<?xml version="1.0" standalone="yes"?>
<Paper uid="C96-2178">
  <Title>Machine Translation Method Using Inductive Learning with Genetic Algorithms</Title>
  <Section position="4" start_page="1021" end_page="1021" type="metho">
    <SectionTitle>
1. Initial population
</SectionTitle>
    <Paragraph position="0"> The system selects the translation rules which can be applied to the source sentence. The set of selected translation rules is called the initial population.</Paragraph>
    <Paragraph position="1"> 2. Determination of fitness value The system calculates the fitness value of the translation rules by the fitness function (1).</Paragraph>
  </Section>
  <Section position="5" start_page="1021" end_page="1021" type="metho">
    <SectionTitle>
3. Selection process
</SectionTitle>
    <Paragraph position="0"> The method of the selection process was described in the section on feedback process.</Paragraph>
  </Section>
  <Section position="6" start_page="1021" end_page="1021" type="metho">
    <SectionTitle>
4, Crossover
</SectionTitle>
    <Paragraph position="0"> The method of crossover was described in the section on learning process.</Paragraph>
  </Section>
  <Section position="7" start_page="1021" end_page="1021" type="metho">
    <SectionTitle>
5. Mutation
</SectionTitle>
    <Paragraph position="0"> The method of mutation was described in the section on learning process.</Paragraph>
  </Section>
  <Section position="8" start_page="1021" end_page="1021" type="metho">
    <SectionTitle>
6. Evaluation of population
</SectionTitle>
    <Paragraph position="0"> The system substitutes the words in the word translation rules for the variables in the sentence translation rules. A translation rule includes a Japanese sentence or words corresponding to an English sentence or words. The system produces a Japanese sentence for the English sentence when the English sentence has the same character string as the source sentence. The Japanese sentence which is produced is the translation result, Figure 4 shows an example of how the translation result is produced.</Paragraph>
    <Paragraph position="1"> The system selects the correct translation result according to two criteria when there are several candidates of translation results: one criterion is the translation rule which has a higher fitness value and the other is the translation rule which is more similar to the source sentence.</Paragraph>
  </Section>
  <Section position="9" start_page="1021" end_page="1022" type="metho">
    <SectionTitle>
3 Experiments for Performance
</SectionTitle>
    <Paragraph position="0"/>
    <Section position="1" start_page="1021" end_page="1021" type="sub_section">
      <SectionTitle>
Evaluation
3.1 Method of Evaluation
</SectionTitle>
      <Paragraph position="0"> The effective translation results are grouped into two categories: (1)The translation result has the same character string as the proofread translation result.</Paragraph>
      <Paragraph position="1"> (2)The translation result has the same structure as the proofread translation result.</Paragraph>
      <Paragraph position="2"> This means that the proofread translation result has the same character string as the translation result with substituted nouns or adjectives for the variables.</Paragraph>
      <Paragraph position="3"> The ineffective translation results are grouped into three categories: (3)The translation result has a different character string than the proofread translation result without unregistered words.</Paragraph>
      <Paragraph position="4"> (4)The translation result has a different character string than the proofread translation result with unregistered words.</Paragraph>
      <Paragraph position="5"> (5)A failed translation.</Paragraph>
      <Paragraph position="6"> The system ranks ten candidates of translation results for the user. The method for determining optimal translation results was described in Sub-section 2.5.</Paragraph>
    </Section>
    <Section position="2" start_page="1021" end_page="1022" type="sub_section">
      <SectionTitle>
3.2 Method of Experiments
</SectionTitle>
      <Paragraph position="0"> In the experiments, 1,810 translation examples were used as data, of which 1,010 examples were taken from a textbook (Hasegawa et al., 1991) for first grade junior high school students, and  800 examples fl'om another textbook (Ota et al., 1991) for first grade junior high school students in Japan. All of these translation examples were processed by the method outlined in Figure 1.</Paragraph>
      <Paragraph position="1"> First, 1,010 translation examples were used for the learning process, and 800 translation exanlples were used for evaluation of the translation. Experiments were carried out with and without genetic algorithms. In the experiments without genetic algorithms, crossover, mutation and the selection process were not performed.</Paragraph>
    </Section>
    <Section position="3" start_page="1022" end_page="1022" type="sub_section">
      <SectionTitle>
3.3 Results of Experiments
</SectionTitle>
      <Paragraph position="0"> The accuracy rate of translation increased fronl 52.8% to 61.9% by applying genetic algorithms.</Paragraph>
      <Paragraph position="1"> Table 1 shows the results of experiments using genetic algorithms. In this table, (1) ~ (5) correspond to (1) ~ (5) in Subsection 3.1. Table 2 shows examples of translation results using genetic algorithms.</Paragraph>
    </Section>
    <Section position="4" start_page="1022" end_page="1022" type="sub_section">
      <SectionTitle>
3.4 Discussion
</SectionTitle>
      <Paragraph position="0"> In the experiments without genetic algorithms, high quality translation results coukl not be obtained due to the requirement of a very large amount of translation examples which are similar to other translatiml examples. Therefore, we applied genetic algorithms to a method of machine translation using inductive learning to automatically produce ncw translation examples which are similar to other translation examples. By using genetic algorithms, the accuracy rate of translation increased from 52.8% to 61.9%.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>