https://elib.bsu.by:443/handle/123456789/159720 2026-04-21T10:10:59Z https://elib.bsu.by:443/handle/123456789/160177 Заглавие документа: Robust multilingual document identification Авторы: Stefanovitch, N. Аннотация: We consider in this paper the problem of detection the language of document when no assumptions are made about a document: it can be of any size and contain zero, one or several languages. Language identification is considered a solved task, but actually, among others shortcomings, does not deal with the case of accurately the presence or absence of several languages in arbitrary documents. In order to tackle these problems, we propose an approach based on word dictionaries using Bayesian statistics and ad-hoc features. We show on two datasets that with sufficient statistics our approach is able to give very satisfying results in dealing with both unsolved tasks: detection of documents with no languages and identification of languages in multilingual documents. 2016-10-25T00:00:00Z https://elib.bsu.by:443/handle/123456789/160174 Заглавие документа: Computer-based system of analysis and interpretation of speech intonation Авторы: Lobanov, B. M.; Zhitko, V. A.; Zdaranok, Y. A. Аннотация: In this paper we describe the importance of intonation in human comprehension of speech and in creation of computer-based systems for analysis, synthesis and understanding of speech signals. We lay out fundamental principles of intonational theory of speech that is based on the concept of universal melodic portraits. In addition, we describe the main algorithms of analysis and interpretation of intonation in an utterance that underlie the developed computer-based system for teaching of speech intonation. We further show the system’s output and discuss its usefulness in teaching foreign language intonation. 2016-10-25T00:00:00Z https://elib.bsu.by:443/handle/123456789/160172 Заглавие документа: Canonization of natural language by means of universal semantic code Авторы: Boyko, I. Аннотация: The article is about canonization of a natural language for representing knowledge in the reasoning computer systems. Reasoning is a very difficult problem and the most important component of artificial intelligence systems. Universal Semantic Coode (USC) is used as a tool of canonization for knowledge representation and reasoning. There is no alternative yet in this kind of development and it is demonstrated in the article. 2016-10-25T00:00:00Z https://elib.bsu.by:443/handle/123456789/160169 Заглавие документа: Автоматическое распознавание аспектных терминов в пользовательских сообщениях на английском языке Авторы: Чернышевич, М. В. Аннотация: Описан подход к автоматическому распознаванию характеристик товаров и услуг в пользовательских сообщениях на английском языке. Предложенный подход базируется на лингвистическом анализе текста и последующем обучении CRF-модели на ряде лексических, синтаксических и статистических признаков. Разработана прототипная система, которая показала хороший результат при тестировании на текстах различной предметной области. 2016-10-25T00:00:00Z