Text Selection Issue For Parallel Corpus

It is known that the basis of any corpus is its units. Typically, texts of different genres are selected as the corpus unit to ensure the representativeness of the corpus. Therefore, when creating any language corpus, first of all, the principles of selection of texts that are part of it should be defined. Parallel corpus units consist of texts that have been translated one or more times from the original. Which topic and genre text to choose for the parallel corpus is determined by the purpose of the compiler?


INTRODUCTION
Parallel corpuses can represent a one-way translation (from Uzbek to English only) or a two-way translation (from Uzbek to English, from English to Uzbek). When selecting a text for the Uzbek-English parallel corpus, it is advisable to collect both direct and indirect translations from Uzbek into English and from English into Uzbek. Thus, the Uzbek-English parallel corpus consists of two blogs: the Uzbek-English language blog (1), the English-Uzbek language blog (2). The materials for these blogs are not the same: the first blog contains works of art translated from Uzbek into English; the second blog consists of works translated from English into Uzbek. Of course, in the search engine, the works selected for search by genre, period of creation, author and translator of the works require selection in this regard. We will focus on these aspects, which are identified on the basis of extralinguistic tags, in the following sections of the work.

LITERATURE REVIEW
M.Kholbekov, a specialist in translation theory and a skilled translator, provides valuable information about translations from foreign languages in the twentieth century. According to him, in the history of world civilization, the twentieth century was called the "century of translation" (P.-F. Kaye). Uzbek translation, which has a thousand-year history, has also achieved unprecedented results in the last century, introducing our people to the masterpieces of world literature. "Today, translation is one of the main factors of interlingual and intercultural communication. In short, "translation introduces the nation to the world," paving the way for us to participate directly in the process of international globalization. We must never forget that." There is a saying, "Shakespeare is infinite". The word belongs to the pen of the great German poet Johann Wolfgang Goethe. Indeed, the legacy of William Shakespeare, a great exponent of English Renaissance literature and theatrical art, has been captivating the peoples of the world for four centuries. In Uzbekistan, too, the name of Shakespeare, his works are well-known and popular, deeply rooted in the hearts of readers and viewers. Almost all of the royal works created by Shakespeare over the past eighty years have been translated into our language and presented to the reader in a number of editions. During the period of independence, the publishing house "Fan" published another three-volume "Selection" of Shakespeare's works, translated by the poet Jamal Kamal, and did another good deed in this regard.
The translation of Shakespeare's creative heritage is also significant in that it served as a material for the creation of the first Uzbek-English parallel corpus. The question arises: if there is a translation of the work, how important is it to re-create a parallel corpus from the finished product? Since the corpus is designed to collect ready-made translations, not to translate, what does it mean for science, the user, the theory of translation? Isn't that an unnecessary process and product?
The answers to the above pertinent questions are as follows: 1) the existing world parallel corpus is based on ready-made translations; 2) combining the translation into a parallel body allows to read the original work and the translation of the whole work side by side; 3) most importantly, the parallel corpus is the most optimal way to observe how this or that unit (simple sentence, compound sentence, phrase, phraseological unit, slang, metaphor) is expressed in the original and in translation. Therefore, in the choice of material for any parallel corpus, the principle of obtaining a variant of the translation made directly from the original prevails. Because the units listed above lose their value in indirect translation, in this case the parallel corpus cannot fully perform its function.

CONCLUSION
We have given the observations of M. Kholbekov in detail above. The reason for this was to recall once again the learning process of translation. The researcher, who aims at the comparative study of translations, carefully examines all its copies; finds and compares each piece from this or that option. The expert searches each passage dozens of times to show the advantages and disadvantages of translation. If the software of the Uzbek-English parallel corpus is created, and the linguistic base collected for the corpus is subordinated to a perfect and convenient search system, the researcher's work will be facilitated by automation; the processes of comparison and comfortable become clearer; the conclusions will be clear.
Therefore, only the results of the linguistic corpus-based analysis of the issues studied in linguistic research conducted abroad -the conclusions based on statistics -are recognized by the scientific community. The conclusion reached without relying on the language corpus is considered non-objective.