<?xml version="1.0" standalone="yes"?>
<Paper uid="C88-2155">
  <Title>Machine Translation for Monolinguals</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1. Background
</SectionTitle>
    <Paragraph position="0"> Under the Alvey Directorate's research programme in natural language processing, an English-Japanese machine translation project was carried out at the Centre for Computational Linguistics, University of Manchester Institute of Science and Technology and the Centre for Japanese Studies, University of Sheffield. The project ran from October 1984 to December 1987, and was also funded by International Computers Limited (ICL). The UMIST group, led by Peter Whitelock, have developed an English-to-Japanese prototype (Ntran), while Jiri Jelinek's group at Sheffield have been working from Japanese into English (Aidtrans).</Paragraph>
    <Paragraph position="1"> The two prototypes, although very different in some aspects of their linguistic and computational approaches, share an important and distinctive design philosophy. Both are interactive, and, unlike present commercially available machine (aided) translation systems, both are designed for end use by a monolingual speaker of English. This paper will discuss the means by which each system achieves this, and the issues involved in tailoring a system for use by a target language or source language speaker.</Paragraph>
  </Section>
class="xml-element"></Paper>