CN100437557C - Machine translation method and apparatus based on language knowledge base - Google Patents
Machine translation method and apparatus based on language knowledge base Download PDFInfo
- Publication number
- CN100437557C CN100437557C CNB2004100011873A CN200410001187A CN100437557C CN 100437557 C CN100437557 C CN 100437557C CN B2004100011873 A CNB2004100011873 A CN B2004100011873A CN 200410001187 A CN200410001187 A CN 200410001187A CN 100437557 C CN100437557 C CN 100437557C
- Authority
- CN
- China
- Prior art keywords
- sentence
- language
- sentence pattern
- translation
- formula
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Machine Translation (AREA)
Abstract
The present invention relates to an English-Chinese machine translation method based on a language knowledge library. The English-Chinese machine translation method is characterized in that according to the characteristics of large-capacity storage and high-speed calculation of a computer, English/Chinese bilingual corpuses are labeled with sentence patterns to establish a bilingual dynamic sentence pattern library; a bilingual conversion mechanism is established by taking the bilingual dynamic sentence pattern library as a core so as to realize automatic machine translation. The present invention also relates to an English-Chinese machine translation which completes translation processes by using the English-Chinese machine translation method.
Description
Technical field:
The present invention relates to the mechanical translation field, particularly based on the machine translation method and the device of language knowledge base.
Background technology:
Utilize computing machine to realize automatic translation technology between the different language, early well known.But because the complicacy of natural language, utilize intertranslation problem between the human different language of computer realization also not have the solution of practicability so far.
Existing machine translation method adopts the method based on rule governing parsing mostly, as " based on the machine translation system and the method for semanteme " (application number 01131689), this method is carried out semantic analysis according to the semantic element representation storehouse to this sentence, thereby obtains this adopted expression formula of sentence; According to the semantic element representation storehouse, the semantic element representation of this adopted expression formula with the purpose languages launched; Sentence after launching is exported as translation.Its essence is and seek a kind of " rule description language ", be used to describe natural language, make it can be by Computer Processing.Because natural language has characteristics such as dirigibility, complicacy and opening, the foundation of rule and to improve difficulty very big.
The another kind of main method that existing interpretation method is adopted is based on the method for corpus statistics, as " English-Chinese translating machine " (application number 00109235), adopts the universal algorithm of matching and replacing templates of any constant and any variable.This method at first will mark the corpus of bilingual alignment, utilizes probabilistic method to set up statistical model then, calculates the conversion of carrying out source language language target language by similarity again.But because the example in the corpus is difficult to contain language phenomenon abundant in the true environment, therefore, this method effect in actual applications is also undesirable.
Summary of the invention:
The purpose of this invention is to provide a kind of machine translation method based on language knowledge base, this method is according to the characteristic of the storage of computing machine high capacity, high-speed computing, language material to different language carries out the sentence pattern mark, set up bilingual dynamic sentence pattern storehouse, be that core is set up bilingual changing the mechanism with the sentence pattern storehouse then, realize automatic translation by computer.This method can provide than the rule and method translation of high-quality more, has again than statistical method adaptability widely.
Another object of the present invention provides a kind of English-Chinese machine translation apparatus based on language knowledge base, and this device utilizes described English-Chinese machine translation method based on language knowledge base to realize automatic translation process.
Machine translation method based on language knowledge base of the present invention may further comprise the steps:
1, sets up dictionary between the different language language, syntax rule storehouse and sentence pattern storehouse; Wherein dictionary is stored the corresponding word of different language language, phrase; The corresponding syntax rule of syntax rule library storage different language language; The corresponding sentence of sentence pattern library storage different language language formula;
2, by input media input source document to be translated;
3, above-mentioned source document is made pauses in reading unpunctuated ancient writings, source document is decomposed into simple sentence one by one;
4, read dictionary information, above-mentioned each simple sentence is carried out cutting, this simple sentence is decomposed into the word or expression of source language;
5, read syntax library information, the simple sentence after the cutting is disappeared hold concurrently and grammatical analysis, form the syntax tree of source language simple sentence;
6, read the sentence pattern library information, according to the syntax tree of source language grammatical analysis formation, with the highest target language sentence formula of matching degree that obtains in the sentence pattern storehouse;
7, according to the sentence formula of above-mentioned target language, generate target language according to the translation rule translation.
Above-mentioned sentence pattern library storage is the translation template from the source language to the target language of unit with the sentence formula, and described sentence formula comprises constant, promptly concrete word, phrase or clause, and variable, promptly replaceable part.
