<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-0127">
  <Title>I : I I I I I</Title>
  <Section position="5" start_page="25130" end_page="25130" type="concl">
    <SectionTitle>
6. Conclusions
</SectionTitle>
    <Paragraph position="0"> In this paper we show a new method for Chinese words classification.</Paragraph>
    <Paragraph position="1"> But it can be applied in multiple language too. It integrates top-down and bottom-up idea in word classification. Thus top-down splitting techniques can learn from bottom-up idea's strong points to offset its obvious weakness and keep the advantage of itself. Especially, unlike other classification methods, this method takes the context-sensitive information which most classification methods do not consider into account and make it reflect the properties of natural language more clearly. Moreover, the probabilities are assigned to the words to demonstrate how well a word belongs to classes. This property is very useful in word class-based language modeling used in speech recognition, for it allows the system to have several powerful candidates to be matched during recognition.</Paragraph>
    <Paragraph position="2"> It, however, is important to consider the limitations of the method.</Paragraph>
    <Paragraph position="3"> The computational cost fs very high. The algorithm's complexity is cubic when we move one word from one class to another. Also, the probabilities the word assign to each class is not global optimal. It reflects the degree of a word belonging to classes approximately. And excessive or insufficient classification may occur because the class number is fixed artificially.</Paragraph>
  </Section>
class="xml-element"></Paper>