Un analyseur de contexte multilingue

Abstract

Man-machine communication in computer science is based essentially on the Latin alphabet and particularly on English, with a one-to-one correspondence between keyboard element, internal code and displayed character. However, most of the living languages do not exhibit this simple correspondence; therefore it is necessary to select a minimum set of representative elements that can fit in existing keyboards and internal code tables and generate, by means of a context analyser, the correct graphic output for each element input to the computer.