Described sentence pattern storehouse divides three classes to make up, and wherein first kind sentence formula is spoken for having only subject and predicate, or has only the sentence pattern of subject and predicate, object; The second class sentence formula is except subject and predicate, object, also has the simple sentence pattern of other sentence elements, and described other compositions are the adverbial modifier, complement or function word; The 3rd class sentence formula is the complicated sentence pattern beyond the above-mentioned two class sentence patterns.
Described sentence formula matching process according to from the 3rd class sentence pattern to the second class sentence pattern, carry out to the order of first kind sentence pattern again.
Described disappearing held concurrently and syntax analysis step is also discerned and write down and not have full noun or the verb phrase of maybe impossible collection in the dictionary.
English-Chinese machine translation apparatus based on language knowledge base of the present invention, the memory storage that comprises dictionary, syntax rule storehouse and the sentence pattern storehouse of the multilingual language of a storage, wherein dictionary is stored the corresponding word of different language language, phrase, the corresponding syntax rule of syntax rule library storage different language language, the corresponding sentence of sentence pattern library storage different language language formula; The input device of source document; The source document memory storage; Translation engine, and central processing unit are finished translation process with the control translation engine, and wherein translation engine comprises that sentence divides module, read source document and source document is made pauses in reading unpunctuated ancient writings; Cutting and pretreatment module, the dictionary in the read storage device carries out cutting with the simple sentence after dividing; Disappear and hold concurrently and syntax Analysis Module, the syntax library in the read storage device disappears to the simple sentence after the cutting and to hold concurrently and grammatical analysis, forms syntax tree; The sentence pattern matching module, the target language sentence formula that matching degree is the highest in the sentence pattern storehouse according to the syntax tree that syntax Analysis Module forms, is found out in the sentence pattern storehouse in the read storage device; The translation generation module, according to sentence pattern matching module result, translation generates corresponding target language translation; Top control module is controlled each module operation.
Above-mentioned machine translation apparatus based on language knowledge base also comprises the translation storer, is used to store translation; The translation output unit is used to export translation.
Disappearing and holding concurrently and noun or the verb phrase of dictionary for login also discerned and stored to syntax Analysis Module,
Above-mentioned sentence pattern library storage is the translation template from the source language to the target language of unit with the sentence formula, and described sentence formula comprises constant, promptly concrete word, phrase or clause, and variable, promptly replaceable part.
Described sentence pattern storehouse divides three classes to make up, and wherein first kind sentence formula is spoken for having only subject and predicate, or has only the sentence pattern of subject and predicate, object; The second class sentence formula is except subject and predicate, object, and described other compositions of simple sentence pattern that also have other sentence elements are the adverbial modifier, complement, or function word; The 3rd class sentence formula is the complicated sentence pattern beyond the above-mentioned two class sentence patterns.
Described sentence pattern matching module according to from three class sentence patterns to two class sentence patterns, carry out sentence pattern coupling to the order of a class sentence pattern again.
Description of drawings:
Below the most preferred embodiment that present invention will be described in detail with reference to the accompanying.
Fig. 1 is an English-Chinese machine translation apparatus architectural schematic of the present invention;
Fig. 2 is the translation engine synoptic diagram in the English-Chinese machine translation apparatus of the present invention;
Fig. 3 is the language knowledge base structural representation in the English-Chinese machine translation apparatus of the present invention;
Fig. 4 is the sentence pattern synoptic diagram in the English-Chinese machine translation apparatus of the present invention;
Fig. 5 is the sentence pattern library structure synoptic diagram in the English-Chinese machine translation apparatus of the present invention;
Fig. 6 is the sentence pattern matching module synoptic diagram in the English-Chinese machine translation apparatus of the present invention;
Fig. 7 is the schematic flow sheet of English-Chinese machine translation method of the present invention;
Fig. 8 is the cutting step synoptic diagram in the English-Chinese machine translation method of the present invention;
Fig. 9 is the syntax analysis step synoptic diagram in the English-Chinese machine translation method of the present invention;
Figure 10 is the sentence pattern coupling step synoptic diagram in the English-Chinese machine translation method of the present invention.
Embodiment:
As shown in Figure 1, embodiment is an English-Chinese machine translation apparatus of the present invention, comprises a linguistry storehouse memorizer 2, is used to store the language knowledge base 1 that comprises dictionary, syntax rule storehouse and sentence pattern storehouse; A text is imported and display device 3, is used for input and shows text to be translated; A central processing unit 4 is used to utilize translation engine 6 to finish translation process; And output of translation result and display device 5, be used for output and show translation.
