CN101957815A - Automatic translation method and system based on correct translation result and corresponding relation - Google Patents
Automatic translation method and system based on correct translation result and corresponding relation Download PDFInfo
- Publication number
- CN101957815A CN101957815A CN2009100546908A CN200910054690A CN101957815A CN 101957815 A CN101957815 A CN 101957815A CN 2009100546908 A CN2009100546908 A CN 2009100546908A CN 200910054690 A CN200910054690 A CN 200910054690A CN 101957815 A CN101957815 A CN 101957815A
- Authority
- CN
- China
- Prior art keywords
- translation
- language material
- target language
- corresponding relation
- authentication
- 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
Images
Landscapes
- Machine Translation (AREA)
Abstract
The invention discloses an automatic translation method based on a correct translation result and corresponding relation, comprising the following steps of: receiving translation briefs: source languages, source language materials, target languages; searching whether the source language materials and target language materials exist or can be combined or not in a multilingual corresponding relation database; if so, showing the target language materials; if not, selecting machine translation or human translation; when the machine translation is selected, acquiring a machine translation result, and then submitting human authentication, translation and analysis; when the human translation is selected, directly submitting the human authentication, translation and analysis; acquiring correct target language materials through the human authentication, translation and analysis; recording the target language materials in the multilingual corresponding relation database; and showing the target language materials.
Description
Technical field
The present invention relates to a kind of interpretation method and device, relate in particular to a kind of multilingual professional automatic translating method and device.
Background technology
The world today is the society of an advanced IT application.Along with the making rapid progress of the infotech headed by the internet, we can be in moment, and is not subjected to geographic far and near restriction, obtains the information from all over the world.But a lot of information wherein are not to represent with your mother tongue or the language that you are familiar with.People have obtained more communication method and channel.This makes that also the influence of aphasis is more obvious, therefore overcomes aphasis and realizes that translation is very urgent automatically.
Now the machine translation tools that much enters into the trial period has been arranged, web page translation service has based on this been arranged on the internet.These machine translation tools all are based on rule (for example by to the parsing of grammer, to the parsing of the meaning etc.) and infer translation result.Their greatest drawback is the correctness that can not guarantee translation result.Colloquial translation brief, the easy more mistake that occurs.Therefore the result who utilizes these translation tools to translate out can only improve the aid of translation efficiency as semi-manufacture as the Professional translator.Be ignorant of the people of target language for those, owing to can't judge the correctness of translation result, so they can't use this class translation tool.From another angle, this class translation tool does not still possess business-like prospect.
For common user (translating the user automatically), the hope of their maximum is to obtain correct translation result.
Therefore, have a kind of technical need in the industry: provide a kind of be applicable to multilingual between intertranslation, guarantee correct automatic translating method and the system of translation result.
Summary of the invention
The purpose of this invention is to provide a kind of automatic translating method and system based on correct translation result and corresponding relation, technical scheme of the present invention is as follows:
A kind of automatic translating method based on correct translation result and corresponding relation is characterized in that, comprises following steps:
Receive translation brief, translation brief comprises: source languages, source language material, target language;
Whether retrieval source language material exists in multilingual corresponding relation database, if exist, judges then whether the authentication result of target language is OK, if then present the target language material;
If not, judge then whether the statement type of source language material and type speech exist,, judge then whether the statement type of target language and the authentication result of type speech are OK, if then present the target language material if exist;
If the source language material does not exist and the statement type of source language material or the type speech does not exist or the statement type of target language or type speech do not exist, then select mechanical translation or human translation; When the selector device is translated, translation brief is submitted to machine translator, obtain the mechanical translation result then, submit mutual authentication, translation and decomposition afterwards to, and with the outcome record of mutual authentication, decomposition in multilingual corresponding relation database; When selecting human translation, directly submit mutual authentication, translation and decomposition to, and with the outcome record of human translation and decomposition in multilingual corresponding relation database.
A kind of automatic translation system based on correct translation result and corresponding relation is characterized in that, comprises:
Multilingual corresponding relation database is preserved one of following at least information: source languages, source language material, target language, target language material, language material attribute, statement rule of combination, whether authenticate OK, the credit level of authentication, authentication people's credit level;
Receiver module receives translation brief;
Retrieval module, whether retrieval source language material, target language material exist in multilingual corresponding relation database, and whether corresponding authentication OK; Whether retrieve statement type and type speech exist;
Express module, present the target language material; For combined target language material, the requirement according to the statement rule of combination of target language behind the type word combination of the statement type of target language and target language, presents;
Module for reading and writing is each module reading and writing database;
Mutual authentication, translation and decomposing module obtain correct translation result, and the language material that can be decomposed into statement type and type speech is decomposed;
Interface module is for receiver module with express module interface with external unit is provided.
