CN105677643A - Translation method combining manpower and machine - Google Patents
Translation method combining manpower and machine Download PDFInfo
- Publication number
- CN105677643A CN105677643A CN201610142419.XA CN201610142419A CN105677643A CN 105677643 A CN105677643 A CN 105677643A CN 201610142419 A CN201610142419 A CN 201610142419A CN 105677643 A CN105677643 A CN 105677643A
- Authority
- CN
- China
- Prior art keywords
- translation
- document
- standard
- written
- statement
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/42—Data-driven translation
- G06F40/47—Machine-assisted translation, e.g. using translation memory
Abstract
The invention discloses a translation method combining a manpower and a machine.The method comprises the following steps that 1, on the basis that it is guaranteed that sentence meanings of a document to be translated are not changed, standard sentence pattern rewriting is conducted manually on sentences in the document to be translated, a standard document is obtained, and identification is automatically conducted on the positions of specialized terms in all standard sentence pattern sentences in the standard document according to grammatical rules; 2, a translator extracts the specialized terms in the standard document and translates the specialized terms, and a specialized term translation is obtained; 3, the specialized term translation corresponding to a specialized term original text is filled in the standard document according to identification, and a first translation document is obtained; 4, the translator conducts correction and format fine tuning on the first translation document according to the document to be translated, and a finished translation document is obtained.According to the translation method combining the manpower and the machine, the technical problems are solved, the translation quantity of the translator can be minimized, the translation efficiency and the translation accuracy are improved, and the best translation achievement is obtained through the least manpower.
Description
Technical field
The present invention relates to translation technology field, relate in particular to a kind of manually in conjunction with the written translation method of machine.
Background technology
From first year 21st century, the method of existing online machine written translation is exactly nothing but the big data type written translation instrument that similar Google Google written translation device, Baidu's written translation device, Yahoo's written translation device etc. are similar, its principle is in that word for word carries out word replacement, and original text word is replaced to translation word. Due to the particularity of language performance, the function of machine written translation device is denounced always repeatly. The principle of machine written translation instrument is when implementing semantic meaning representation, macaronic semantic meaning representation word is carried out correspondence one by one fast accurate, is stored after correspondence. When identical original text word occurs, it is replaced without thinking. Unlimited multiformity due to language performance, each word, word, phrase or simple sentence be the corresponding more than one meaning often, the translation using machine written translation device obtained completely is generally not capable of normal expression original text implication, and therefore artificial written translation remains the guarantee obtaining height written translation quality.
In prior art, for a project or long document, in a team, it is divided into many parts often and translates, but owing to interpreter translates the difference of habit, the situation that different interpreter translates a word of identical meanings and causes translation inconsistent often occurs. It addition, this modes adopting team to be divided into many parts of written translations, cause that interpreter often repeats to translate and has the word of identical meanings, phrase or simple sentence, not only significantly increase the written translation intensity of interpreter, also greatly reduce written translation efficiency.
Summary of the invention
It is an object of the invention to solve the above-mentioned problems in the prior art, there is provided a kind of manually in conjunction with the written translation method of machine, present invention solves the technical problem that and be able to minimize interpreter's written translation amount, and improve written translation efficiency and improve translation accuracy rate, thus with the minimum written translation achievement manually getting the best in return.
For achieving the above object, the technical solution used in the present invention is as follows:
A kind of manually in conjunction with the written translation method of machine, it is characterised in that to comprise the following steps:
(1), treat in written translation document on the constant basis of the statement meaning ensureing, the statement manually treated in written translation document carries out standard clause rewriting, obtain the standard document with standard clause statement, further according to grammatical rules, the position of the proprietary term in standard clause statement each in standard document is identified automatically, wherein, described proprietary term includes word, phrase and simple sentence;
(2), by interpreter extract the proprietary term in standard document, and translate accordingly, obtain proprietary term translation;
(3), according to the mark in step (1), proprietary term translation corresponding for proprietary term original text is filled in standard document by mark, obtains just translation shelves;
(4), interpreter first translation shelves are examined and revised and after form fine setting according to until written translation document, namely obtain written translation document finished product.
Described step (1) treats statement in written translation document according to semantic and the tone and carries out standard clause rewriting,
Or use written translation aid extraction proprietary term artificial by interpreter in described step (2).
Employing it is an advantage of the current invention that:
One, the present invention utilizes the unlimited multiformity of the language meaning of one's words, before written translation, ensureing on the basis that original text each statement meaning is constant, to treat that written translation original text carries out standard clause rewriting, select proprietary terminology bank more manually or mechanically, and apply mechanically the standard clause meeting target language, correctly carry out the mutual deduction of original text and translation implication, thus reaching most basic semantic meaning representation, minimize interpreter's written translation amount, improve written translation efficiency and improve translation accuracy rate, with the minimum written translation achievement manually getting the best in return. Compared with prior art, mechanical assistance function and human thinking's logic can be carried out seamless combination by the present invention, thus by artificial translation amount mechanization simply, translation amount and time are greatly reduced, improve translation efficiency beyond imaginationly.
