Earticle

초록

영어

The purpose of this paper is to describe a means of extending a unification-based syntactic processor so that it is capable of inferring appropriate internal representations of the syntactic properties of newly encountered words. We begin with a description of the parser in section 2, focusing on those aspects most relevant for the learning procedure. The hypothesis construction part of the procedure is discussed in section 3, the hypothesis manager in section 4. Section 5 discusses the instantiation of the procedure under categorial grammar. Section 6 discusses formal foundations and definitions of the notions presented in the body of the paper

목차

Abstract
 1. Introduction
 2. The Unicorn Parsing System
 3. Encountering an Unknown Lexeme
 4. Hypothesis Management
  4.1. Combining lower bounds
  4.2. Combining upper bounds
 5. The Implementation
  5.1. An Example Learning Episode
  5.2. Special cases
 6. Formal foundations
  6.1. Feature structures
  6.2. Ordering feature structures
  6.3. Combining information
 7. Conclusion
 References

저자

• Jae-yoon Kim [ Silla University ]

참고문헌