Advantage of the present invention is:
1. use certified translation content, guarantee that translation result is correct;
2. the increase of language material relies on the accumulation of using the user, so accumulating rate is fast, is convenient to commercialization;
3. the every increase of system is once used and will be added up language materials more or increase credit rating more, therefore has the growth ability;
4. language material increases the use that relies on the user, does not therefore need to spend special typing manpower, saves artificial;
5. the mode that receives translation brief is various, the convenient use;
6. present translation result and can be convenient to utilization and extention of the present invention for multiple modes such as website, phone, mobile phone, PDA, computers.
Description of drawings
Below in conjunction with accompanying drawing,, will make technical scheme of the present invention and other beneficial effects apparent by detailed description to the specific embodiment of the present invention.
Fig. 1 is the schematic flow sheet that the present invention is based on the automatic translating method of correct translation result and corresponding relation;
Fig. 2 is the composition synoptic diagram that the present invention is based on the automatic translation system of correct translation result and corresponding relation.
Embodiment
In order further to understand feature of the present invention, see also following about detailed description of the present invention and accompanying drawing, yet appended graphic only provide with reference to the usefulness of explanation, be not to be used for protection scope of the present invention is limited.
As shown in Figure 1, be the schematic flow sheet of the automatic translating method that the present invention is based on correct translation result and corresponding relation.Step 101 at first: receive translation brief.Receive mode can be for one of following: website, phone, mobile phone, PDA, computer.Translation brief specifically comprises: source languages, source language material, target language.The alleged language material of the present invention is: word, statement, phrase or paragraph.
Execution in step 102 then, and whether retrieval source language material exists in multilingual corresponding relation database.If the source language material exists, execution in step 103 judges whether the authentication result of target language is OK, if, the target language material that has target language in the multilingual corresponding relation database then is described, then execution in step 104, present the target language material.The statement type of source language material and type speech do not exist if the source language material does not exist, judge then whether target language exists statement type and type speech, if all exist, then present the target language material, the target language material of this moment is determined by statement type, type speech and statement organization regulation three parts.That is to say that language material of the present invention has two kinds of manifestation modes: complete language material form; The form of statement type+type speech, for a kind of mode in back, we are called combined language material, promptly according to the requirement of statement rule of combination, statement type and the synthetic complete language material of type phrase, present.
Such as: complete language material is: I will go to the pudong airport.The statement type of combined language material is: go; The type speech is me, the pudong airport.If there is not " I will go to the pudong airport " this language material in the database, but the type that in statement type, has " will go ", in the type speech of person, " I " speech is arranged, " pudong airport " speech is arranged in the type speech of place name, according to the statement rule of combination: it is called+go+place name; This language material just is combined as: I will go to the pudong airport.
For another example, complete language material is: I came off duty at 5 o'clock.The statement type of combined language material is: come off duty; The type speech of person is me; The type speech of time was 5 o'clock.If there is not " I came off duty at 5 o'clock " this language material in the database, but the type that in statement type, has " coming off duty ", in the type speech of person, have the speech of " I ", in the type speech of time, have the speech at " 5 o'clock ", according to the statement rule of combination: the person+time+next; This language material just is combined as: I came off duty at 5 o'clock.
By above-mentioned two examples as can be seen, according to the statement type of storing in the database, type speech and statement rule of combination, also can be combined into the target language material even without the target language material.
In an embodiment of the present invention, the target language material in the database is presented to the user can the multiple mode that presents, and for example: webpage presents, the mobile phone literal presents, voice present, PDA presents, computer presents.In the mode that voice present, have a multilingual words in the multilingual corresponding relation database and read aloud instrument, therefore the target language material can be presented with voice.
When can not find the target language material and also can't the composite object language material time, then execution in step 105, select human translation or mechanical translation by the user.If selection result is a mechanical translation, then execution in step 106, and translation brief is submitted to machine translator.The alleged machine translator of the present invention provides the website or the software of automatic translation for all, for example: google translation, Yahoo translation, MSN translation etc.
Execution in step 107 then, obtain the mechanical translation result.