As shown in Figure 2, a translation engine most preferred embodiment comprises that top control module 7 is used to manage and control the work of each module of translation engine; Module 8 divided in sentence, is used for English sentence branch to be translated is broken into character string; Cutting and pretreatment module 9, being used for an english sentence is cut into the phrase is the character string sequence of unit; Disappear and hold concurrently and syntax Analysis Module 10, be used for, make the english sentence after the cutting form syntax tree, and then carry out sentence pattern and mate by eliminating ambiguous category and relative simple grammatical analysis; Sentence pattern matching module 11 is used for the sentence pattern in source language and sentence pattern storehouse is mated, thereby realizes mechanical translation; And translation generation module 12, be used on the basis of sentence pattern coupling, generating translation.
As shown in Figure 3A, the most preferred embodiment of a language knowledge base 1 comprises that a dictionary 13 is used to store the english-chinese bilingual dictionary; A syntax library 14 is used to store syntax rule, for cutting, disappear and hold concurrently and use during grammatical analysis; And a sentence pattern storehouse 15, be used to store sentence pattern, use when mating for sentence pattern.Shown in Fig. 3 B, the entry of preserving in the dictionary 13 marks by the requirement of translation system, has indicated relevant semantic attribute.Shown in Fig. 3 C, the syntax rule of preserving in the syntax library 14 has been stipulated the translation rule of word or phrase according to the requirement of translation system.Shown in Fig. 3 D, the sentence pattern of preserving in the sentence pattern storehouse 15 marks English sentence and standard translation thereof according to the requirement of translation system sentence pattern mark.
As shown in Figure 4, the most preferred embodiment of sentence pattern is to be the translation template from the English source language to the Chinese target language of unit with the sentence, and it is the sequence that variable is formed by constant (speech, phrase or clause) and replaceable part.At first define the sentence pattern symbol when making sentence pattern, such as definition { MAN1} represents the people, { N[I] } express time speech etc.Sentence pattern symbol and English word have been formed sentence pattern.
Reason such as: that is why{SV}--＞Here it is %1
MAN1}{blame}{MAN2}for{Y}--＞%1%2[] %4 is attributed to %3
In sentence pattern, the part in { } is that replaceable part is a variable, and { } part in addition is a constant.When keeping verb prototype in { }, as: blame}, and represent its variable for the various tenses (present tense, past tense, future tense) of this verb, voice (by dynamically, carry out attitude, finish attitude) and the tone (subjunctive mood) form.
Sometimes sentence pattern can be made up of constant fully, as: Speaking by telephone--＞in telephone talk.
As shown in Figure 5, the most preferred embodiment in a sentence pattern storehouse 15 divides three classes to make up, and wherein a class sentence pattern storehouse 16 is preserved and had only the subject and predicate language, or has only the sentence pattern of subject and predicate, object; Preserve except subject and predicate, object in two class sentence pattern storehouses 17, also has the simple sentence pattern of other sentence elements (as the adverbial modifier, attribute, function word); The complicated sentence pattern beyond the above-mentioned two class sentence patterns is preserved in three class sentence pattern storehouses 18.
As shown in Figure 6, the most preferred embodiment of a translation engine sentence pattern matching module, its sentence pattern matching process according to from three class sentence patterns to two class sentence patterns, carry out to the order of a class sentence pattern again.
As shown in Figure 7, a most preferred embodiment of English-Chinese translation method of the present invention comprises, is that source language text is made pauses in reading unpunctuated ancient writings by unit with the sentence, and the sentence after will making pauses in reading unpunctuated ancient writings based on the english-chinese bilingual dictionary again carries out cutting; Sentence after cutting handled disappears and holds concurrently and grammatical analysis; Offset the syntax tree that forms after double and the grammatical analysis according to the sentence pattern storehouse then and carry out the sentence pattern coupling, sentence translation is also exported according to the translation of the sentence pattern on the coupling.
