ПРОБЛЕМА ПОШУКУ ГРУП СЛІВ У ДВОМОВНОМУ ПАРАЛЕЛЬНОМУ КОРПУСІ
Journal Title: INTERNATIONAL ACADEMY JOURNAL "Web of Scholar" - Year 2017, Vol 1, Issue 3
Abstract
The article considers algorithms of word combination fuzzy search in text. Tanimoto, Levenshein, Jaro and Jaro-Winkler metrics of text similarity have been considered and investigated experimentally using real data from document storage of document support system. Research has confirmed the pros and cons of methods. Jaro-Winkler distance has been recognized as the most powerful for prefix search in systems with low quantity of mistakes. Investigation results have been applied to multi-language document support system for purpose of search across parallel corpora as human-aided machine translation module. SMT can help in improving translation result but should not be the key component.
Authors and Affiliations
Б. В. Жуковецький
Keywords
Related Articles
- EP ID EP458024
- DOI -
- Views 152
- Downloads 0
НЕКОТОРЫЕ СВОЙСТВА БАКТЕРИОЦИНОПОДОБНОГО ИНГИБИРУЮЩЕГО ВЕЩЕСТВА (БПИВ) ШТАММА LACTOBACILLUS PARACASEI SPP. PARACASEI BN ATS 10W
Lactobacillus possesses a pronounced antagonist activity with respect to pathogens and conditionally pathogens as well as to other kinds and even to Lactobacillus strains. Lactic acid bacteria widely distributed in the n...
ETYMOLOGY OF SLANG: ITS ORIGIN AND DEFINITION
The English language can be categorized in different social varieties, and one of important subsystems of modern English is slang. The topicality of my investigation is conditioned by the necessity to study different col...
СОСТОЯНИЕ ФИЗИЧЕСКОГО РАЗВИТИЯ, СОМАТИЧЕСКОГО И РЕПРОДУКТИВНОГО ЗДОРОВЬЯ ДЕВОЧЕК-ПОДРОСТКОВ КЫРГЫЗСКОЙ РЕСПУБЛИКИ
In order to assess the physical development, general and reproductive health of kyrgyz nation adolescent girls, a survey of 390 high school students in Bishkek was conducted. Most of the examined girls have average indic...
METHODS IN RESEARCH WORKS OF EDUCATION MANAGEMENT PROBLEMS
The article focuses on the problem of methods application in research of education management problems in accordance with tendencies of science development. Attention is drawn to choice dependence of research methods sys...
SPACE TECHNOLOGY FOR MODELING METEOROLOGICAL CONDITIONS
This article discusses one of the most important topics in the space industry, including in the space industry of the Republic of Kazakhstan. The article contains information on how to use space technology to track and p...