![]() |
|||
![]()
|
![]() |
![]() |
![]() |
Chapter 19
|
1. | Introduction | ||
2. | History | ||
2.1. | The Origins | ||
2.2. | The ALPAC Report and the First AI Programs | ||
2.3. | The "Conceptual" Approach | ||
2.4. | Modern Times | ||
3. | Challenges and Solutions | ||
3.1. | Introduction | ||
3.2. | The Standard Paradigm for NLP | ||
3.3. | Morphological Analysis | ||
3.4. | Syntactic Analysis | ||
3.4.1. | Formal Grammars | ||
3.4.2. | Unification Grammars | ||
3.4.3. | Technical Issues | ||
3.4.4. | Linguistic Problems That Affect the Parse Operations | ||
3.5. | Semantic Analysis | ||
3.5.1. | Use of Semantic Procedures to Supplement the Syntactic Analysis | ||
3.5.2. | Shallow Semantic Representations | ||
3.5.3. | High-Level Languages for Meaning Representation | ||
3.6. | Discourse and Pragmatic Analysis | ||
3.7. | Language Generation | ||
3.8. | Interactive Systems and Spoken Language Understanding (SLU) | ||
4. | State of the Art and Future Trends in the NLP Domain | ||
5. | Summary and Conclusion | ||
References |
Natural language processing (NLP) can be defined, in a very general way, as the discipline having as its ultimate, very ambitious goal that of enabling people to interact with machines using their "natural" faculties and skills. This means, in practice, that machines should be able to understand spoken or written sentences constructed according to the rules of some natural language (NL), and should be capable of generating in reply meaningful sentences in this language. Computers able to fully decipher NL would, inter alia, remove the need to become computer literate, given that it would be possible to give the machine instructions directly in some natural, native language.
In its most ambitious form, this sort of endeavor is an extremely difficult one -- a full, fluent, and totally automated man/machine interaction is surely beyond the possibilities of the current technology. Nevertheless, in these last 20 years, some successes in dealing in an (at least partial) automatic way with NL have been achieved in several specific fields, and NLP techniques have gained considerably in popularity, attracting an increasing number of researchers and practitioners. Consequently, the NLP domain is now split into a multitude of lively subfields -- from speech recognition to handwriting analysis, from parsing to written and spoken language generation, from discourse and dialog modeling to document retrieval and mechanical translation, to mathematical methods, etc. -- which ask, in turn, for the intervention of several disciplines like linguistics, of course, but also psychology (see, e.g., the "user models"), mathematics, logics, physics, engineering, computer science, etc.
In this Chapter, we will focus on the NLP subfields that are of specific interest for ESs (and, more in general, for knowledge-based systems, KBSs). After having presented, in Section 2, a (short) historical account of the development of the NLP field, and in Section 3 the challenges proper to the NLP endeavor and the standard technical solutions adopted in this domain, we will describe briefly, in Section 4, the "state of the art" and the future trends of the domain. Section 5 concerns the summary and the conclusion, and is followed by References.
Previous | Table of Contents | Next |
![]() |
|
Use of this site is subject certain Terms & Conditions. Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details. |