中文
校园招聘信息
请以中国出版集团的校招通知为准
All>News Center> view

GTCOM's high-quality machine translation engine supports medical research information translation

NewsSource:GTCOMDate: 12 March 2020views:37

On February 28, the World Health Organization (WHO) raised the risk level of the global COVID-19 epidemic to "very high", calling on all countries to unite and strengthen their prevention-and-control practices in order to prevent the epidemic from spreading on an even greater scale. In appreciation of China's efforts on epidemic prevention and control, the Report of theWHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19), issued on February 29, asserts that China has taken the most courageous, flexible, proactive epidemic prevention-and-control measures in history, providing important experience for global response to COVID-19.

Amid the global outbreak of COVID-19, GTCOM's YeeKit translation platform updates its corpus data and upgrades its machine-translation engine correspondingly. The high-quality medical machine-translation engine can assure instant translation of the global medical research information and epidemic information, help other countries to acquire a rapid understanding of China's experience in the control of epidemic pathogens, help China to obtain insight regarding the global epidemic situation and thereby promote the research on COVID-19. The platform is opening its machine-translation services to the global public free of charge.

1584931130(1).jpg

YeeKit online machine-translation platform
(fanyi.yeekit.com)


Based on Professional Epidemic Corpus Data,Improve the Translation Quality of Medical Information

Amid the COVID-19 outbreak, the YeeKit R&D team has rapidly collected, collated and analyzed the hot words, new words and application contexts related to the epidemic and added them to the epidemic corpus after being translated by professional personnel and processed and reviewed by the corpus team. Meanwhile, the quality of the translation of COVID-19 epidemic information has improved through the use of advanced machine learning, deep neural network and other AI technologies to train the machine-translation engine. This improvement has been enhanced with the adoption of detailed long-sentence processing rules and strategies to humanize the translation results. In addition to textual corpus, the platform has updated more than 400 new and hot words related to the epidemic, including proper names and professional terms, such as "International Committee on Taxonomy of Viruses", "covert infection" and "COVID-19".

image.png

Examples of hot words related to COVID-19 included in the corpus of YeeKit translation platform


Multilingual Translation ToolHelp You Read Global Epidemic Information in Your Native Language

We selected an English-language report titled "The Vice Minister of Health of Iran Is Infected with COVID-19" and a Chinese report regarding domestic epidemic prevention and control as samples for translation.

image.png

YeeKit Translation Platform


Example about English-Chinese Translation of Epidemic Information


image.png

YeeKit Translation Platform

Example of Chinese-English Translation of Epidemic Information

It can be seen from the above two news reports that the real-time translation results fed back by YeeKit represent a relatively high-level machine translation in terms of the accurate processing of vocabulary and phrases as well as in the holistic consideration of the meanings of long sentences and paragraphs. Additionally, supporting the automatic detection of source language, the platform can, immediately after users input the texts to be translated and select the target language, complete the translation in a real-time manner. Thus, the tool makes it possible for users to read global epidemic information in their own languages.

If it's inconvenient for you to copy and paste, you can use our webpage plug-in tool, which can be integrated with more than 10 major browsers to achieve instant webpage translation with a single click. It also facilitates bilingual reading, thus offering you an unprecedented reading experience.

image.png

Webpage plug-in - translation sample

(Downloading link: web.yeekit.com)


A Rapid Rise Due to Accumulated Strength,A Top Player in Machine Translation

In the competitive field of machine translation, it is difficult to keep going. As one of the top players, GTCOM launched the YeeKit Chinese-English machine-translation engine in June 2014, thus becoming one of the first Chinese enterprises to independently develop a machine-translation engine, following Google, Microsoft and Baidu. Currently, the YeeKit translation platform supports the translation between more than 50 languages, of which the official languages cover 90% of the world's population.

image.png

Languages supported by GTCOM's machine-translation technologies

Generally, data is one of the main factors in determining the quality of machine translation. The volume and precision of the data can also influence the effectiveness of a machine-translation engine. Based on its basic data, totaling five billion sentence pairs, the YeeKit online machine-translation platform continues its rapid expansion. It adds two billion high-quality sentence pairs annually through multiple open-source channels such as enterprise translation production, LBDA (Language Big Data Alliance) data sharing and open online assisted translation platforms, becoming one of the largest parallel corpora worldwide. Thus, the application of professional corpus data serves to enhance the accuracy of machine translation. For example, by relying on its corpus resources and technological advantages, GTCOM's machine-translation engine was an outstanding performer at the Conference on Machine Translation. It has become one of the top players in the field.

image.png

GTCOM machine-translation engine

Contribute Multiple Outstanding Performances in the Conference on Machine Translation

Given the worldwide spread of the COVID-19 epidemic, linguistic services will experience heavy demand in aspects ranging from the preparation of anti-epidemic supplies at home and abroad to the emergency management of the epidemic; from the science popularization of COVID-19 worldwide to the reporting of domestic and foreign epidemic information, and to the sharing of effective prevention, control and treatment schemes. As the only 4A language-service enterprise in China, as well as a leading provider of cross-language big-data technology and language technologies, GTCOM will maximize its machine translation advantages and, through innovative language technologies and means, keep providing emergency language services to support epidemic prevention and control. Thus, users who log into "fanyi.yeekit.com" can freely access the YeeKit translation platform, acquire instant translation of the global epidemic information, learn about the new progress in epidemic research, get real-time insight into successful practices and jointly fight the tough war against COVID-19.

Enterprise or institution users who need such services as the tools for batched document translation, the customization of exclusive machine translation models, the machine-translation engine for multiple vertical fields such as finance, science and technology and medical care, or the deployment of private cloud for machine translation, please feel free to contact us:

2020nlp@gtcom.com.cn

Scanning two dimensional code to share WeChat

Contact Us