Execution in step 108 then, submit mutual authentication, translation to and decompose, and whether specifically estimate the mechanical translation result by the authentication people correct, revise translation result till correctly.The artificial authentication of the present invention can be a personal authentication, also can be that a plurality of people authenticate, revise successively (coming across under the not accurate enough situation of translation result under the situation that a plurality of people revise) successively.The present invention decomposes the language material that can be decomposed into statement type and type speech.The method of decomposing is: complete language material is decomposed into: statement type and type speech are stored in the database respectively.Has the statement rule of combination in the statement type, it is the declarative statement type, the type speech of stipulating what type could make up and built-up sequence, such as: " I will go to the pudong airport " this language material is decomposed into the statement type of " will go ", the people of " I " claims the place name type speech of type speech, " pudong airport " and the statement rule of combination of " it is called+go+place name ", is stored in the database respectively.Like this, behind the displacement type speech, statement type and new type word combination remain a smoothness, correct language material.Be replaced into Nanjing Road Walk Way such as place name type speech, according to rule of combination, complete language material is: I will remove Nanjing Road Walk Way.It also is clear and coherent language material.
Final correct result presents the target language material and gives the user via step 104.For combined target language material, the requirement according to the statement rule of combination of target language behind the type word combination of the statement type of target language and target language, presents.Simultaneously, execution in step 109, write data storehouse: be about to target language and target language material and write multilingual corresponding relation database, and the authentication result field in the multilingual corresponding relation database is changed to OK.
For the language material that can be decomposed into statement type and type speech,, also to append the step that statement type and type speech and rule of combination is write multilingual corresponding relation database except target language and target language material are write the multilingual corresponding relation database.1) writing mode of statement type: in " the language material attribute " of source languages, write statement type, write that part of language material that is divided into statement type at " source language material ", in " the statement rule of combination " of source languages, write statement rule of combination (for example: the people claims type speech+statement type+place name type speech), in " the language material attribute " of target language, write statement type, write that part of language material that is divided into statement type at " target language material ", in " the statement rule of combination " of target language, write statement rule of combination (for example: the people claims type speech+statement type+place name type speech); 2) writing mode of type speech: in " the language material attribute " of source languages, write the type speech, write that part of language material that is divided into the type speech at " source language material ", " the statement rule of combination " of source languages is set at sky, in " the language material attribute " of target language, write the type speech, write that part of language material that is divided into the type speech at " target language material ", " the statement rule of combination " of target language is set at sky.If a plurality of type speech are arranged just according to 2) step append repeatedly.
The present invention carries out credit rating for authenticating and authenticate the people, shown in step 110.Credit rating to authentication is specially: the number of times that utilizes according to translation result, whether translation result there is the record of raising an objection and revising, and the situations such as difference between translation result and the final correct translation provide credit level.For example rank is totally 5 grades, and the inventive method writes database storing with the rank of authentication.Credit rating to the authentication people is specially: provide credit level according to this authentication people's the translation and number of times, the accuracy of authentication.For example, every increase is once translated, and then rating result is plused fifteen; Translation result is carried an objection, then rating result minus fifteen.The use translation result that the user can be more relieved like this.
If the user selected human translation, then direct execution in step 108 is submitted to mutual authentication, translation and decomposing module with translation brief, is directly translated by the authentication people and decomposes.
By above-mentioned to explanation of the present invention as seen: if retrieve identical language material, corresponding " authentication result=OK ", and Dui Ying target language is not empty with it, under the situation of corresponding " authentication result=OK ", can be judged to be correct translation result (target language material).The target language material is fed back to the user just finished translation process.
If one of following situation all be considered as do not have correct translation result.A. in the row of " row name=source languages ", do not retrieve the language material identical, do not retrieve the statement type identical yet with the part of source language material with source language; B. with source languages corresponding " authentication result is not equal to OK "; C. in the corresponding target language is empty; D. with target language corresponding " authentication result is not equal to OK ".
Under the situation that does not have correct translation result (target language material), the translation brief that the user is proposed feeds back to existing machine translation tools, and result's (semi-manufacture) of mechanical translation is carried out mutual authentication, has just begun to obtain correct translation result's process.
If it is correct that the authentication people judges translation result, just translation result is directly deposited in multilingual corresponding relation database.Deposit source language in according to " row name=source languages ", set source languages corresponding " authentication result=OK "; Deposit target language material, target setting languages corresponding " authentication result=OK " in according to " row name=target language ".
If the authentication people judges that translation result is wrong, just revise mistake by authentication people (can be that same authentication artificially also can not be same authentication people), then amended translation result is deposited in multilingual corresponding relation database.Deposit source language in according to " row name=source languages ", set source languages corresponding " authentication result=OK "; Deposit target language, target setting languages corresponding " authentication result=OK " in according to " row name=target language ".
