Application of programming languages for processing natural languages with special reference to english and tamil

Abstract

The program development has led to the following applications: i) Linguists working on various aspects of language, and ii) computer users who would like to use one of their own languages for linguistic data processing. Among the linguistic applications, the following are noted:

Declension of nouns and conjugation of verbs and identification of any given form occurring in a text as belonging to one of the declension or conjugation groups with details of parsing. Identification of parts of a word (i.e., prefix, suffix, ending, etc.) to get grammatical information about that word. Automatic parsing of a sentence and demarcation of the units in a sentence based on the PCG-STRUCTURE (P-Structure, C-Structure Grammar) Theory. Further analysis of meter in Tamil. The same method can be followed for other Indian languages for metre analysis.

The program packages presented here can be used at any computer centre for linguistic data processing as the programs are written in the most widely available programming language, namely FORTRAN. A combination of the programs can be used for any other type of linguistic analysis. For example, the frequency counting program included in the chapter on syntactic analysis could be used to analyse the style of the author or to enumerate the number of times a particular structure occurs in a text or in different texts. It is hoped that these programs, now readily made available to the linguists in this presentation, will be widely made use of by them.