|
|
Evaluation of machine translation from French to Czech depending on the type of the text
KUPKOVÁ, Aneta
The aim of this bachelor thesis is to test and to evaluate the quality of machine translation from French to Czech depending on the type of the text, using free - Microsoft, Bing Translator and Google Translate. The work contains three parts. The first, theoretical part is divided into several chapters which are focused on the history and use of the machine translation, introduction to basic strategies of machine translation (Rule-Based Machine Translation, Statistical Machine Translation, Hybrid Machine Translation, Neural machine translation), development and history of online machine translation, presentation of two online translators - Microsoft Bing Translator and Google Translate. Furthermore, first part proposes an introduction to the MT evaluation, which is divided into human evaluation methods and automatic evaluation metrics. Finally, theoretical part includes the development and division of different functions of text in the Czech language. The second, methodological part contains the hypothesis of success of the machine translation. Furthermore, this part presents ten texts, which are analysed in the third part and their brief characterisation. Finally, there is a model of evaluation for the practical part. The last, practical part includes the evaluation of Google 's and Bing's translations using the automatic evaluation system - BLEU.
|
guest ::
RSS feed.