Machine Translation (MT) is a pattern that convertsone kind of language symbols to another using computer. As Machine Learning(ML) algorithms have gotten more complex and powerful, Natural Language Process(NLP) techniques have been widely applied. Technical giants like Google andBaidu have developed progressive MT system and perform well. Although whetherMT will ultimately exceed human translation remains controversial, it becomesan irreversible trend that they should cooperate to better serve thetranslation industry.

The assessment of MT system is a cross-disciplinaryfield in both Computer Science and Linguists. By exploring how MT algorithmsprocess natural languages, it becomes clearer how they can be improved, and howthe human language work (Doddington, 2002). Atpresent most of studies are concentrating on how Machine Translation algorithmsiterate and help improve the quality of translation work. Wolk (2015) states ina study that Modern neutral network-based Machine Translation (NMT) isfundamentally different from previous statistical or rule-based systems and theperformance is constantly improving. But how to apply linguistic theories andthe functions of corpus has always been ignored. Thus, in addition to algorithmoptimization, we should pay attention to linguistics and its assistance ontranslation work.

At present, most MT evaluation tools use a statisticalmethod to count the consistency between the translation and the referencetranslation. The better they match with each other, the higher quality thetranslation will have. The implementation of this method requires aprecondition: there must be enough high-quality reference translation work. Thesystem has two methods of evaluating comparability, based on character stringsor on syntax structures. The former one used more frequently, such as BLEU andNIST. BLEU [1] is a statistical method to count the number of matching n-gramsbetween the translation and the reference translation, while NIST is a methodimproved on BLEU on the weight of n-grams.

The multidimensional (MD) analysis was invented byBiber(1988). Its initial purpose was to make comparisons between spoken andwritten English and has been gradually developed as an analytical method fortextual comparison. Many groundbreaking achievements have been obtained in manyresearch fields. Based on TOEFL 2000 Spoken and Written Academic LanguageCorpus, Biber (2002) analyzed spoken and written teaching materials in Americanuniversities using the MD approach. The study shows that the two types ofmaterials in American universities are in consistently polarized pattern. Inaddition to spoken and written register studies, the MD approach has also beenused in the field of English for specific purposes (ESP). After investigatingthe patterns of linguistic features in philanthropic direct mail letters,Connor and Upton (2003) found that direct mail letters are a unique genre,whose linguistic features are distinct from other ordinary genres. Helt (2001)makes a conclusion that spoken language of American English is not as formal asthat of British English after he studies the spoken variation of the twonational varieties. More recently, Koteyko (2015) focused on the analysis ofconsumer advertisements in British press. The research shows thatfemale-targeted advertising messages frequently elaborate products with lessscientific focus while male-oriented ones employ more disjunctive grammaticalstructures.

This paper chooses the Chineseand English version of "Legends of the Condor Heroes" and itsresearch material which is a classic literature work written by Jin Yong whichwas published by Guangzhou Press on April 1, 2013 (B0747BP96F) and it's namedChinese "Lord of the Rings". The English version of this chosen bookis translated and beautified by Anna Holmwood, an English translator, which isaccepted by Jin Yong. This translated work was published by MacLehose Press on February22, 2018(0857053000).