As shown in Figure 8, the most preferred embodiment of the cutting step of English-Chinese translation method of the present invention adopts maximum matching method, that is: earlier the character string of importing is scanned, what this technology adopted is forward scan, and promptly first word is taken out in from left to right scanning, in the cutting dictionary, search this speech then, if can find this speech or can find with this speech is the phrase of initial speech, just get second word again, go to consult the dictionary with first combinations of words together, so circulation, up to can not finding this speech in the cutting dictionary, and can not to find with this speech be the phrase of initial speech, at this moment, just begin backtracking, remove last word of this speech earlier, in the cutting dictionary, search this speech then, if can find this speech, then first phrase has just cut out, otherwise,, consult the dictionary again with regard to removing last word of this speech again.After cutting out first phrase, repeat above process, up to the word that a sentence is cut into one by one.With The United States is very big. is example, and the participle program is read in the The word earlier, consults the dictionary, can find, that is to say that independent The word is a speech, but owing to be maximum matching method, so can not think that The is exactly that a speech reads in the United word again in this sentence, form The United, consult the dictionary, can not find, but there is phrase in the dictionary with The United beginning, read in States again, form The UnitedStates and consult the dictionary, can find, read in is again, form The United States is, consult the dictionary, can not find, and the speech with The United States is beginning does not have yet, coupling has been to the end so forward, needs backtracking then, removes last is, look into again, so repeatedly, find out real speech, last word segmentation result is The United States/is/very/big/.
As shown in Figure 9, the most preferred embodiment purpose of the syntax analysis step of English-Chinese translation method of the present invention is by simple relatively grammatical analysis, makes the english sentence after the cutting form syntax tree, and then carries out the sentence pattern coupling.Its effect also is by not having maybe can not collect complete noun or verb phrase in the simple lexical analysis identification dictionary.Such as so structure is arranged: the U.S-led war in Iraq, at first, this structure can be placed in many sentences, it all is equivalent to a noun composition, if included this speech in the dictionary, cutting procedure can correctly cut out it, and can translate translation exactly: the war in Iraq of U.S. leader.But, if do not include this structure in the dictionary, and only included war in Iraq/ the war in Iraq, in this case, a kind of method adds this speech the U.S.-led war in Iraq exactly, also having a kind of method is exactly lexical analysis, what adopt is the method for rule match: such as so rule is arranged: definite article+adjective+noun==noun phrase, and the U.S.-led war in Iraq, just in time can mate this rule, so this noun phrase has just identified.
As shown in figure 10, the most preferred embodiment of the sentence pattern of English-Chinese translation method of the present invention coupling step according to from three class sentence patterns to two class sentence patterns, carry out to the order of a class sentence pattern again.For example sentence: " We cannot rest onour success sofar.SARS may return and we should be ready for it; " said Shigeru Omi, the WHO ' s regional director for the Western Pacific.
Through punctuate, pre-service and cutting, disappear double and syntax analysis step, the form below forming:
″We【cannot〔rest?on〕】【our?success?】〔so?far〕.SARS【may?return】and?we【should?〔be?ready?for〕】it，″said〔Shigeru?Omi〕，【〖the?WHO’sregional?director〗for?〔the?western?pacific〕】
With full sentence search sentence pattern storehouse, mate last the 3rd class sentence pattern then:
I] }, { N[A] ﹠amp; ﹠amp; S_SEM[B|D] }--＞%3%2 claims, " %1 " ∥ weight=-52
Wherein, x} is corresponding to We[cannot (rest on)] and [our success] (so far) .SARS[may return] and we[should (be ready for)] it; Said is corresponding to said; I] } corresponding to (Shigeru Omi), { N[A] ﹠amp; ﹠amp; S SEM[B|D] } corresponding to [" the WHO ' s regionaldirector " for (the western pacific)].
Continue in the sentence pattern storehouse, to search for first subordinate sentence of front with subordinate sentence then:
we【cannot〔rest?on〕】【our?success】〔so?far〕.SARS?【may?return】and?we【should?〔be?ready?for〕】it.
Mate last the 3rd class sentence pattern:
I] } so far{X}--＞up to now, %TRAN[%1, %2, %3, %4] //weight=-38
Every corresponding relation is as follows:
I] } corresponding to We; { V} is corresponding to [cannot (rest on)]; N[! I] } corresponding to [our success]; So far is corresponding to (so far); X} is corresponding to SARS[may return] and and we[should (be ready for)] it.
Continue the search clause:
We【cannot〔rest?on〕】【our?success?】.SARS【may?return】and?we?【should〔be?ready?for〕】it.
Mate last the 3rd class sentence pattern:
{X}.{X}--＞％1。％2//weight＝-9
Every corresponding relation is as follows:
X} is corresponding to We[cannot (rest on)] [our success]; X} is corresponding to SARS[mayreturn] and and we[should (be ready for)] it.
Continue the search clause:
I] }
{VOI[A]&&E_VAL[1]}
{MAN2||N[！I]}--＞％1％VOF[％2，％3]//weight＝-15
Every corresponding relation is as follows:
I} is corresponding to We; { VOI[A] ﹠amp; ﹠amp; E_VAL[1] } corresponding in [cannot (rest on)]; I] } corresponding to [our success].
