CN110705318B - Machine translation engine evaluation optimization method and system - Google Patents
Machine translation engine evaluation optimization method and system Download PDFInfo
- Publication number
- CN110705318B CN110705318B CN201910834491.2A CN201910834491A CN110705318B CN 110705318 B CN110705318 B CN 110705318B CN 201910834491 A CN201910834491 A CN 201910834491A CN 110705318 B CN110705318 B CN 110705318B
- Authority
- CN
- China
- Prior art keywords
- machine translation
- evaluation
- text
- translated
- translation engine
- 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.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Machine Translation (AREA)
Abstract
The invention discloses a machine translation engine evaluation optimization method and a system, wherein according to an evaluation word number range of a selected text to be translated, each machine translation engine performs translation evaluation, keyword evaluation and language model confusion evaluation on the text to be translated in the range to obtain a plurality of scores, then performs weighted summation on the scores to obtain a weighted sum value of each machine translation engine, and selects a translation result of a machine translation engine output sentence with the highest weighted sum value, so that the whole translation text is integrated. By the method, the machine translation engine evaluation optimization service can be provided for the user in a plurality of complicated machine translation engines with uneven translation quality, the translation efficiency is improved, the subsequent workload of the user is reduced, and the high-quality machine translation service is provided.
Description
Technical Field
The present disclosure relates to the field of machine translation, and in particular, to a method and system for evaluating a preference of a machine translation engine.
Background
Today, with the rapid development of artificial intelligence, machine translation technology has made breakthrough progress. A number of businesses have emerged in the marketplace to offer machine translation services including microsoft, google, hundred degrees, dog search, channel, tencel translation monarch, etc. Many complex machine translation engines are diverse in translation quality, so it is a necessary and necessary thing how to provide users with automatic preference machine translation engine services, and how to objectively evaluate machine translation engines currently on the market.
At present, there is no automatic preferred service of a machine translation engine in the market, and a user is in a stage of failing to distinguish between the best and bad translation engines, so that the translation result of the machine translation engine is not so satisfactory to the user, and the obtained result may need to be checked and modified again by the user, thereby affecting the efficiency of life and work.
Disclosure of Invention
The invention provides a method and a system for evaluating optimization of a machine translation engine, and aims to solve the problems that in the prior art, machine translation causes uneven translation quality and cannot obtain the optimal translation result.
To solve or at least partially solve the above-mentioned problems, in one embodiment of the present application, a method for evaluating a preference of a machine translation engine is provided, wherein the machine translation engine has a plurality of machine translation engines, and the method is characterized in that the method includes:
step one, selecting an evaluation word number range of the text to be translated for the uploaded text to be translated;
translating texts within the evaluation word number range of the text to be translated by using each machine translation engine to obtain corresponding translated texts;
thirdly, evaluating the translated text of each machine translation engine according to the translation, keyword and language model confusion degree to obtain a plurality of scores;
step four, carrying out weighted summation on the scores of all the machine translation engines to obtain the weighted summation value of all the machine translation engines on the translated text;
and fifthly, selecting the machine translation engine with the highest weighted sum value to output the whole translation text of the file to be translated.
In yet another embodiment of the present application, there is also provided a machine translation engine evaluation preference system, wherein the machine translation engine has a plurality, characterized in that the system comprises:
the selecting module is used for selecting an evaluation word number range of the text to be translated for the uploaded text to be translated;
the initial translation module is used for translating the text within the evaluation word number range of the text to be translated by using each machine translation engine to obtain a corresponding translated text;
the evaluation module is used for evaluating the translated text of each machine translation engine according to the translation, keyword and language model confusion, and obtaining a plurality of scores correspondingly;
the calculation module is used for carrying out weighted summation on the scores of the machine translation engines to obtain the weighted sum value of the machine translation engines on the translated text;
and the translation module is used for selecting the machine translation engine with the highest weighted sum value to output the whole translation text of the file to be translated.