If the authentication people judges that the source language material can decompose according to identical mode with the target language material, just by authentication people 1) part that is decomposed into statement type is write multilingual corresponding relation database.According in " the language material attribute " of source languages, writing statement type, write that part of language material that is divided into statement type at " source language material ", in " the statement rule of combination " of source languages, write statement rule of combination (for example: the people claims type speech+statement type+place name type speech), set source languages corresponding " authentication result=OK ".In " the language material attribute " of target language, write statement type, write that part of language material that is divided into statement type at " target language material ", in " the statement rule of combination " of target language, write statement rule of combination (for example: the people claims type speech+statement type+place name type speech), target setting languages corresponding " authentication result=OK "; 2) part that is decomposed into the type speech is appended write multilingual corresponding relation database.According in " the language material attribute " of source languages, writing the type speech, write that part of language material that is divided into the type speech at " source language material ", write the statement rule of combination in " the statement rule of combination " of source languages and be set at sky, set source languages corresponding " authentication result=OK ".Write the type speech in " the language material attribute " of target language, write that part of language material that is divided into the type speech at " target language material ", " the statement rule of combination " of target language is set at sky, target setting languages corresponding " authentication result=OK ".If a plurality of type speech are arranged just according to 2) step append repeatedly.
So just realized the birth process of multilingual corresponding relation database.Therefore as seen, use the present invention, not only the translation result of Huo Deing is correct, and the user is many more, and the correct result of the present invention's accumulative total and study is many more, is beneficial to commercialization.
Fig. 2 is the composition synoptic diagram that the present invention is based on the automatic translation system of correct translation result and corresponding relation.201 is multilingual corresponding relation database, wherein preserve all data of native system, as: source languages, source language material, target language, target language material, language material attribute, statement rule of combination, authenticate the credit level of whether OK, authentication, authentication people's credit level.
Whether retrieval module 203 is retrieved the source language material and is existed in multilingual corresponding relation database 201.If the source language material exists, or if not, but exist by the statement type that the part of source language material constitutes, and exist by the type speech that the part of source language material constitutes, authentication result is the words of OK, the correct translation result who has target language in the multilingual corresponding relation database 201 then is described, from multilingual corresponding relation database 201, reads the target language material or present to expression module 207 by the combined target language material that statement type+type speech constitutes by module for reading and writing 204.In the embodiments of the invention, according to express the different of interface module 208 that module 207 links to each other, therefore the multiple mode that presents is arranged, for example: webpage presents, the mobile phone literal presents, voice present, PDA presents, computer presents.In the mode that voice present, have a multilingual words in the multilingual corresponding relation database 201 and read aloud instrument, therefore language material can be presented with voice.
If the source language material does not exist, then submit to mechanical translation or submit to mutual authentication, translation and decomposing module 205 by interface module 208.Submitting to the translation result that obtains behind the machine translator is semi-manufacture, inaccurate usually, so the result of mechanical translation also submits to mutual authentication, translation and decomposing module 205.The alleged machine translator of the present invention provides the website or the software of automatic translation for all, for example: google translation, Yahoo translation, MSN translation etc.Mutual authentication is whether estimate the mechanical translation result by authentication people (or claim interpreter) correct, revises translation result till correct.The artificial authentication of the present invention can be a personal authentication, also can be that a plurality of people authenticate, revise successively (situation that a plurality of people revise comes across under the not accurate enough situation of translation result more) successively.Be statement type and a plurality of type speech, the statement type of target language and the process of a plurality of type speech that according to rule the source language material that can decompose and target language material is decomposed into source language by the authentication people artificial the decomposition.
Final correct result and those language materials that can be decomposed into statement type and type speech writes multilingual corresponding relation database 201 via module for reading and writing 204, presents to the user via expressing module 207 simultaneously.
Authentication people of the present invention is the people who understands translation, the present invention carries out credit rating for authenticating and authenticate the people, shown in 206 among Fig. 2, specifically: according to the number of times that utilizes of translation result, whether translation result is had the record of raising an objection and revising, and the situations such as difference between translation result and the final correct translation provide credit level.The user can use translation result more relievedly like this.
Embodiment 1:
The translation brief that the user submits to is: " I will go to the pudong airport " requires to translate into Japanese.The source languages are Chinese in other words, and (the row name: Chinese), the source language material is " I will go to the pudong airport "; Target language is " Japanese " (row name: Japanese).In the source of multilingual corresponding relation database languages=Chinese (row name: retrieve in the language material after authentication Chinese).
Result for retrieval: in Chinese languages (row name: found identical language material and Chinese authentication (row name: Authentication_CN=OK) Chinese), and corresponding target language (Japanese, row name: be not empty Japanese) and Japanese authentication (row name: Authentication_JP=OK).This explanation has the correct translation result through authentication.Particular content is: " the capable I of private は Pu East airport To い In The.”
Japanese (row name: (the capable I of the private は Pu East airport To い In The of the content Japanese) with correspondence.) feed back to the user, just finished the overall process that translator of Chinese becomes Japanese.
