<?xml version="1.0" standalone="yes"?>
<Paper uid="A94-1017">
  <Title>Real-Time Spoken Language Translation Using Associative Processors</Title>
  <Section position="5" start_page="105" end_page="105" type="metho">
    <SectionTitle>
4 Related works
</SectionTitle>
    <Paragraph position="0"> Up to now, some systems using a massively parallel machine in the field of natural language processing, such as a parsing system (Kitano and Higuchi, 1991b) and translation systems, e.g., Dm-SNAP (Kitano et al., 1991), ASTRAL (Kitano and Higuchi, 1991a), MBT3n (Sato, 1993), have been proposed. They have demonstrated good performance; nonetheless, they differ from our proposal. For the first three systems, their domain is much smaller than our domain and they do not perform structural disambiguation or target word selection based on the semantic distance between an input expression and each example. For the last system, it translates technical terms i.e. noun phrases, but not sentences.</Paragraph>
  </Section>
class="xml-element"></Paper>