The invention discloses a machine translation engine evaluation optimization method and a system, wherein according to an evaluation word number range of a selected text to be translated, each machine translation engine performs translation evaluation, keyword evaluation and language model confusion evaluation on the text to be translated in the range to obtain a plurality of scores, then performs weighted summation on the scores to obtain a weighted sum value of each machine translation engine, and selects a translation result of a machine translation engine output sentence with the highest weighted sum value, so that the whole translation text is integrated. By the method, the machine translation engine evaluation optimization service can be provided for the user in a plurality of complicated machine translation engines with uneven translation quality, the translation efficiency is improved, the subsequent workload of the user is reduced, and the high-quality machine translation service is provided.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following description will briefly introduce the drawings that are required to be used in the description of the embodiments or the prior art. It is evident that the figures in the following description are only intended to illustrate some embodiments of the present application, and that it is possible for a person skilled in the art to obtain technical features, connections or even method steps not mentioned in the other figures from these figures without inventive effort.
FIG. 1 is a flow chart of a preferred method of evaluating a machine translation engine according to an embodiment of the present invention;
FIGS. 2a, 2b, and 2c are schematic diagrams illustrating a portion of a preferred method for evaluating translations by a machine translation engine according to one embodiment of the present invention;
FIG. 3 is a schematic diagram of a machine translation engine evaluation preference system according to another embodiment of the present invention.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The terminology used in the embodiments of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, the "plurality" generally includes at least two, but does not exclude the case of at least one.
It should be understood that although the terms first, second, third, etc. may be used in embodiments of the present application to describe XXX, these XXX should not be limited to these terms. These terms are only used to distinguish XXX from each other. For example, a first XXX may also be referred to as a second XXX, and similarly, a second XXX may also be referred to as a first XXX, without departing from the scope of embodiments of the present application.
The words "if", as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination" or "in response to monitoring", depending on the context. Similarly, the phrase "if determined" or "if monitored (stated condition or event)" may be interpreted as "when determined" or "in response to determination" or "when monitored (stated condition or event)" or "in response to monitoring (stated condition or event), depending on the context.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a product or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such product or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a commodity or system comprising such elements.
In one embodiment, as shown in FIG. 1, a preferred method of evaluating a machine translation engine is provided, wherein the machine translation engine includes, but is not limited to, microsoft, google, hundred degrees, dog search, channel, tencel translation monarch, etc.; translation languages include, but are not limited to, chinese english, chinese, japanese, english, japanese, chinese, german, chinese, method, chinese, russia, chinese, korean, etc.
The method comprises the following steps:
step one, selecting an evaluation word number range of the text to be translated for the uploaded text to be translated.
The user uploads the file to be translated, and the system can automatically select a plurality of lines or sections of texts in the text to be translated, or select a certain word number range as a translation evaluation range.
And secondly, translating the text within the evaluation word number range of the text to be translated by using each machine translation engine to obtain the corresponding translated text.
And translating the text within the range of the translation evaluation in the previous step by using each machine translation engine to obtain the corresponding translated text.
And thirdly, evaluating the translated text of each machine translation engine according to the translation, keyword and language model confusion degree to obtain a plurality of scores.
As shown in fig. 2a-2b, the first aspect: the translation evaluation comprises the following steps:
1) Setting an error type and an error level for the text to be translated, wherein the error type has a plurality of sub-error types, and each error level has a corresponding penalty score;
2) Revising each translated text, and revising and selecting the sub-error type and the error level for each position to automatically generate a corresponding penalty score;
3) Summing the penalty scores corresponding to the different sub-error types in the revision to obtain a score sum value of each translated text, as shown in FIG. 2 c;
4) Ordering all the machine translation engines in a descending order according to the scores and the values;
5) In the descending order, the sequence of each machine translation engine is the score of the machine translation engine on the translation evaluation.
Wherein, the second aspect: the method for evaluating the keywords comprises the following steps:
1) A professional translator gives keywords which the text to be translated should contain; the translated text and the keywords form an evaluation set together;
2) Detecting whether keywords appear in each machine translation result, if not, subtracting one score, wherein the initial score of each machine translation engine is 0;
3) Sorting all the machine translation engines in ascending order according to the scores;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on the keyword evaluation.
The method comprises the following steps of:
1) Translating by using each machine translation engine to obtain the translated text to form a test set;
2) Scoring the test set of each machine translation engine by using the trained neural network language model;
3) According to the scores, sequencing all the machine translation engines in ascending order;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on the confusion degree evaluation of the language model.
Further, the above-mentioned evaluation scores in three aspects are listed, and other evaluation score methods in other aspects may be added in the art, and the evaluation scores may be evaluation scores in terms of user preference, average editing cost, average editing time, and the like.
Wherein, user preference, its evaluation step is:
1) Historical translation data of each machine translation engine in the technical field of the text to be translated is used as an evaluation set.
