|
|
Hybrid Machine Translation Approaches for Low-Resource Languages
Kamran, Amir ; Popel, Martin (advisor) ; Kuboň, Vladislav (referee)
In recent years, corpus based machine translation systems produce significant results for a number of language pairs. However, for low-resource languages like Urdu the purely statistical or purely example based methods are not performing well. On the other hand, the rule-based approaches require a huge amount of time and resources for the development of rules, which makes it difficult in most scenarios. Hybrid machine translation systems might be one of the solutions to overcome these problems, where we can combine the best of different approaches to achieve quality translation. The goal of the thesis is to explore different combinations of approaches and to evaluate their performance over the standard corpus based methods currently in use. This includes: 1. Use of syntax-based and dependency-based reordering rules with Statistical Machine Translation. 2. Automatic extraction of lexical and syntactic rules using statistical methods to facilitate the Transfer-Based Machine Translation. The novel element in the proposed work is to develop an algorithm to learn automatic reordering rules for English-to-Urdu statistical machine translation. Moreover, this approach can be extended to learn lexical and syntactic rules to build a rule-based machine translation system.
|
|
|
Hybrid Machine Translation Approaches for Low-Resource Languages
Kamran, Amir ; Popel, Martin (advisor) ; Kuboň, Vladislav (referee)
In recent years, corpus based machine translation systems produce significant results for a number of language pairs. However, for low-resource languages like Urdu the purely statistical or purely example based methods are not performing well. On the other hand, the rule-based approaches require a huge amount of time and resources for the development of rules, which makes it difficult in most scenarios. Hybrid machine translation systems might be one of the solutions to overcome these problems, where we can combine the best of different approaches to achieve quality translation. The goal of the thesis is to explore different combinations of approaches and to evaluate their performance over the standard corpus based methods currently in use. This includes: 1. Use of syntax-based and dependency-based reordering rules with Statistical Machine Translation. 2. Automatic extraction of lexical and syntactic rules using statistical methods to facilitate the Transfer-Based Machine Translation. The novel element in the proposed work is to develop an algorithm to learn automatic reordering rules for English-to-Urdu statistical machine translation. Moreover, this approach can be extended to learn lexical and syntactic rules to build a rule-based machine translation system.
|
|
|
Evaluation of SMT translation systems (Google translate, Bing) from French to Czech: collocations related to "Waste management"
VÍŠKOVÁ, Barbora
This bachelor thesis is focused on a Evaluation of SMT translation systems (Google Translate, Bing) from French to Czech: collocations related to "Waste management". At first, the work describes a brief history of machine translation. Then the work describes the basic principles of machine translation like the rule-based machine transla-tion, the statistical machine translation, the hybrid machine translation and the by com-puter aided translation. The online translators GT and MB are introduced as well. These translators are completed by specific studies of correctness their evaluations. The automatic translations of these tranlators are evaluated on the based of outlined typology of resulting mistakes. These automatic translations are the most important ob-ject of this bachelor thesis.
|
|
|
Evaluation of SMT translation systems (Google translate, Bing) from French to Czech: collocations related to "Food safety"
ŠVARCOVÁ, Zora
The aim of this bachelor thesis is to test and evaluate the translation quality of selected terminological collocations related to "food safety" using free available translators (Microsoft Bing Translator, Google Translate). The work is divided into two parts theoretical and practical. Theoretical part is divided into several chapters, which focus on the development, history and use of statistical machine translation, the basic strategies of machine translation systems (rule-based machine translation, statistical machine translation, hybrid machine translation and computer-aided translation). Furthermore, the first part deals with the history and characteristic of two online translators Microsoft Bing Translator and Google Translate. Finally, there is a brief description of the term "collocation ". Practical part is focused on the evaluation of translations of selected collations, whose quality is evaluated by comparison with the available terminology databases (IATE).
|
|
|
Evaluation of SMT translation systems (Google translate, Bing) from French to Czech: collocations related to "banking"
SEKALOVÁ, Tereza
The aim of this bachelor´s thesis is to test and evaluate the translation quality of selected terminological collocations in the area of "banking" using freely available translators (Google Translate, Bing Translator). The thesis consists of the necessary theoretical knowledge of machine translation and translation evaluation. The theoretical knowledge includes historical development of machine translation, translation types (rule-based machine translation, statistical machine translation, a hybrid machine translation and computer aided translation) and approaches some basic information about Google Translate and Microsoft Bing Translator. The evaluation of the translations was done by analysing machine-translated collocations and their subsequent comparison with the available terminology databases (esp. IATE) and with specialized dictionaries.
|
guest ::
