Lingvist oxirgi shakl 1-son indd
Download 1.51 Mb. Pdf ko'rish
|
Lingvist OXIRGI SHAKL 1-son
Aziza RAKHMANOVA
156 language: 1. English-German translation corpus 2. Anglo-Norwegian Parallel Corps (ENPC) 3. English-Swedish Parallel Corpus (ESPC), created in 1993. It has now become an important resource for learning English and Swedish. The corpus database contains 64 English texts and trans- lations, 72 Swedish texts and translations. The corpus contains 2.8 million words. The texts are adapted as much as possible in terms of text type, topic, and style, so they can be used as a two-way paral- lel corpus and as a comporative corpus. Served as a resource for re- search on epistemic modality and adverbial conjunctions in English and Swedish. Parallel cases are also available in structured form. For exam- ple, the CLARIN infrastructure provides access to 87 parallel corpus, many of which are available for download from national databases, as well as concordances such as Corp, Corpuscle, and KonText. The CLARIN infrastructure contains 47 bilingual corpus, mainly Europe- an language pairs, but also Hindi, Tamil and Vietnamese corpus. 40 corpus in the system are multilingual, and 5 contain texts in more than 50 languages. Almost half of the corpus base is almost 100% adapted in terms of translation, and such sources serve as analysis material in comparative and cross-sectional studies. The user of the parallel corpus can find translations of a cer- tain word or phrase from Russian to other languages or from other languages to Russian. Currently English-Russian, Russian-English, German-Russian, Russian-German, French-Russian, Russian-French, Spanish-Russian, Russian-Spanish, Italian-Russian, Russian-Ital- ian, Polish-Russian, Russian-Polish, Ukrainian -Russian, Rus- sian-Ukrainian, Belarusian-Russian, Russian-Belarusian and multi- lingual parallel corpus search is available. Multimedia corpus containing fragments of video and au- dio films from 1930-2000 have been created. Such corpus can be searched not only for spoken text, but also for gestures, gestures (nodding, shrugging, etc.) and type of speech act (agreement, sar- casm, etc.) [Johansson, 1994]. The ability to create these corpora and the search engine will be enhanced with the help of artificial intelligence. A corpus of parallel texts is an electronic representation of works of art, manuals, mass media, various documents in two or more languages. According to the character of the database and text features, the corpus of parallel texts can be classified as follows: The role of parallel corpora in the improvement of artificial intelligence 157 1. Parallel corpus of scientific style texts. The corpus com- piled according to the text of manuals and educational literature are intended to illuminate the features of the scientific method and pro- vide theoretical information on a specific field. Such corpus serve as a resource for creating educational corpus. The corpus of parallel texts can be used as a linguistic database in the language education and language teaching system. Such corpus expand the possibilities of artificial intelligence in the translation of educational literature, editing of scientific text for educational purposes. The formation of parallel corpus based on the text of scientific articles is of practical importance in mastering the basics of science, improving the quality of practical training, and improving the inde- pendent education process. It is desirable to place scientific articles in a parallel corpus in a themed, systematic manner. The creation of parallel corpus of international scientific journal series and confer- ence materials serves the wide spread of scientific ideas and views across cultures. 2. Parallel corpus based on journalistic style. The corpus of parallel texts compiled within mass media provides complete infor- mation about the type and content of mass publications. It makes it possible to master the content of the issues covered in the press pub- lications. Placing oral versions of television and radio texts (audio texts) in the corpus of parallel texts increases the level of presenta- tion of the database. Newspaper texts, scientific-popular, social-po- litical magazine texts serve to spread publicist style texts widely. 3. Parallel corpus based on formal style. Parallel texts relat- ed to official business documents are important in determining the style of official business in different languages, the format of regula- tory and legal documents. The placement of the Uzbek language ad- ministrative system in parallel corpus makes it possible to compare the form, content, purpose, and types of normative documents with the target language. 4. Parallel corpus based on artistic style. The corpus of paral- lel texts is of great importance in mastering the content of works of art, conceptual analysis, learning the basics of text linguistics, text features of different languages. The corpus based on literary texts also includes units specific to certain languages - real units. The cor- pus of parallel texts makes it possible to fully express lacunar and real units between the languages being translated, and to determine the principles of translation of real units. It will be possible to get information about the translation of real units using methods such Download 1.51 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling