IMPACT OF ONLINE MACHINE TRANSLATION SYSTEMS ON LIFELONG LEARNERS
IMPACT OF ONLINE MACHINE TRANSLATION SYSTEMS ON LIFELONG LEARNERS

Summary/Abstract: The economic growth, but also the provision and storage of multilingual texts, in considerable quantity, by the development of the Web and powerful computers, have increased the demand for automatic translations for different pairs of languages, within companies and administrations worldwide. The automatic translation is an entirely automatic process of translating a text from a source language to a text in a target language. No human intervention is required. However, current machine translation (MT) systems generate a lot of errors and obviously the human output intervention is still necessary in order to obtain a high quality for a given translation. If the purpose of the translation is not its high quality, but rather the comprehension of the transmitted message, online machine translation systems are tough very effective for quick translations, in real time, within companies and administrations. Thus, these systems can help professionals to improve their knowledge in their various areas of activity, from multilingual Web content, but also to learn foreign languages or to write technical documents in several languages. Machine translation systems are based on linguistic approaches or on purely statistical methods. Many of recent systems are hybrid. They combine statistical techniques and linguistic information. If linguistic systems provide effective results by requiring tough a major human and material effort, the hybrid systems can provide similar results by using less expensive resources. In this paper, we present the state of the art in machine translation field. The advantages and the disadvantages of different MT methods are also presented. The particular case of Romanian, which is a language less rich in electronic resources, is also discussed.

• Details
• Contents