International scientific conference "INFORMATION TECHNOLOGIES, NETWORKS AND
TELECOMMUNICATIONS" ITN&T-2022 Urgench, 2022y April 29-30
468
care,
fraud prevention, and support for more efficient administrative and clinical
workflows.
Sources used
1.
Werbos, P. J. (1988), "Generalization of backpropagation with application to a
recurrent gas market model", Neural Networks, 1 (4): 339–356, doi:10.1016/0893-
6080(88)90007-X.
2.
Nikos Paragios and Yunmei Chen and Olivier Faugeras (2005). Handbook of
Mathematical Models in Computer Vision. Springer. ISBN 978-0-387-26371-7.
THE USE OF ARTIFICIAL INTELLIGENCE IN THE
IMPLEMENTATION OF THE MULTILINGUAL WEBSITE OF THE
DORUL HIKMAT PROJECT
Khujaev Otabek, Rustamov Oybek, Allambayeva Yorqinoy
Urgench branch of the Tashkent University of Information Technologies named
after Muhammad Al-Khwarizmi.
h_q_otash@mail.ru, oybekrustamov@icloud.com
Annotation: This article discusses the implementation of the multilingualism of the
Dorul Hikmat project website with the use of artificial intelligence for text
translation from the Yandex company of the Yandex Translator service.
Keywords:
multilingual, yandex, api, development.
When developing a web application for the Doral Hikmat project, it was
decided to implement content for different languages of the world. to make it
easier for Internet users to understand information on the site in their own
language. Several options for implementing multilingualism
on the site were
proposed.This is how multilingual websites began to be created in order to
facilitate work with a web application for different people of our planet. Here are a
few of them:
using different pages for different languages;
using multiple language content on one page;
using artificial intelligence to translate pages.
When using multiple pages for different languages, it was not a very optimal
solution, since in order for the site visitor to find the necessary information for
himself, the visitor had to go to a certain page with his native language and this
was not a very good solution, since the visitor may not understand exactly where
his page
with his language is located, since the visitor may not understand or in
many cases not find the information, even though it is in front of his face, but in a
different language.
Multilingualism is the use of more than one language, either by an individual
speaker or by a group of speakers. It is believed that multilingual speakers
outnumber monolingual speakers in the world’s population.[1]
International scientific conference "INFORMATION TECHNOLOGIES, NETWORKS AND
TELECOMMUNICATIONS" ITN&T-2022 Urgench, 2022y April 29-30
469
Using multiple language texts on the same page can slow down the web
application as it will load more information when you first visit the site. Firstly, it
is inconvenient for the visitor, and secondly, it is not
beautiful to have Chinese
words, for example, with Russian content.
These two types of implementations of multilingualism for a web
application are not so suitable, since they have a lot of negative reviews from the
programmers themselves. Firstly, it is very time-consuming, and secondly, it takes
a lot of time.
In computer science, artificial neural networks, also referred to as ANNs, are
a computational information processing approach inspired by the mammalian
nervous system. ANNs consist of a large number of interconnected elementary
processing units called artificial neurons, whose structure
and functionality are
modeled after the biological neurons.[2]
The use of Artificial Intelligence was the most optimal for the
implementation of multilingualism on a web application. It has the following
advantages:
1.
Automatically determining the visitor’s country and the output of content
already translated into the desired language, which contributes to automatism
and does not switch the language by the visitor himself;
2.
There is no need to reload the website, as artificial intelligence can translate on
the fly;
3.
There are many services for translating web application
content on the fly
using Artificial Intelligence, such as Yandex Translate, Google Translate and
others.
For the implementation, a self-written script
was used to implement
multilingualism using artificial intelligence integrated into the Yandex Translator
text translation system. To connect it to the site, there is no need to rewrite the
entire application, you just need to connect the script
to the site and add the
element to the HTML code.
The source code is available on my GitHub repository with instructions:
https://github.com/oybekrustamov/yandex-translate-custom-widget
Do'stlaringiz bilan baham: 