2) Counting the total number of times that the user selects each machine translation engine to translate on the evaluation set;
3) According to the total translation times of each machine translation engine, sequencing each machine translation engine in ascending order;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on the user preference.
Wherein, average editing cost, its evaluation step is:
1) And extracting a certain amount of data from historical translation data of each machine translation engine in the technical field of the text to be translated as an evaluation set. The data volume is 2 ten thousand, and the data of each machine translation engine is 2 ten thousand.
2) According to the difference of the machine translation engines, calculating average editing cost respectively;
wherein the average editing cost is defined as:
where len (x) is the length of the character string x, MT (x) is the result of machine translation of sentence segment x, PE (x) is the result of manual modification of sentence segment x, and ED (x) is the edit distance from the result of machine translation of sentence segment x to the result of manual modification.
3) According to the average editing cost, ordering all the machine translation engines in a descending order;
4) In the above descending order, the sequence of each machine translation engine is the score of the machine translation engine on the average editing cost.
The average editing time is measured by the following steps:
1) And extracting a certain amount of data from historical translation data of each machine translation engine in the technical field of the text to be translated as an evaluation set. The data volume is 2 ten thousand, and the data of each machine translation engine is 2 ten thousand.
2) According to the difference of the machine translation engines, calculating average editing time respectively;
wherein the average edit time is defined as:
wherein the time it takes for an ET (x) sentence fragment x translator to begin translation to acknowledge.
3) According to the average editing time, ordering all the machine translation engines in a descending order;
4) In the above descending order, the sequence of each machine translation engine is the score of the machine translation engine on the average editing time.
And step four, weighting and summing the scores of the machine translation engines to obtain the weighted sum value of the machine translation engines on the translated text.
The weights of the translation evaluation, the keyword evaluation and the language model confusion degree evaluation in the weighted summation are respectively set to be 0.4, 0.3 and 0.3.
Further, if other evaluation methods in the foregoing are required to be combined, the person skilled in the art can reasonably set the weights of the user preference, the average editing cost and the average editing time.
And fifthly, selecting the machine translation engine with the highest weighted sum value to output the whole translation text of the file to be translated.
After the highest weighted sum value is obtained, namely the optimal machine translation engine, the optimal machine translation engine is used for translating the file to be translated uploaded by the user into a required language, and then the language is returned to the user.
According to the method for evaluating and optimizing the machine translation engines, according to the evaluation word number range of the selected text to be translated, each machine translation engine evaluates the text to be translated in the range, evaluates the keyword and evaluates the confusion degree of the language model to obtain a plurality of scores, then performs weighted summation on the scores to obtain weighted sum values of each machine translation engine, selects the translation result of the machine translation engine output sentence with the highest weighted sum value, and integrates the translation result to obtain the whole translation text. By the method, the machine translation engine evaluation optimization service can be provided for the user in a plurality of complicated machine translation engines with uneven translation quality, the translation efficiency is improved, the subsequent workload of the user is reduced, and the high-quality machine translation service is provided.
In another embodiment, as shown in FIG. 3, a machine translation engine evaluation preference system is provided, wherein the machine translation engine includes, but is not limited to, microsoft, google, hundred degrees, dog search, channel, tencel translation monarch, and the like; translation languages include, but are not limited to, chinese english, chinese, japanese, english, japanese, chinese, german, chinese, method, chinese, russia, chinese, korean, etc.
The system comprises a selection module, an initial translation module, an evaluation module, a calculation module and a translation module.
And the selection module is used for selecting the evaluation word number range of the text to be translated for the uploaded text to be translated.
The user uploads the file to be translated, and the system can automatically select a plurality of lines or sections of texts in the text to be translated, or select a certain word number range as a translation evaluation range.
And the initial translation module is used for translating the text within the evaluation word number range of the text to be translated by using each machine translation engine to obtain the corresponding translated text.
And translating the text within the range of the translation evaluation in the previous step by using each machine translation engine to obtain the corresponding translated text.
And the evaluation module is used for evaluating the translated text of each machine translation engine according to the translation, keyword and language model confusion, and obtaining a plurality of scores correspondingly.
