Lexical Interference as an Automatic Translation Problem

Bilyalova, Albina Anvarovna
D.Sc. in Philology, Professor, Head of Philology Department Kazan Federal University, Kazan, Russia
abill71@mail.ru

Bulatova, Ilgina Zamilevna
Postgraduate Student, Philology Department, Kazan Federal University, Kazan, Russia
IlZBulatova@kpfu.ru


Abstract
Automatic or machine translation is perhaps one of the most challenging tasks of artificial intelligence, taking into account the flexibility of human language. Machine translation technology is a multidisciplinary problem due to its close connection to the problems of modeling the processes of understanding and generating texts in natural language. The subject of this study is the problem of lexical interference and its impact on the quality of machine translation. Lexical interference is considered in the framework of Runglish, which is formed by transforming English words or word combinations into Russian through the addition of prefixes, suffixes, and endings in order to adapt English vocabulary for use in everyday speech. Based on the definition of Runglish, it is clear that the condition for the emergence of language interference is the contact of the languages spoken by the speaker. The result of language interference is the transfer of phonetic, lexical, grammatical, ethnic features of the source language, that is reflected in deviations from language norms, which lead to linguistic neologisms. The task of research is to study such neologisms caused by lexical interference, which is observed in Runglish, because such neologisms represent a certain difficulty in their analysis by artificial intelligence.
The features of the Runglish language are presented at the level of acrolect, mesolect and basilect. The authors consider examples of doublet xeno-lexics, hybrid words, jargonisms, varvarisms, etc. in the Runglish language.

Keywords: artificial intelligence, automatic translation, lexical interference, Runglish language