anguage processing is directly related to artificial intelligence (although it can be used with traditional computer techniques, as well). Concepts of an intelligent machine are inherently linked to its use and the manipulation of the primary representation of knowledge: words (May, 1991, p. 120). The goal of natural language processing is to have computers understand and respond to commands and inquiries issued in noncryptic, colloquial language, either typed or spoken.
Natural language processing is a specialty or subfield in the area of artificial intelligence. It includes members of the artificial intelligence community and the computational linguists. These experts, in turn, evolved from linguists and computer scientists who had traded expertise with one another. Each group of scientists continues to develop along their own theoretical bases, but there has been a combining of efforts over the past ten years (Pritchard-Schoch, 1993, p. 34). Because natural language development has commercial aspects, scientists working in the field are reluctant to share their information because of proprietary concerns. This makes understanding the current state of the technology difficult, and is likely to lead to several different philosophies being used in commercial natural language processors before a single standard finally emerges.
Early work in natural language processing focused on keyboard input. The software works through parsing and pattern matching techniques. The program first parses (diagrams the sentence structure) using rules of grammar, and then uses internal dictionaries to understand the words. Natural language processing eliminates the requirement of proper syntax, semantics and structure that Boolean logic requires, and enables the user to enter commands in common English.
Natural language processors come with built-in vocabularies of several hundred words; some include prepositions and articles. G
...