Continue the search clause: SARS[may return] sentence pattern on and we [should (be ready for)] the it coupling:
I] } and VO}{and}{MAN1}{VO}.--＞%TRAN[%1, %2], and %TRAN[%4, %5].//weight＝-27
Every corresponding relation is as follows:
I] corresponding to SARS; { VO} is corresponding to [may return]; { and} is corresponding to and; { MAN1} is corresponding to we; { VO} is corresponding to [should (be ready for)] it.
Continue the search clause:
SARS【may?return】we【should〔be?ready?for〕】it
Sentence pattern on the coupling:
I}--＞%1%2[is] %3 is ready //weight=-25
Every corresponding relation is as follows:
I] } corresponding to we; { be ready for} is corresponding to [should (be ready for)]; I] } corresponding to it.
According to the relevant translation rule of each sentence pattern, the result of sentence pattern coupling is translated into the translation of target language, the translation result of last above-mentioned source language is:
The regional responsible official's Shigeru Omi of the World Health Organization (WHO) of being responsible for the West Pacific region claims that " up to now, we can not rest in our success.SARS may return, and we should be ready for it.”
More than by specific embodiments of the invention principle of the present invention and feature are described.Be to be understood that the present invention is not limited only to above-mentioned specific embodiment, multiple variation can also be arranged, and concrete implementation step also can be had any different.Protection scope of the present invention is only defined by the appended claims.
Claims (10)
1. machine translation method based on language knowledge base may further comprise the steps:
1) sets up dictionary between the different language language, syntax rule storehouse and sentence pattern storehouse; Wherein dictionary is stored the corresponding word of different language language, phrase; The corresponding syntax rule of syntax rule library storage different language language; The corresponding sentence of sentence pattern library storage different language language formula;
2) by input media input source document to be translated;
3) above-mentioned source document is made pauses in reading unpunctuated ancient writings, source document is decomposed into simple sentence one by one;
4) read dictionary information, above-mentioned each simple sentence is carried out cutting, this simple sentence is decomposed into the word or expression of source language;
5) read syntax library information, the simple sentence after the cutting is disappeared hold concurrently and simple grammatical analysis, form the syntax tree of source language simple sentence;
6) read the sentence pattern library information, the syntax tree according to the source language grammatical analysis forms obtains the highest target language sentence formula of matching degree in the sentence pattern storehouse;
7) according to the sentence formula of above-mentioned target language, generate target language according to the translation rule translation.
2. the machine translation method based on language knowledge base as claimed in claim 1, it is characterized in that described sentence pattern library storage is the translation template from the source language to the target language of unit with the sentence formula, described sentence formula comprises constant, promptly concrete word, phrase or clause, and variable, promptly replaceable part.
3. the machine translation method based on language knowledge base as claimed in claim 1 or 2 is characterized in that described sentence pattern storehouse divides three classes to make up, and wherein first kind sentence formula is spoken for having only subject and predicate, or has only the sentence pattern of subject and predicate, object; The second class sentence formula is except subject and predicate, object, also has the simple sentence pattern of other sentence elements, and described other compositions are the adverbial modifier, complement or function word; The 3rd class sentence formula is the complicated sentence pattern beyond the above-mentioned two class sentence patterns.
4. the machine translation method based on language knowledge base as claimed in claim 3, it is characterized in that described sentence formula matching process according to from the 3rd class sentence pattern to the second class sentence pattern, carry out to the order of first kind sentence pattern again.
5. the machine translation method based on language knowledge base as claimed in claim 1 is characterized in that described disappearing hold concurrently and syntax analysis step is also discerned and write down and not have the full noun phrase of maybe impossible collection in the dictionary.
6. machine translation apparatus based on language knowledge base, the memory storage that comprises dictionary, syntax rule storehouse and the sentence pattern storehouse of the multilingual language of a storage, wherein dictionary is stored the corresponding word of different language language, phrase, the corresponding syntax rule of syntax rule library storage different language language, the corresponding sentence of sentence pattern library storage different language language formula; The input device of source document; The source document memory storage; Translation engine, and central processing unit are finished translation process with the control translation engine, and wherein translation engine comprises that sentence divides module, read source document and source document is made pauses in reading unpunctuated ancient writings; Cutting and pretreatment module, the dictionary in the read storage device carries out cutting with the simple sentence after dividing; Disappear and hold concurrently and syntax Analysis Module, the syntax library in the read storage device disappears double and simply grammatical analysis to the simple sentence after the cutting, forms syntax tree; The sentence pattern matching module, the target language sentence formula that matching degree is the highest in the sentence pattern storehouse according to the syntax tree that syntax Analysis Module forms, is found out in the sentence pattern storehouse in the read storage device; The translation generation module, according to sentence pattern matching module result, translation generates corresponding target language translation; Top control module is controlled each module operation.