Wherein, in the first aspect: the translation evaluation comprises the following steps:
1) Setting an error type and an error level for the text to be translated, wherein the error type has a plurality of sub-error types, and each error level has a corresponding penalty score;
2) Revising each translated text, and revising and selecting the sub-error type and the error level for each position to automatically generate a corresponding penalty score;
3) Summing the penalty scores corresponding to the different sub-error types in the revision to obtain a score sum value of each translated text;
4) Ordering all the machine translation engines in a descending order according to the scores and the values;
5) In the descending order, the sequence of each machine translation engine is the score of the machine translation engine on the translation evaluation.
Wherein, the second aspect: the method for evaluating the keywords comprises the following steps:
1) A professional translator gives keywords which the text to be translated should contain; the translated text and the keywords form an evaluation set together;
2) Detecting whether keywords appear in each machine translation result, if not, subtracting one score, wherein the initial score of each machine translation engine is 0;
3) Sorting all the machine translation engines in ascending order according to the scores;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on the keyword evaluation.
The method comprises the following steps of:
1) Translating by using each machine translation engine to obtain the translated text to form a test set;
2) Scoring the test set of each machine translation engine by using the trained neural network language model;
3) According to the scores, sequencing all the machine translation engines in ascending order;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on the confusion degree evaluation of the language model.
And the calculation module is used for carrying out weighted summation on the scores of the machine translation engines to obtain the weighted sum value of the machine translation engines on the translated text.
The weights of the translation evaluation, the keyword evaluation and the language model confusion degree evaluation in the weighted summation are respectively set to be 0.4, 0.3 and 0.3.
And the translation module is used for selecting the machine translation engine with the highest weighted sum value to output the whole translation text of the file to be translated.
After the highest weighted sum value is obtained, namely the optimal machine translation engine, the optimal machine translation engine is used for translating the file to be translated uploaded by the user into a required language, and then the language is returned to the user.
According to the above-mentioned machine translation engine evaluation optimization system, according to the evaluation word number range of the selected text to be translated, each machine translation engine performs translation evaluation, keyword evaluation and language model confusion evaluation on the text to be translated in the range to obtain a plurality of scores, then performs weighted summation on the scores to obtain weighted sum values of each machine translation engine, and selects the translation result of the machine translation engine output sentence with the highest weighted sum value, so that the whole translation text is integrated. By the system, the machine translation engine evaluation optimization service can be provided for the user in a plurality of complicated machine translation engines with uneven translation quality, the translation efficiency is improved, the subsequent workload of the user is reduced, and the high-quality machine translation service is provided.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a product or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such product or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a commodity or system comprising such elements.
While the foregoing description illustrates and describes the preferred embodiments of the present invention, it is to be understood that the invention is not limited to the forms disclosed herein, but is not to be construed as limited to other embodiments, and is capable of numerous other combinations, modifications and environments and is capable of changes or modifications within the scope of the inventive concept as described herein, either as a result of the foregoing teachings or as a result of the knowledge or technology in the relevant art. And that modifications and variations which do not depart from the spirit and scope of the invention are intended to be within the scope of the appended claims.
Claims (4)
1. A machine translation engine evaluation preference method, wherein the machine translation engine has a plurality of, the method comprising:
step one, selecting an evaluation word number range of the text to be translated for the uploaded text to be translated;
translating the text within the evaluation word number range of the text to be translated by using each machine translation engine to obtain a corresponding translated text;
thirdly, evaluating the translated text of each machine translation engine according to the translation, keyword and language model confusion degree to obtain a plurality of corresponding scores;
step four, carrying out weighted summation on the scores of all the machine translation engines to obtain weighted summation values of all the machine translation engines on the translated text;
step five, selecting the machine translation engine with the highest weighted sum value to output the whole translation text of the text to be translated;
the translation evaluation method comprises the following steps:
1) Setting an error type and an error level for the text to be translated, wherein the error type has a plurality of sub-error types, and each error level has a corresponding penalty score;
2) Revising each of said translated text, selecting said sub-error type and said error level for each of said revisions, automatically generating said corresponding penalty score;
3) Summing the penalty scores corresponding to the sub-error types which are different in the revision to obtain a score sum value of each translated text;
4) Sorting the machine translation engines in a descending order according to the scores and the values;
5) In the descending order sorting, the sequence of each machine translation engine is the score of the machine translation engine on the translation evaluation;
the keyword evaluation step comprises the following steps:
1) The professional translator gives the keywords which the text to be translated should contain; the translated text and the keywords form an evaluation set together;
2) