Recently MeroSpark is lunched with more features and services, now you can ask your question, sell your books, share your notes and many more. Visit now and create your account to take full advantage of MeroSpark.

Natural Language Generation | Cognitive Science & AI | BSc.CSIT | 4th and 5th Semester

Download our Android App from Google Play Store and start reading Reference Notes Offline.

natural language generationNatural Language Generation
Subject: Cognitive Science and AI | BSc.CSIT (TU)
Fourth And Fifth Semester | Tribhuvan University

Natural Language Generation
“Natural Language Generation (NLG), also referred to as text generation, is a subfield of natural language processing (NLP; which includes computational linguistics)

Natural Language Generation (NLG) is the natural language processing task of generating natural language from a machine representation system such as a knowledge base or a logical form.

In a sense, one can say that an NLG system is like a translator that converts a computer based representation into a natural language representation. However, the methods to produce the final language are very different from those of a compiler due to the inherent expressivity of natural languages.

NLG may be viewed as the opposite of natural language understanding. The difference can be put this way: whereas in natural language understanding the system needs to disambiguate the input sentence to produce the machine representation language, in NLG the system needs to make decisions about how to put a concept into words.

The different types of generation techniques can be classified into four main categories:

  • Canned text systems constitute the simplest approach for single-sentence and multi-sentence text generation. They are trivial to create, but very inflexible.
  • Template systems, the next level of sophistication, rely on the application of predefined templates or schemas and are able to support flexible alterations. The template approach is used mainly for multi-sentence generation, particularly in applications whose texts are fairly regular in structure.
  • Phrase-based systems employ what can be seen as generalized templates. In such systems, a phrasal pattern is first selected to match the top level of the input, and then each part of the pattern is recursively expanded into a more specific phrasal pattern that matches some sub-portion of the input. At the sentence level, the phrases resemble phrase structure grammar rules and at the discourse level they play the role of text plans.
  • Feature-based systems, which are as yet restricted to single-sentence generation, represent each possible minimal alternative of expression by a single feature. Accordingly, each sentence is specified by a unique set of features. In this framework, generation consists in the incremental collection of features appropriate for each portion of the input. Feature collection itself can either be based on unification or on the traversal of a feature selection network. The expressive power of the approach is very high since any distinction in language can be added to the system as a feature. Sophisticated feature-based generators, however, require very complex input and make it difficult to maintain feature interrelationships and control feature selection.

Many natural language generation systems follow a hybrid approach by combining components that utilize different techniques.

(Visited 188 times, 1 visits today)

Posted By : Digvijay | Comment RSS | Category : Fifth Semester, Fourth Semester
Tag : ,

Post a Comment

Your email is never published nor shared. Required fields are marked *


Wordpress DMCA
Community | Toolbar | Android App | Founder/Developer : Hari Prasad Chaudhary | CSIT Portal Manager : Digvijay Chaudhary