<?xml version="1.0" standalone="yes"?>
<Paper uid="C92-1055">
  <Title>Syntactic Ambiguity Resolution Using A Discrimination and Robustness Oriented Adaptive Learning Algorithm</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1. Introduction
</SectionTitle>
    <Paragraph position="0"> Ambiguity resolution has long been the focus in natural language processing. Many rule-based approaches have been proposed in the past, However, when applying such approaches to large scale applications, they usually fail to offer satisfactory pertbrmance. As a huge amount of fine-grained knowledge is required to solve the ambiguity problem, it is quite difficult for rule-based approach to acquire the huge and fine-grained knowledge, and maintain consistency among them by human \[Su 90a\].</Paragraph>
    <Paragraph position="1"> Probabilistic approaches attack these problems by providing a more objective measure on the preference to a given interpretation. Then, these approaches acquire huge and fine grained knowledge, or parameters in statistic terms from the corpus automatically. The uncertainty problem in linguistic phenomena is resolved on a more solid basis if a probabilistic approach is adopted. Moreover, the knowledge acquired by the statistical method is always consistent because the knowledge is acquired by jointly considering all the data in the corpus at the same time. Hence, the time for knowledge acquisition and the cost to maintain consistency are significantly reduced by adopting those probabilistic approaches.</Paragraph>
    <Paragraph position="2"> To resolve the problems resulting from syntactic ambiguities, a unified statistical approach for ambiguity resolution has been proposed by Su \[Su 88, 92b\]. In that approach, all knowledge sources, including lexical, syntactic and semantic knowledge, are encoded by a unifiedprobabilistic score function with a uniform formulation. This uniform probabilistic score function has been successfully applied in spoken language processing \[Su 90b, 91b, 92a\] and machine translation systems \[Chen 91\] to integrate different knowledge sources for ambiguity resolution.</Paragraph>
    <Paragraph position="3"> In implementing this unified probabilistic score function, values of score functions are estimated from the data in the training corpus. However, due to the problem of insufficiency of training data and incompleteness of model knowledge, the statistical variations between the training corpus and the real application are usually not covered by this approach.</Paragraph>
    <Paragraph position="4"> Therefore, the performance in the testing set sometimes gets poor in the real application.</Paragraph>
    <Paragraph position="5"> To enhance the capability of discrimination and robustness of those proposed score function, a discrimination-oriented adaptive learning is proposed in this paper. And then, the robustness of this proposed adaptive learning procedure is enhanced by enlarging the margin between the correct candidate and its confusing candidates to achieve maximum separation between different candidates.</Paragraph>
    <Paragraph position="6"> Since the implementation of this adaptive learning procedure is based on the uniform probabilistic score function, we will first briefly review the unified probabilistic score function. Readers who are ACTES DE COLING-92, NANTas, 23-28 AO~r 1992 3 5 2 PROC. OF COLING-92, NANTES, AuG, 23-28. 1992 interested in the details about the uniform probabilistic score function please refer \[Chen 91, Su 91b, 92a, 92b\].</Paragraph>
  </Section>
class="xml-element"></Paper>