7. the machine translation apparatus based on language knowledge base as claimed in claim 6, it is characterized in that described sentence pattern library storage is the translation template from the source language to the target language of unit with the sentence formula, described sentence formula comprises constant, promptly concrete word, phrase or clause, and variable, promptly replaceable part.
8. as claim 6 or 7 described machine translation apparatus based on language knowledge base, it is characterized in that described sentence pattern storehouse divides three classes to make up, wherein first kind sentence formula is spoken for having only subject and predicate, or has only the sentence pattern of subject and predicate, object; The second class sentence formula is except subject and predicate, object, and described other compositions of simple sentence pattern that also have other sentence elements are the adverbial modifier, complement, or function word; The 3rd class sentence formula is the complicated sentence pattern beyond the above-mentioned two class sentence patterns.
9. the machine translation apparatus based on language knowledge base as claimed in claim 8, it is characterized in that described sentence pattern matching module according to from three class sentence patterns to two class sentence patterns, carry out sentence pattern coupling to the order of a class sentence pattern again.
10. the machine translation apparatus based on language knowledge base as claimed in claim 6 is characterized in that described disappearing hold concurrently and the uncollected noun phrase of dictionary is also discerned and write down to syntax Analysis Module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100011873A CN100437557C (en) | 2004-02-04 | 2004-02-04 | Machine translation method and apparatus based on language knowledge base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100011873A CN100437557C (en) | 2004-02-04 | 2004-02-04 | Machine translation method and apparatus based on language knowledge base |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1652106A CN1652106A (en) | 2005-08-10 |
| CN100437557C true CN100437557C (en) | 2008-11-26 |
Family
ID=34867058
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100011873A Expired - Fee Related CN100437557C (en) | 2004-02-04 | 2004-02-04 | Machine translation method and apparatus based on language knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100437557C (en) |
Families Citing this family (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100437557C (en) * | 2004-02-04 | 2008-11-26 | 北京赛迪翻译技术有限公司 | Machine translation method and apparatus based on language knowledge base |
| CN101131689B (en) * | 2006-08-22 | 2010-08-18 | 苗玉水 | Bidirectional mechanical translation method for sentence pattern conversion between Chinese language and foreign language |
| CN100543727C (en) * | 2006-12-21 | 2009-09-23 | 中国科学院计算技术研究所 | A kind of interpretation method that has merged sentence pattern template and statistical machine translation technology |
| CN101201820B (en) * | 2007-11-28 | 2010-06-02 | 北京金山软件有限公司 | Method and system for filtering bilingualism corpora |
| CN101303692B (en) * | 2008-06-19 | 2012-08-29 | 徐文和 | All-purpose numeral semantic library for translation of mechanical language |
| JP5501581B2 (en) * | 2008-07-25 | 2014-05-21 | シャープ株式会社 | Information processing apparatus and information processing method |
| CN101770458A (en) * | 2009-01-07 | 2010-07-07 | 三星电子（中国）研发中心 | Mechanical translation method based on example phrases |
| CN101739395A (en) * | 2009-12-31 | 2010-06-16 | 程光远 | Machine translation method and system |
| CN101777044B (en) * | 2010-01-29 | 2012-07-25 | 中国科学院声学研究所 | System for automatically evaluating machine translation by using sentence structure information and implementing method |
| CN102214166B (en) * | 2010-04-06 | 2013-02-20 | 三星电子（中国）研发中心 | Machine translation system and machine translation method based on syntactic analysis and hierarchical model |
| CN102567310B (en) * | 2010-12-17 | 2016-08-31 | 张龙哺 | Networking artificial intelligence's translation system based on Intelligence repository and interpretation method thereof |
| CN102789464B (en) * | 2011-05-20 | 2017-11-17 | 陈伯妤 | Natural language processing methods, devices and systems based on semantics identity |
| CN103793375A (en) * | 2012-10-31 | 2014-05-14 | 上海勇金懿信息科技有限公司 | Method for accurately replacing terms and phrases in automatic translation processing |
