<?xml version="1.0" standalone="yes"?>
<Paper uid="C94-1027">
  <Title>PART-OF-SPEECH TAGGING WITH NEURAL NETWORKS</Title>
  <Section position="7" start_page="775" end_page="775" type="concl">
    <SectionTitle>
7 CONCLUSIONS
</SectionTitle>
    <Paragraph position="0"> In this paper, the Net-Tagger was presented, a part-of-speech tagger which is based on a MLP-network.</Paragraph>
    <Paragraph position="1"> A comparison of the tagging results with those of a trigram tagger and a IIMM tagger showed that the accuracy is as high as that of the trigram tagger and the robustness on small training corpora is as good as that of the HMM tagger. Thus, the Net-Tagger combines advantages of both of these methods.</Paragraph>
    <Paragraph position="2"> The Net-Tagger has the additional advantage that problematic decisions between tags are easy to detect, aDue to the large training times needed to train the 3-layernetwork, no further tests have been conducted.</Paragraph>
    <Paragraph position="3"> o Less than 60 % of the tagging errors were made in common by both taggers.</Paragraph>
    <Paragraph position="4"> so that in these cases an additional tag can be given in the output. In this way, the final decision can be delayed to a later processing stage, e.g. a parser.</Paragraph>
    <Paragraph position="5"> A disadvantage of the presented method may be its lower processing speed compared to statistical methods. In the light of the high speed of present computer hardware, however, this does not seem to be a serious drawback.</Paragraph>
  </Section>
class="xml-element"></Paper>