Research On English And Chinese Interactive Machine Translation Technique Base On Word Prediction

Machine Translation(MT) technology has been widely used in the world after development for decades. Especially, statistical machine translation approaches based on large-scale corpus have made a series of remarkable achievements. However, up till now, neither rule-based nor corpus-based MT systems can solve the complex language issues such as vagueness and polysemy etc. Therefore, MT systems cannot replace human to generate high-quality translation.Under such circumstances, Computer Aided Translation(CAT) is proposed, which provides a platform for the cooperation of machine and human translators. In order to further improve the interaction between machine and human translators, Interactive Machine Translation(IMT) is proposed. This technology allows human translators to carry out manual intervention and guidance in the translation progress, and reduces the repetitive and mechanical operations of manual translation by means of translation memory, online dictionary etc., consequently improving the translation efficiency.The main work of this paper is to study and analyze the method and strategy of the current interactive machine translation technology. To deal with the problem of frequent switches between mouse and keyboard when the users search bilingual words in the process of translation, a prediction model oriented to word searching behavior was proposed. The word searching behavior was transformed into the translation selection problem under the condition of current translation, and the prediction of the word searching behavior was achieved with high accuracy using the alignment model, translation model and the language model. And then we transform the mouse selection behavior in the interactive process into the word alignment information of the partial translation, dynamically labels the word alignment of the translation, and add the word alignment information to the next prediction process. Experiments on English-Chinese machine translation show that the introduction of dynamic word alignment information can improve the prediction accuracy of interactive machine translation and improve the efficiency of translation.