The Research On Machine Translation From English To Chinese Of Long English Sentence

With the rapid development of globalization, machine translation has experienced along period of time at the evolution and development. At first, machine translationconverts one language into another language. Over the past few centuries of thecontinuous improvement of information technology and computer technology, thecontents of machine translation is no longer a simple synthesis of computer science andlinguistics, which will involve corpus, semantics, natural language processing,mathematics and scientific computing, biological science and artificial intelligence andmany other fields. Although the level of machine translation development has had therapid development, the translation quality of machine translation is still not very high, itis also hard to satisfies the needs of users. One of the most major problem is theprocessing of English long sentences. Complex long sentences in English are verycommon, even throughout the English articles. The results of the analysis of thesecomplex long sentence will seriously affect the quality and readability of the translationmachine. So the processing of English long sentences is still a key problem in the field ofmachine translation.The methods of analyzing English long sentences and querying asingle word from the dictionary are very different. Vocabulary translation only requireslemmas and tags, and then to find the original form of the word in the database. Butsimple word or words analysis program is an initial part of the long sentence analysis. Inthe article, it also needs involving context, ambiguity of some words. On the analysis oflong sentence structure in the program, some questions such as subordinate clause,long-range correlation and their relationship are to be taken into account. So thesecomplex English long sentences can be analyzed correctly and translated into the targetlanguage of the users, and it has become the key factor influencing the rapid development of machine translation. On the other hand, the whole translation system resource faces aserious insufficient. And this will directly affects the ability of the users to capture usefulinformation. So the analysis of long sentences will be a key issue of completinginformation strategy.This article is mainly looking for ways which is good for the translation of Englishlong sentences on the basis of the existing English-Chinese machine translation. Basedon the part of speech and the improving of the rules, it will matches more sentencemodels in order to improve the existing quality of machine translation. The topic mainlyconducts and researches long English sentence segmentation, so as to achieve the goal ofsegmentation, which can effectively shorten the length of the sentence and is helpful toform the correct translation. Through the improving of the regular matching method,itcan achieve to the sentence segmentation and simplify the sentence patterns. The rules inthe traditional rule-based machine translation model are heavily depended on linguistsummarizes debugging. Now machine translation system based on rules built in thispaper is mainly to build enough translation information base and rule base. And the rulesare the combination of corpus knowledge to summarize in this paper. And the rule-basedmachine translation method can express a language or a statement syntax, then acomputer can understand natural language in accordance with the regulations set by thepeople.