Embodiment 2:
The translation brief that the user submits to is: " I will go to the pudong airport " requires to translate into English.The source languages are Chinese in other words, and (the row name: Chinese), the source language material is " I will go to the pudong airport "; Target language is " English " (row name: English).In the source of multilingual corresponding relation database languages=Chinese (row name: retrieve in the language material after authentication Chinese).
Result for retrieval: in Chinese languages (row name: do not find identical language material Chinese).Implement the retrieval of the 2nd step, in Chinese languages (row name: retrieve in statement type Chinese), found identical statement type (Chinese=" will go "), corresponding Chinese authentication (row name: Authentication_CN=OK), and corresponding target language (the row name: be not empty English=" going to "), English authentication (row name: Authentication_EN=OK).The retrieval of the 2nd step has the result in other words, then prepares to implement the retrieval of the 3rd step.If the result of the 2nd step retrieval is: do not find identical statement type in the languages of source, perhaps corresponding target language is empty, and just being judged to be does not have translation result, does not carry out for the 3rd step and has retrieved.Implement the retrieval of the 3rd step, in Chinese languages (row name: retrieval in Chinese) " people claims the type speech ", found identical people to claim type speech (Chinese=" I "), corresponding Chinese authentication (row name: Authentication_CN=OK), and corresponding target language (the row name: be not empty English=" I ' m "), English authentication (row name: Authentication_EN=OK).Implement the retrieval of the 4th step, in Chinese languages (row name: retrieval in Chinese) " place name type speech ", found identical place name type speech (Chinese=" pudong airport "), corresponding Chinese authentication (row name: Authentication_CN=OK), and corresponding target language (the row name: be not empty English=" Pudong Airport "), English authentication (row name: Authentication_EN=OK).So far, statement type all finds with relevant type speech, " the statement rule of combination " stipulated in the statement type according to target language=" people claims the type speech "+" statement type "+" place name type speech ", the result who implements to obtain after the combination is: " I ' m going to Pudong Airport. ".It is fed back to the user, just finished translator of Chinese and become English overall process.
Embodiment 3:
The translation brief that the user proposes is: " the Shanghai new world there? ", require to translate into English (row name: English).In other words the source languages be Chinese (row names: Chinese), the source language material is " the Shanghai new world there? " Target language is " English " (row name: English).In the source of multilingual corresponding relation database languages=Chinese (row name: retrieve in the language material after authentication Chinese).
Result for retrieval: in Chinese languages (row name: found identical language material Chinese), with Chinese authentication (row name: Authentication_CN=OK), but corresponding target language (English, row name: be empty and English authentication (row name: Authentication_EN=" ") also do not have English).Carry out the retrieval of the second step statement type, the result does not have identical statement type, and this explanation does not have translation result.The translation result that the present invention recommends machine translation tools to draw, and submit mutual authentication to.The translation result authenticator draws translation result (semi-manufacture) to machine translation tools and authenticates, if translation result is correct, then deposits English corresponding in the multilingual corresponding relation database (row name: English) in as target language.(the row name: blank Authentication_EN) changes OK into corresponding English authentication simultaneously.
If translation result is wrong, then submit to the translation result reviser to right a wrong, then correct translation result is deposited in English corresponding in the multilingual corresponding relation database (row name: English) as target language.(the row name: blank Authentication_EN) changes OK into corresponding English authentication simultaneously.
Can decompose according to identical mode with the target language material if judge the source language material, respectively source language material and target language material are decomposed into statement type, type speech, and write multilingual corresponding relation database.Be specially: the first step is the statement type of source languages (row name: Chinese=" there "), language material attribute (row name: Attribute_CN=" statement type "), statement rule of combination (row name: Rule_CN=" place name type speech "+" statement type "), whether authenticate OK (row name: Authentication_CN=OK) and the statement type of corresponding target language (row name: English=" Where is "), language material attribute (row name: Attribute_EN=" statement type "), statement rule of combination (row name: Rule_EN=" statement type "+" place name type speech "), whether authenticate OK (row name: Authentication_EN=OK); Second step is the type speech of source languages (row name: Chinese=" Shanghai new world "), language material attribute (row name: Attribute_CN=" place name type speech "), statement rule of combination (row name: Rule_CN=" "), whether authenticate OK (row name: Authentication_CN=OK) and the type speech of corresponding target language (row name: English=" Shanghai Xintiandi "), language material attribute (row name: Attribute_EN=" place name type speech "), statement rule of combination (row name: Rule_EN=" "), whether authenticate OK (row name: Authentication_EN=OK).This process, the process of appending language material exactly to multilingual corresponding relation database.Many more to use of the present invention, accumulation and study is many more among the present invention, is convenient to commercialization.