| CN103116578A (en) * | 2013-02-07 | 2013-05-22 | 北京赛迪翻译技术有限公司 | Translation method integrating syntactic tree and statistical machine translation technology and translation device |
| CN103714053B (en) * | 2013-11-13 | 2017-05-10 | 北京中献电子技术开发中心 | Japanese verb identification method for machine translation |
| CN104679735A (en) * | 2013-11-30 | 2015-06-03 | 赵会军 | Pragmatic machine translation method |
| CN103823796A (en) * | 2014-02-25 | 2014-05-28 | 武汉传神信息技术有限公司 | System and method for translation |
| CN105320650B (en) * | 2014-07-31 | 2019-03-26 | 崔晓光 | A kind of machine translation method and its system based on corpus matching and syntactic analysis |
| CN106383819A (en) * | 2016-01-11 | 2017-02-08 | 陈勇 | Speech convertor |
| CN107783965B (en) * | 2016-08-31 | 2021-07-02 | 深圳市青锋全能教育培训有限公司 | Sentence pattern structure-based translation method and device |
| CN107943794A (en) * | 2016-10-12 | 2018-04-20 | 阿里巴巴集团控股有限公司 | A kind of interpretation method and system |
| CN107451114A (en) * | 2017-06-28 | 2017-12-08 | 广州尚恩科技股份有限公司 | A kind of archaic Chinese semantic analysis and its system |
| CN108255818B (en) * | 2018-01-23 | 2021-08-17 | 广州番禺职业技术学院 | Combined machine translation method using segmentation technology |
| CN108153743B (en) * | 2018-01-23 | 2021-12-17 | 甲骨易(北京)语言科技股份有限公司 | Intelligent off-line translation machine based on similarity |
| CN108280066B (en) * | 2018-01-23 | 2021-09-24 | 江苏省舜禹信息技术有限公司 | Off-line translation method from Chinese to English |
| CN108491397A (en) * | 2018-03-26 | 2018-09-04 | 安阳师范学院 | A kind of English Translation electronic system |
| CN108563644A (en) * | 2018-03-29 | 2018-09-21 | 河南工学院 | A kind of English Translation electronic system |
| CN109740168B (en) * | 2019-01-09 | 2020-10-13 | 北京邮电大学 | Traditional Chinese medicine classical book and ancient sentence translation method based on traditional Chinese medicine knowledge graph and attention mechanism |
| CN111931524B (en) * | 2020-07-15 | 2023-08-08 | 北京百度网讯科技有限公司 | Method, apparatus, device and storage medium for outputting information |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5640575A (en) * | 1992-03-23 | 1997-06-17 | International Business Machines Corporation | Method and apparatus of translation based on patterns |
| JPH1166069A (en) * | 1997-08-12 | 1999-03-09 | Ricoh Co Ltd | Machine translation device |
| CN1212406A (en) * | 1997-09-23 | 1999-03-31 | 金旗 | Automatic accurate translating system for computer |
| CN1212407A (en) * | 1997-09-22 | 1999-03-31 | 信快达电脑公司 | Method for translating cultural subtleties in machine translation |
| US6182026B1 (en) * | 1997-06-26 | 2001-01-30 | U.S. Philips Corporation | Method and device for translating a source text into a target using modeling and dynamic programming |
| CN1067782C (en) * | 1997-07-02 | 2001-06-27 | 华建机器翻译有限公司 | Inference technology on imperfect knowledge |
| US20010029443A1 (en) * | 2000-03-27 | 2001-10-11 | International Business Machines Corporation | Machine translation system, machine translation method, and storage medium storing program for executing machine translation method |
| US20030018469A1 (en) * | 2001-07-20 | 2003-01-23 | Humphreys Kevin W. | Statistically driven sentence realizing method and apparatus |
| CN1108572C (en) * | 1992-03-23 | 2003-05-14 | 日本电气株式会社 | Mechanical Chinese to japanese two-way translating machine |
| US20030233226A1 (en) * | 2002-06-07 | 2003-12-18 | International Business Machines Corporation | Method and apparatus for developing a transfer dictionary used in transfer-based machine translation system |
| CN1652106A (en) * | 2004-02-04 | 2005-08-10 | 北京赛迪翻译技术有限公司 | Machine translation method and apparatus based on language knowledge base |
-
2004
- 2004-02-04 CN CNB2004100011873A patent/CN100437557C/en not_active Expired - Fee Related
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5640575A (en) * | 1992-03-23 | 1997-06-17 | International Business Machines Corporation | Method and apparatus of translation based on patterns |
| CN1108572C (en) * | 1992-03-23 | 2003-05-14 | 日本电气株式会社 | Mechanical Chinese to japanese two-way translating machine |
| US6182026B1 (en) * | 1997-06-26 | 2001-01-30 | U.S. Philips Corporation | Method and device for translating a source text into a target using modeling and dynamic programming |
| CN1067782C (en) * | 1997-07-02 | 2001-06-27 | 华建机器翻译有限公司 | Inference technology on imperfect knowledge |
