022">
  <Title>ALIGNING SENTENCES IN PARALLEL CORPORA</Title>
  <Section position="7" start_page="174" end_page="175" type="evalu">
    <SectionTitle>
RESULTS
</SectionTitle>
    <Paragraph position="0"> We aplflied the alignment algorithm of Sect.ions 3 and 4 to the Ca.na.dian Hansa.rd data described in Section 2. The job ran for l0 clays on au IBM Model 3090 mainframe under an operating system that permitted access to 16 mega.bytes of virtual memory. The most probable alignment contained 2,869,041 el-beads. Some of our colleagues helped us  And love and kisses to you, too.</Paragraph>
    <Paragraph position="1"> ... mugwumps who sit on the fence with their mugs on one side and their wumps on the other side and do not know which side to come down on.</Paragraph>
    <Paragraph position="2"> At first reading, she may have.</Paragraph>
    <Paragraph position="3"> Pareillelnent.</Paragraph>
    <Paragraph position="4"> ... en voulant m&amp;lager la ch~vre et le choux ils n'arrivent 1)as k prendre patti.</Paragraph>
    <Paragraph position="5"> Elle semble en effet avoir un grief tout a fait valable, du moins au premier abord.</Paragraph>
    <Paragraph position="6"> Table 4: Unusual but correct alignments examine a random sample of 1000 of these beads, and we found 6 in which sentences were not translations of one another. This is consistent with the expected error rate ol 0.9% mentioned above. In some cases, the algorithm correctly aligns sentences with very different lengths. Table 4 shows some interesting examples of this.</Paragraph>
  </Section>
class="xml-element"></Paper>