Embodiment 4:
The translation brief that the user proposes is: " I ' m going to Pudong Airport. " requires to translate into Japanese.The source languages are that (the row name: English), the source language material is " I ' m going to Pudong Airport. " to English in other words; Target language is " Japanese " (row name: Japanese).In the source of multilingual corresponding relation database languages=English (row name: retrieve in the language material after authentication English).
The result for retrieval of multilingual corresponding relation database: in English (row name: found identical language material and English authentication (row name: Authentication_EN=OK) English), and corresponding target language (Japanese, row name: be not empty Japanese) and Japanese authentication (row name: Authentication_JP=OK).This explanation has the correct translation result through authentication.Particular content is: " the capable I of private は Pu East airport To い In The.”
Corresponding Japanese (row name: content (the capable I of private は Pu East airport To い In The Japanese).) feed back to the user, just finished the overall process that translator of English is become Japanese.
Can see by above-mentioned explanation, use the inventive method to translate automatically and can guarantee that the result is correct and be convenient to commercialization the embodiment of the invention.
Should be understood that above-mentioned detailed description to embodiment, can be improved or conversion those of ordinary skills according to the above description only in order to understand the present invention.So long as reach all improvement of this purpose and the protection domain that conversion all should belong to claims of the present invention.
Claims (10)
1. the automatic translating method based on correct translation result and corresponding relation is characterized in that, comprises following steps:
Receive translation brief, this translation brief comprises: source languages, source language material, target language;
Whether retrieval source language material exists in multilingual corresponding relation database, if exist, judges then whether the authentication result of target language is OK, if then present the target language material;
If not, judge then whether the statement type of source language material and type speech exist,, judge then whether the statement type of target language and the authentication result of type speech are OK, if then present the target language material if exist;
If the source language material does not exist and the statement type of source language material or the type speech does not exist or the statement type of target language or type speech do not exist, then select mechanical translation or human translation;
When the selector device is translated, translation brief is submitted to machine translator, obtain the mechanical translation result then, submit mutual authentication, translation and decomposition afterwards to, and with the outcome record of mutual authentication, decomposition in multilingual corresponding relation database;
When selecting human translation, directly submit mutual authentication, translation and decomposition to, and with the outcome record of human translation and decomposition in multilingual corresponding relation database.
2. the automatic translating method based on correct translation result and corresponding relation according to claim 1 is characterized in that, also comprises:
Whether the number of times that utilizes according to translation result has the record of raising an objection and revising to translation result, and the difference between translation result and the final correct translation is estimated credit level for authenticating and authenticate the people.
3. the automatic translating method based on correct translation result and corresponding relation according to claim 1 is characterized in that, the mode of this reception translation brief is one of following: website, phone, mobile phone, PDA, computer; This mode that presents this target language material is one of following: website, phone, mobile phone, PDA, computer.
4. the automatic translating method based on correct translation result and corresponding relation according to claim 1 is characterized in that, this translation brief specifically comprises: source languages, source language material, target language.
5. the automatic translating method based on correct translation result and corresponding relation according to claim 1 is characterized in that, this language material is one of following: word, statement, phrase or paragraph.
6. the automatic translating method based on correct translation result and corresponding relation according to claim 1 is characterized in that, this machine translator is one of following: google translation, Yahoo translation, MSN translation.
7. the automatic translating method based on correct translation result and corresponding relation according to claim 1, it is characterized in that, this with the step of mutual authentication outcome record in multilingual corresponding relation database is: target language and target language material are stored in the database, and the authentication result field in the multilingual corresponding relation database is changed to OK.
8. the automatic translating method based on correct translation result and corresponding relation according to claim 1 is characterized in that this is decomposed into: complete language material is decomposed into statement type and type speech, has the statement rule of combination in the statement type.
9. the automatic translation system based on correct translation result and corresponding relation is characterized in that, comprises:
Multilingual corresponding relation database is preserved one of following at least information: source languages, source language material, target language, target language material, language material attribute, statement rule of combination, whether authenticate OK, the credit level of authentication, authentication people's credit level;
Receiver module receives translation brief;
Retrieval module, whether this source language material of retrieval exists in this multilingual corresponding relation database, and whether corresponding authentication OK; Whether retrieve statement type and type speech exist;
Express module, present this target language material; For combined target language material, the requirement according to the statement rule of combination of target language behind the type word combination of the statement type of target language and target language, presents;
Module for reading and writing is each module reading and writing database;
Human translation and authentication module obtain correct target language material; The language material that can be decomposed into statement type and type speech is decomposed;
Interface module is for this receiver module and this expression module provide interface with external unit.