Two, the present invention is suitable for Office Hierarchy format document, and threshold is low, can process by mass, and interpreter's treating capacity per capita reaches 24 hours 100,000 words.
Three, compatible 60 polyglots of the present invention, it is possible to for the mutual written translation of polyglot, applied widely.
Four, the present invention translate after translation form and unified at text formatter before translating, typesetting difficulty is low.
Five, the present invention carries out data interaction without server, and only computer achieves that the process of document word.
Accompanying drawing explanation
Fig. 1 is the flow chart of the present invention.
Detailed description of the invention
Below to treat that written translation document is for Chinese, translation illustrates specific embodiment of the invention process for English.
A kind of manually in conjunction with the written translation method of machine, comprise the following steps:
(1), treat in written translation document on the constant basis of the statement meaning ensureing, namely treat on the basis that in written translation document, the semanteme of each statement and the tone are constant ensureing, the statement manually treated in written translation document carries out standard clause rewriting, obtain the standard document with standard clause statement, further according to grammatical rules, the position of the proprietary term in standard clause statement each in standard document is identified automatically.
In this step, described proprietary term includes word, phrase and simple sentence, and each standard clause statement in revised standard document is by one or more compositions in word, phrase, simple sentence.
In this step, described standard clause refers to and comprises SVO in statement or subject-predicate determines the clause of guest. The statement treated in written translation document carries out standard clause rewriting and then refers to: in conjunction with each statement meaning in treating written translation document, simplification, elliptical word, phrase or simple sentence in each statement being added in statement, there is SVO or subject-predicate determines the standard clause statement of guest thus being formed. Such as, treat that the statement in written translation document is " Zhang San goes to work in factory ", by analysis, this statement is unsatisfactory for SVO or subject-predicate determines the standard clause of guest, in conjunction with this statement treat written translation document in the meaning, the standard clause of " workman that Zhang San is factory " need to be rewritten into, after rewriting completes, further according to the grammatical rules of English, the position of " Zhang San ", the "Yes" in the workman of factory " Zhang San be ", " factory " and " workman " these proprietary term places is identified automatically by the grammatical rules of English.
It addition, in this step, if the statement treated in written translation document has been standard clause statement, then this statement is no longer needed for rewriting. Such as, treat that the statement in written translation document is " Zhang San is workman ", meet the standard clause of SVO, this statement is standard clause statement, need not rewrite, have only to the grammatical rules according to English, the position at " Zhang San ", "Yes" and " workman " these proprietary term places in " Zhang San is workman " is identified automatically.
(2), manually or used written translation aid to extract the proprietary term in standard document by interpreter, and translate accordingly, obtain proprietary term translation. Such as, by " Zhang San ", the "Yes" in above-mentioned standard clause statement, " factory " and " workman " these proprietary term extraction out after, it is translated, the translation of proprietary term can be obtained.
(3), according to the mark in step (1), proprietary term translation corresponding for proprietary term original text is filled in standard document by mark correspondence, just translation shelves can be obtained. Such as, the translation of above-mentioned " Zhang San ", "Yes", " factory " and " workman " these proprietary terms is automatically filled in standard clause statement corresponding in standard document by mark.
(4), interpreter first translation shelves are examined and revised and after form fine setting according to until written translation document, namely obtain written translation document finished product.
In the present invention, due to original text word to be translated with wait to have between the target text translated mutually corresponding clause storehouse, therefore, the position of the proprietary term in standard clause statement each in standard document is automatically identified referring to by step (1) according to grammatical rules: according to the grammatical rules of target text, the position of the proprietary term in each standard clause statement is identified automatically, so, the standard clause statement being automatically filled in standard document by proprietary term translation accordingly by mark can obtain form and the constant translation of order.
The present invention is in actual written translation process, have only to manually treat written translation document and carry out standard clause rewriting, after rewriting completes, can all be translated by machine, and ensure that translation content is consistent with waiting for translating shelves content, written translation industry is had great progress, there is substantial progress and overturn the huge reform of industry. Meanwhile, the situation that external translation supplementary text handling implement monopolization Chinese Translators' Journal industry was up to 15 years is changed.
Claims (3)
1. one kind manually in conjunction with the written translation method of machine, it is characterised in that comprise the following steps:
(1), treat in written translation document on the constant basis of the statement meaning ensureing, the statement manually treated in written translation document carries out standard clause rewriting, obtain the standard document with standard clause statement, further according to grammatical rules, the position of the proprietary term in standard clause statement each in standard document is identified automatically, wherein, described proprietary term includes word, phrase and simple sentence;
(2), by interpreter extract the proprietary term in standard document, and translate accordingly, obtain proprietary term translation;
(3), according to the mark in step (1), proprietary term translation corresponding for proprietary term original text is filled in standard document by mark, obtains just translation shelves;
(4), interpreter first translation shelves are examined and revised and after form fine setting according to until written translation document, namely obtain written translation document finished product.
2. as claimed in claim 1 a kind of manually in conjunction with the written translation method of machine, it is characterised in that: described step (1) treats statement in written translation document according to semantic and the tone and carries out standard clause rewriting.
