INFLUENCE OF WORD ORDER ON QUALITY OF AUTOMATIC ANALYSIS OF TEXT


Keywords: automatic syntactic text processing, word order, syntactic ambiguity, homonymous component, dominant

Abstract

The relevance of the article is due to the interest of applied linguists in the process of creating special programs for teaching modern computer systems the rules of automatic text processing that are necessary for conducting high-quality automatic analysis and synthesis of texts.

The purpose of this article is to consider the word order as the main factor in the occurrence of syntactically ambiguous statements that adversely affect the efficiency of automatic text processing systems, as well as the analysis of homonymous constructions using the automatic syntactic analysis of sentences.

The object of the research in the article is syntactically ambiguous sentences; the subject of analysis is the word order in statements as the reason for the syntactic homonyms appearance.

The study of the material confirms the fact that human speech is linear, as a result of which words connected grammatically or meaningfully are arranged in a chained statement. However, the grammatical structure is too complex and cannot be expressed only by the linear following relation. Therefore, a part of grammatical meaning is expressed by the word order, and the other part is expressed by morphological categories, auxiliary words or intonation. Violation of the rules of following words in a sentence can lead to a change in its meaning or to grammatical incorrectness of expression.

In the process of speech, the same statement may acquire different meanings depending on its communicative task. However, a change in the direct word order often leads to the appearance of such a negative linguistic phenomenon as syntactic homonymy. In constructions containing syntactic homonyms, a double bond is realized, in which one word or group of words can simultaneously obey two dominants, which leads to a change in the utterance semantics. The presence of syntactic homonyms in the text makes the communication process difficult, since it puts the addressee of speech in front of the need to choose one of several meanings of the utterance. Polysemy is a serious problem in the implementation of machine translation and automatic text processing, since at the present stage of artificial intelligence development, a machine can work effectively only with structures that have a strictly unambiguous structure.

The results of this study showed that, despite the achievements of modern IT technologies, computer systems are still not perfect; they cannot replace a human being  where flexibility of thinking is required. At the present stage of computational linguistics, automatic analysis of texts containing sentences with direct word order is effective. In addition, for a correct understanding of the semantics of a statement and its syntactic structure, a computer program must have access to the corpuses of texts that allow it, taking into account the context, to determine the true meaning of the statement.

We associate the prospects for further research with the development of such a field as computational linguistics, which deals with the problem of creating artificial intelligence that could process texts with the same efficiency as a human being.

Author Biography

O.V. Shkurko, Oles Honchar Dnipro National University, Ukraine;

Candidate of Sciens (philology), Associate professor

References

Литература:

1. Власов М. С., Савостьянов А. Н., Сапрыгин А. Е., Астахова Т. Н. Разрешение глобальной синтаксической неоднозначности двуязычными испытуемыми в условиях психолингвистического эксперимента / Вестник Томского государственного университета. Филология. 2015. №4 (36). С. 5-18.
2. Кобзарева Т. Ю. Иерархизация синтаксического анализа на основе свойств линейной структуры русского предложения: диссертация на соискание ученой степени к. филол. н. Москва : РГГУ. 192 с.
3. Кудашина В. Л. Коммуникативные функции порядка синтаксических сегментов в современном русском языке: диссертация на соискание ученой степени к. филол. н.: специальность: 10.02.01 - русский язык. Ставрополь : СГУ. 2003. 190 с.
4. Митренина О. В. Проблемы неоднозначности синтаксического анализа: диссертация на соискание ученой степени кандидата филологических наук: специальность 10.02.01 – русский язык. Санкт-Петербург. 2005. 133 с.
5. Шкурко Е. В. Лингвистические и психологические особенности синтаксического анализа неоднозначных предложений / Вісник університету ім. А. Нобеля. Серія: Педагогика. Психологія. Вип. 1(15). Дніпро : Вид-во Ун-ту ім. А. Нобеля, 2018. С. 214-218.
6. Шкурко Е. В., Горб О. В. Влияние синтаксической омонимии на качество автоматического анализа предложений / Сучасна вища освіта: проблеми та перспективи: VI Всеукраїнська науково-практична конференція студентів, аспірантів і науковців: тези доповідей. Дніпро : Ун-т ім. А.Нобеля, 2018. С. 213-215.
Published
2019-04-30
How to Cite
Shkurko, O. (2019). INFLUENCE OF WORD ORDER ON QUALITY OF AUTOMATIC ANALYSIS OF TEXT. Linguistics. Lingvoculturology, 13, 182-190. Retrieved from http://lingvodnu.com/index.php/journal/article/view/149