10. the automatic translation system based on correct translation result and corresponding relation as claimed in claim 9, it is characterized in that, also comprise: the credit rating module, the number of times that utilizes according to translation result, whether translation result is had the record of raising an objection and revising, and the difference between translation result and the final correct translation provides authentication and authentication people's credit level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100546908A CN101957815A (en) | 2009-07-13 | 2009-07-13 | Automatic translation method and system based on correct translation result and corresponding relation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100546908A CN101957815A (en) | 2009-07-13 | 2009-07-13 | Automatic translation method and system based on correct translation result and corresponding relation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101957815A true CN101957815A (en) | 2011-01-26 |
Family
ID=43485148
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100546908A Pending CN101957815A (en) | 2009-07-13 | 2009-07-13 | Automatic translation method and system based on correct translation result and corresponding relation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101957815A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102637167A (en) * | 2011-12-23 | 2012-08-15 | 东莞康明电子有限公司 | Multilingual inter-translation method |
| CN102929865A (en) * | 2012-10-12 | 2013-02-13 | 广西大学 | PDA (Personal Digital Assistant) translation system for inter-translating Chinese and languages of ASEAN (the Association of Southeast Asian Nations) countries |
| CN103294682A (en) * | 2012-02-24 | 2013-09-11 | 摩根全球购物有限公司 | Multi-language retrieving method, computer readable storage medium and network searching system |
| CN104391841A (en) * | 2014-12-05 | 2015-03-04 | 北京国双科技有限公司 | Data field translation processing method and device |
| CN104391840A (en) * | 2014-11-24 | 2015-03-04 | 上海迈外迪网络科技有限公司 | Translation method and device |
| CN105302800A (en) * | 2015-10-19 | 2016-02-03 | 黄阳光 | Method and apparatus for performing remote translation on commodity conversation information through internet |
| CN105677706A (en) * | 2015-12-28 | 2016-06-15 | 武汉传神信息技术有限公司 | Corpus retrieval matching method |
| CN106202060A (en) * | 2016-06-24 | 2016-12-07 | 深圳信息职业技术学院 | A kind of character input method and device |
| CN106202061A (en) * | 2016-06-28 | 2016-12-07 | 大连民族大学 | Chinese-English illiteracy towards cross-border e-commerce platform is hidden and is tieed up multi-lingual machine translation system and method |
| CN106372065A (en) * | 2016-10-27 | 2017-02-01 | 新疆大学 | Method and system for developing multi-language website |
| CN107301173A (en) * | 2017-06-22 | 2017-10-27 | 北京理工大学 | A kind of automatic post-editing system and method for multi-source neutral net that mode is remixed based on splicing |
| CN107402918A (en) * | 2017-06-19 | 2017-11-28 | 上海青橙实业有限公司 | The character string translation processing method and device of electric terminal |
| CN111221965A (en) * | 2019-12-30 | 2020-06-02 | 成都信息工程大学 | Classification sampling detection method based on bilingual corpus of public identification words |
| CN111753560A (en) * | 2020-06-28 | 2020-10-09 | 语联网（武汉）信息技术有限公司 | Self-adaptive translation system under parallel multi-source input mode |
| CN112784612A (en) * | 2021-01-26 | 2021-05-11 | 浙江香侬慧语科技有限责任公司 | Method, apparatus, medium, and device for synchronous machine translation based on iterative modification |
-
2009
- 2009-07-13 CN CN2009100546908A patent/CN101957815A/en active Pending
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102637167B (en) * | 2011-12-23 | 2015-02-18 | 曾敏生 | Multilingual inter-translation method |
| CN102637167A (en) * | 2011-12-23 | 2012-08-15 | 东莞康明电子有限公司 | Multilingual inter-translation method |
| CN103294682A (en) * | 2012-02-24 | 2013-09-11 | 摩根全球购物有限公司 | Multi-language retrieving method, computer readable storage medium and network searching system |
| CN102929865A (en) * | 2012-10-12 | 2013-02-13 | 广西大学 | PDA (Personal Digital Assistant) translation system for inter-translating Chinese and languages of ASEAN (the Association of Southeast Asian Nations) countries |
| CN104391840A (en) * | 2014-11-24 | 2015-03-04 | 上海迈外迪网络科技有限公司 | Translation method and device |
| CN104391841A (en) * | 2014-12-05 | 2015-03-04 | 北京国双科技有限公司 | Data field translation processing method and device |
| CN104391841B (en) * | 2014-12-05 | 2018-11-09 | 北京国双科技有限公司 | data field translation processing method and device |
| CN105302800A (en) * | 2015-10-19 | 2016-02-03 | 黄阳光 | Method and apparatus for performing remote translation on commodity conversation information through internet |