3. as claimed in claim 1 a kind of manually in conjunction with the written translation method of machine, it is characterised in that: or use written translation aid extraction proprietary term artificial by interpreter in described step (2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610142419.XA CN105677643A (en) | 2016-03-14 | 2016-03-14 | Translation method combining manpower and machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610142419.XA CN105677643A (en) | 2016-03-14 | 2016-03-14 | Translation method combining manpower and machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105677643A true CN105677643A (en) | 2016-06-15 |
Family
ID=56308268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610142419.XA Pending CN105677643A (en) | 2016-03-14 | 2016-03-14 | Translation method combining manpower and machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105677643A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108519979A (en) * | 2018-02-28 | 2018-09-11 | 成都优译信息技术股份有限公司 | The method and system that translation memory library and MT are combined in a kind of CAT systems |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101196883A (en) * | 2008-01-07 | 2008-06-11 | 刘莎 | Internet information natural language translation general method and system |
| CN101510194A (en) * | 2009-03-15 | 2009-08-19 | 刘树根 | Statement component apparatus and multilingual professional translation method based on statement component |
| CN102609410A (en) * | 2012-04-12 | 2012-07-25 | 传神联合（北京）信息技术有限公司 | Authority file auxiliary writing system and authority file generating method |
| CN102662937A (en) * | 2012-04-12 | 2012-09-12 | 传神联合（北京）信息技术有限公司 | Automatic translation system and automatic translation method thereof |
| CN104331399A (en) * | 2014-07-25 | 2015-02-04 | 一朵云（北京）科技有限公司 | Dictionary tree translation method |
| CN104933041A (en) * | 2015-06-25 | 2015-09-23 | 武汉传神信息技术有限公司 | File extraction and reduction method favorable for translation work |
-
2016
- 2016-03-14 CN CN201610142419.XA patent/CN105677643A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101196883A (en) * | 2008-01-07 | 2008-06-11 | 刘莎 | Internet information natural language translation general method and system |
| CN101510194A (en) * | 2009-03-15 | 2009-08-19 | 刘树根 | Statement component apparatus and multilingual professional translation method based on statement component |
| CN102609410A (en) * | 2012-04-12 | 2012-07-25 | 传神联合（北京）信息技术有限公司 | Authority file auxiliary writing system and authority file generating method |
| CN102662937A (en) * | 2012-04-12 | 2012-09-12 | 传神联合（北京）信息技术有限公司 | Automatic translation system and automatic translation method thereof |
| CN104331399A (en) * | 2014-07-25 | 2015-02-04 | 一朵云（北京）科技有限公司 | Dictionary tree translation method |
| CN104933041A (en) * | 2015-06-25 | 2015-09-23 | 武汉传神信息技术有限公司 | File extraction and reduction method favorable for translation work |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108519979A (en) * | 2018-02-28 | 2018-09-11 | 成都优译信息技术股份有限公司 | The method and system that translation memory library and MT are combined in a kind of CAT systems |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Negri et al. | ESCAPE: a large-scale synthetic corpus for automatic post-editing | |
| CN100437557C (en) | Machine translation method and apparatus based on language knowledge base | |
| Pinnis et al. | Neural machine translation for morphologically rich languages with improved sub-word units and synthetic data | |
| KR20150036041A (en) | Phrase-based dictionary extraction and translation quality evaluation | |
| CN110678868B (en) | Translation support system, translation support apparatus, translation support method, and computer-readable medium | |
| CN105760368B (en) | A kind of deep treatment method of document text | |
| CN112766000B (en) | Machine translation method and system based on pre-training model | |
| CN106156013B (en) | A kind of two-part machine translation method that regular collocation type phrase is preferential | |
| KR20140049150A (en) | Automatic translation postprocessing system based on user participating | |
| Peng | The impact of machine translation and computer-aided translation on translators | |
| Gough et al. | Robust large-scale EBMT with marker-based segmentation | |
| CN105573994B (en) | Statictic machine translation system based on syntax skeleton | |
| CN103793375A (en) | Method for accurately replacing terms and phrases in automatic translation processing | |
| CN105677643A (en) | Translation method combining manpower and machine | |
| Denoual | Analogical translation of unknown words in a statistical machine translation framework | |
| CN111178060A (en) | Korean word segmentation reduction method based on language model | |
| Phuoc et al. | Building a bidirectional english-vietnamese statistical machine translation system by using moses | |
| Tran et al. | Linguistic-relationships-based approach for improving word alignment | |
| CN107590132A (en) | A kind of method of automatic corrigendum segment word is judged by English part of speech | |
| CN103268314A (en) | Method and device for acquiring sentence punctuating rules of Thai language | |
| JP2013054608A (en) | Translation device, method and program, and translation model learning device, method and program | |
| Läubli et al. | Combining statistical machine translation and translation memories with domain adaptation | |
| Zhao et al. | Vietnamese to Chinese machine translation via Chinese character as pivot | |
| CN108537246A (en) | A kind of method and system that parallel corpora is classified by translation quality | |
| Papli | Morpheme-aware subword segmentation for neural machine translation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160615 |