<?xml version="1.0" standalone="yes"?>
<Paper uid="J90-2001">
  <Title>WIDE-RANGE RESTRUCTURING OF REPRESENTATIONS IN MACHINE INTERMEDIATE TRANSLATION</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 INTRODUCTION
</SectionTitle>
    <Paragraph position="0"> Much effort has been devoted to research into, and development of, machine translation since the 1950s (Slocum 1985). However, the quality of the output sentences produced by most machine translation systems is not high enough to have any marked effect on translation productivity. null A machine translation system produces a variety of expressions in the target language, including good, fair, and poor expressions. In this paper, we define these distinctions as follows. &amp;quot;Good&amp;quot; sentences can be easily understood and have high readability because of their naturalness, &amp;quot;fair&amp;quot; sentences can be understood but their readability is low, and &amp;quot;poor&amp;quot; sentences cannot be understood without referring to the source sentences.</Paragraph>
    <Paragraph position="1"> To improve the quality of translation, the following major functions should be implemented:  1. selection of equivalents for words; 2. reordering of words; and 3. improvement of sentence styles.</Paragraph>
    <Paragraph position="2">  A machine translation system that does not have Function 3 often produces &amp;quot;good&amp;quot; output in the case of translation between languages in the same linguistic group if Functions 1 and 2 are appropriately achieved. However, most output will be &amp;quot;fair&amp;quot; or &amp;quot;poor&amp;quot; in the case of translation between languages in different linguistic groups, because of the stylistic gaps between them. We need to enhance Function 3 as well as Functions 1 and 2 in order to change &amp;quot;fair&amp;quot; or &amp;quot;poor&amp;quot; sentences to &amp;quot;good&amp;quot; or &amp;quot;fair&amp;quot; ones. Note that in this paper, style means a preferable grammatical form that successfully conveys a correct meaning. First, let us consider how to select target language equivalents for words in the source language. An appropriate part of speech for a word is first determined by the grammatical constraints provided by the analysis grammar. Nouns and the verb in a simple sentence can then be appropriately translated according to the combinative constraints between the case frame of the verb and the semantic markers of the nouns. This is a well-known mechanism for practical semantic processing in machine translation.</Paragraph>
    <Paragraph position="3"> However, this way of selecting equivalents has the limitation that we cannot classify verbs and nouns in sufficient detail, because of ambiguities in the definitions and usage of words. Many researchers in machine translation claim that a much more powerful semantic processing mecha-Computational Linguistics Volume 16, Number 2, June 1990 71 Taijiro Tsutsumi Intermediate Representations in Machine Translation nism with a knowledge base is required for this purpose.</Paragraph>
    <Paragraph position="4"> This is a long-range research project in the field.</Paragraph>
    <Paragraph position="5"> Second, there seem to be few critical problems in reordering words if structural transfer is appropriately carried out. In a simple sentence, for example, we can usually reorder words correctly on the basis of the case frame of the verb.</Paragraph>
    <Paragraph position="6"> Thus, the improvement of sentence styles is one of the crucial functions required for further enhancement of the present translation quality. Even when the output is &amp;quot;fair&amp;quot; in quality, it has to be read carefully, because the readability is often low. We expect an improvement in sentence styles to result in an improvement in translation quality from &amp;quot;poor&amp;quot; or &amp;quot;fair&amp;quot; to &amp;quot;good.&amp;quot; Improving sentence styles seems to be easier than selecting better equivalents for words, because there are syntactic clues to help us make the styles more natural.</Paragraph>
    <Paragraph position="7"> This paper focuses on an approach to the improvement of sentence styles by wide-range restructuring of intermediate representations. Wide-range restructuring in this paper means the global restructuring of intermediate representations, usually including the replacement of some class words (i.e., noun, adjective, verb, and adverb). Some papers have mentioned limited restructuring of intermediate representations (Bennett and Slocum 1985; Vauquois and Boitet 1985; Isabelle and Bourbeau 1985; Nagao et al.</Paragraph>
    <Paragraph position="8"> 1985; McCord 1985; Nomura et al. 1986). For example, LMT (McCord 1985) has a restructuring function after the transfer phase, to form a bridge between the basic styles of English and German. The Mu system (Nagao et al.</Paragraph>
    <Paragraph position="9"> 1985) has two specific restructuring functions, before and after the transfer phase, mainly to handle exceptional cases.</Paragraph>
    <Paragraph position="10"> However, few machine translation systems have so far had a comprehensive component for wide-range restructuring (Slocum 1985), mainly because many systems are presently designed to produce output that is at best &amp;quot;fair,&amp;quot; and little effort has been devoted to obtaining &amp;quot;good&amp;quot; output or natural sentences. As a matter of fact, wide-range restructuring functions are usually scattered over the analysis, transfer, and generation phases in an ad hoc way.</Paragraph>
    <Paragraph position="11"> Because of complicated implementations, the systems come to have low maintainability and efficiency.</Paragraph>
    <Paragraph position="12"> Few papers so far have systematically discussed the crucial stylistic gaps between languages, the importance of wide-range restructuring of intermediate representations to bridge these gaps, and effective mechanisms for restructuring. A restructuring mechanism is necessary even when a machine translation system is based on the semantic-level transfer or pivot method (Carbonell et al. 1981).</Paragraph>
    <Paragraph position="13"> In this paper, we first discuss stylistic gaps between languages, and the importance of dealing with them effectively in order to generate natural target sentences. We then propose a restructuring mechanism that successfully bridges the stylistic gaps and preserves a high maintainability for the transfer phase of a machine translation system. Last, we discuss the implementation of the wide-range restructuring function.</Paragraph>
  </Section>
class="xml-element"></Paper>