| CN105677706A (en) * | 2015-12-28 | 2016-06-15 | 武汉传神信息技术有限公司 | Corpus retrieval matching method |
| CN106202060A (en) * | 2016-06-24 | 2016-12-07 | 深圳信息职业技术学院 | A kind of character input method and device |
| CN106202061B (en) * | 2016-06-28 | 2018-09-14 | 大连民族大学 | Chinese-English illiteracy towards cross-border e-commerce platform, which is hidden, ties up multi-lingual machine translation system and method |
| CN106202061A (en) * | 2016-06-28 | 2016-12-07 | 大连民族大学 | Chinese-English illiteracy towards cross-border e-commerce platform is hidden and is tieed up multi-lingual machine translation system and method |
| CN106372065A (en) * | 2016-10-27 | 2017-02-01 | 新疆大学 | Method and system for developing multi-language website |
| CN106372065B (en) * | 2016-10-27 | 2020-07-21 | 新疆大学 | Multi-language website development method and system |
| CN107402918A (en) * | 2017-06-19 | 2017-11-28 | 上海青橙实业有限公司 | The character string translation processing method and device of electric terminal |
| CN107301173A (en) * | 2017-06-22 | 2017-10-27 | 北京理工大学 | A kind of automatic post-editing system and method for multi-source neutral net that mode is remixed based on splicing |
| CN107301173B (en) * | 2017-06-22 | 2019-10-25 | 北京理工大学 | A kind of automatic post-editing system and method for multi-source neural network remixing mode based on splicing |
| CN111221965A (en) * | 2019-12-30 | 2020-06-02 | 成都信息工程大学 | Classification sampling detection method based on bilingual corpus of public identification words |
| CN111753560A (en) * | 2020-06-28 | 2020-10-09 | 语联网（武汉）信息技术有限公司 | Self-adaptive translation system under parallel multi-source input mode |
| CN111753560B (en) * | 2020-06-28 | 2024-05-10 | 语联网（武汉）信息技术有限公司 | Self-adaptive translation system in parallel multi-source input mode |
| CN112784612A (en) * | 2021-01-26 | 2021-05-11 | 浙江香侬慧语科技有限责任公司 | Method, apparatus, medium, and device for synchronous machine translation based on iterative modification |
| CN112784612B (en) * | 2021-01-26 | 2023-12-22 | 浙江香侬慧语科技有限责任公司 | Method, device, medium and equipment for synchronous machine translation based on iterative modification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101957815A (en) | Automatic translation method and system based on correct translation result and corresponding relation | |
| CN101582075B (en) | Web information extraction system | |
| CN101816000B (en) | Autocompletion and automatic input method correction for partially entered search query | |
| US20140115439A1 (en) | Methods and systems for annotating web pages and managing annotations and annotated web pages | |
| US20110184960A1 (en) | Methods and systems for content recommendation based on electronic document annotation | |
| CN109446341A (en) | The construction method and device of knowledge mapping | |
| US20130013616A1 (en) | Systems and Methods for Natural Language Searching of Structured Data | |
| US10853564B2 (en) | Operation for copied content | |
| CN101833581A (en) | SEO website construction realizing method and system capable of optimizing search engine | |
| CN101561725B (en) | Method and system of fast handwriting input | |
| CN105335487A (en) | Agricultural specialist information retrieval system and method on basis of agricultural technology information ontology library | |
| CN104715032A (en) | Mapping system and method of Chinese and English table name and field name of report system | |
| US20150142811A1 (en) | Content Categorization System | |
| CN103390045A (en) | Time sequence storage method and time sequence storage device for monitoring system | |
| CN101976246A (en) | Classification retrieval method for encyclopedia entries | |
| CN104679642A (en) | Language independent processing of logs in a log analytics system | |
| CN116245177B (en) | Geographic environment knowledge graph automatic construction method and system and readable storage medium | |
| WO2022142635A1 (en) | Service information inputting method and apparatus, and server and storage medium | |
| CN103177039A (en) | Data processing method and data processing device | |
| CN109753638A (en) | Multi-layer document file management system and its management method | |
| CN103838763A (en) | Object file generation system and method | |
| CN112650529B (en) | System and method for configurable generation of mobile terminal APP codes | |
| CN102609455B (en) | Method for Chinese homophone searching | |
| US9916348B1 (en) | Answer facts from structured content | |
| JP6095487B2 (en) | Question answering apparatus and question answering method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| DD01 | Delivery of document by public notice | ||
| DD01 | Delivery of document by public notice |
Addressee: Bai Jinshi Document name: Decision of Rejection |
|
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110126 |