Computer system of knowledge control on punctuation of Russian language

Abstract

The paper proposes a concept of the computer system for controlling knowledge of punctuation in the Russian language. The system is based on the algorithm for the identifying a suitable comma rule in the given sentence. The mentioned algorithm uses a formal classification of comma rules derived from the special information extracted from the sentence. The use of a large number of the abbreviations and the simplified words in the electronic correspondence has actually led to the loss of skills of literate writing. Often people make mistakes in the application of the rules of the Russian language, but the rules are not only an indicator of the degree of human development as individuals, sometimes they play an important role for the accurate understanding written, as in the famous example, "Execute not pardon". The problem generated by IT-technologies can be solved, at least partially, by means of these technologies, formalizing the process of the identifying suitable punctuation rules.

For the purpose of the automatically identifying suitable rule, it is convenient to divide the rulebook into some groups. The classification of the rules is the basis of the model version of the knowledge control system for punctuation of the Russian language. At the user's choice, the system can operate in two modes: test mode and training mode. The classification of the comma placement rules was originally developed for the use in the knowledge control program system, but also it can be considered as the independent guide for a big part of the rules of the Russian grammar. The model version of the knowledge control system for punctuation of the Russian language was tested on a fairly representative set of the examples that allow us to judge the prospects of the proposed approach.