<?xml version="1.0" standalone="yes"?>
<Paper uid="C00-1064">
  <Title>Structural Feature Selection For English-Korean Statistical Machine Translation</Title>
  <Section position="3" start_page="439" end_page="439" type="intro">
    <SectionTitle>
2 Motivation
</SectionTitle>
    <Paragraph position="0"> In order to devise parameters for statistical modeling of translation, we started our research from the IBM model which has bee:: widely used by :nany researches. The IBM model is represented with the</Paragraph>
    <Paragraph position="2"> Here, n is the fertility probability that an English word generates n h'end: words, t is tim aligmnent probability that the English word c generates the French word f, and d is the distortion probability that an English word in a certain t)osition will generate a lh'ench word in a certain 1)osition. This formula is Olm of many ways in which p(f, ale ) can tie writtm.</Paragraph>
    <Paragraph position="3"> as the product of a series of conditional prot)at)ilities.</Paragraph>
    <Paragraph position="4"> In above model, the distortion probability is re-lated with positional preference(word order). Since Korean is a free order language, the probability is not t~asible in English-Korean translation.</Paragraph>
    <Paragraph position="5"> Furthermore, the difference between two languages leads to the discordance between words that the one-to-one correst)ondence between words generally does not keel). The n:odel (1), however, as-sumed that an English word cat: be connected with multiple French words, but that each French word is connected to exactly one English word inch:ding the empty word. hl conclusion, many-to-:nany :nap-pings are not allowed in this model.</Paragraph>
    <Paragraph position="6"> According to our ext)eri:nent, inany-to-nmny mappings exceed 40% in English and Korean lexical aligninents. Only 25.1% of then: can be explained by word for word correspondences. It means that we need a statistical model which can lmndle phrasal mat) pings.</Paragraph>
    <Paragraph position="7"> In the case of the phrasal mappings, a lot of parameters should be searched eve:: if we restrict the length of word strings. Moreover, in order to prop-erly estimate t)arameters we need much larger voI-ume of bilingual aligned text than it in word-for-word modeling. Even though such a large corpora exist sometimes, they do not come up with the lex-ical alignments.</Paragraph>
    <Paragraph position="8"> For this problem, we here consider syntactic features which are importmlt in determining structures. A structural feature means here a mapt)ing between tag sequences in bilingual parallel sentences.</Paragraph>
    <Paragraph position="9"> If we are concerned with tag sequence alignments, it is possible to estimate statistical t)armneters in a relatively small size of corpora. As a result, we can remarkably reduce the problem space for possible lexical alignments, a sort of t probability in (1), which improve the complexity of a statistical machine translation model.</Paragraph>
    <Paragraph position="10"> If there are similarities between corresponding tag sequences in two language, tile structural features would be easily computed or recognized. However, a tag sequence in English can be often translated into a completely different tag sequence in Korean as follows.</Paragraph>
    <Paragraph position="12"> It nmans that similarities of tag features between two languages are not; kept all the time and it is necessaw to get the most likely tag sequence mappings that reflect structural correspondences between two languages.</Paragraph>
    <Paragraph position="13"> In this paper, the tag sequence mappings are obtaind by automatic feature selection based on the maximum entropy model.</Paragraph>
  </Section>
class="xml-element"></Paper>