| JPH1166069A (en) * | 1997-08-12 | 1999-03-09 | Ricoh Co Ltd | Machine translation device |
| CN1212407A (en) * | 1997-09-22 | 1999-03-31 | 信快达电脑公司 | Method for translating cultural subtleties in machine translation |
| CN1212406A (en) * | 1997-09-23 | 1999-03-31 | 金旗 | Automatic accurate translating system for computer |
| US20010029443A1 (en) * | 2000-03-27 | 2001-10-11 | International Business Machines Corporation | Machine translation system, machine translation method, and storage medium storing program for executing machine translation method |
| US20030018469A1 (en) * | 2001-07-20 | 2003-01-23 | Humphreys Kevin W. | Statistically driven sentence realizing method and apparatus |
| US20030233226A1 (en) * | 2002-06-07 | 2003-12-18 | International Business Machines Corporation | Method and apparatus for developing a transfer dictionary used in transfer-based machine translation system |
| CN1652106A (en) * | 2004-02-04 | 2005-08-10 | 北京赛迪翻译技术有限公司 | Machine translation method and apparatus based on language knowledge base |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1652106A (en) | 2005-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100437557C (en) | Machine translation method and apparatus based on language knowledge base | |
| CN110543644B (en) | Machine translation method and device containing term translation and electronic equipment | |
| Saad et al. | Osac: Open source arabic corpora | |
| Pettersson et al. | An SMT approach to automatic annotation of historical text | |
| KR101762866B1 (en) | Statistical translation apparatus by separating syntactic translation model from lexical translation model and statistical translation method | |
| KR101266361B1 (en) | Automatic translation system based on structured translation memory and automatic translating method using the same | |
| CN103116578A (en) | Translation method integrating syntactic tree and statistical machine translation technology and translation device | |
| Kumar et al. | Factored Statistical Machine Translation System for English to Tamil Language. | |
| Ranathunga et al. | Si-ta: Machine translation of sinhala and tamil official documents | |
| KR101794274B1 (en) | Method and apparatus for filtering translation rules and generating target word in hierarchical phrase-based statistical machine translation | |
| WO1997040452A1 (en) | Automated natural language translation | |
| CN113343717A (en) | Neural machine translation method based on translation memory library | |
| CN107491441B (en) | Method for dynamically extracting translation template based on forced decoding | |
| Stepanov et al. | Language style and domain adaptation for cross-language SLU porting | |
| CN108255818B (en) | Combined machine translation method using segmentation technology | |
| Wang et al. | High-quality speech-to-speech translation for computer-aided language learning | |
| Siahbani et al. | Simultaneous translation using optimized segmentation | |
| Gu et al. | Concept-based speech-to-speech translation using maximum entropy models for statistical natural concept generation | |
| CN112085985B (en) | Student answer automatic scoring method for English examination translation questions | |
| CN114861628A (en) | System, method, electronic device and storage medium for training machine translation model | |
| Singh et al. | English-Dogri Translation System using MOSES | |
| Dhore et al. | Survey on machine transliteration and machine learning models | |
| JP2006004366A (en) | Machine translation system and computer program for it | |
| CN108153743B (en) | Intelligent off-line translation machine based on similarity | |
| CN108280066B (en) | Off-line translation method from Chinese to English |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CI01 | Correction of invention patent gazette |
Correction item: Inventor Correct: Liu Honglie|Xu Xiaolan|Song Jinping|Xiao Jian|Yuan Qi False: Song Jinping Number: 48 Page: 1022 Volume: 24 |
|
| CI03 | Correction of invention patent |
Correction item: Inventor Correct: Liu Honglie|Xu Xiaolan|Song Jinping|Xiao Jian|Yuan Qi False: Song Jinping Number: 48 Page: The title page Volume: 24 |
|
| ERR | Gazette correction |
Free format text: CORRECT: INVENTOR; FROM: SONG JINPING TO: LIU HONGLIE XU XIAOLAN SONG JINPING XIAO JIAN - YUAN QI |
|
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081126 Termination date: 20120204 |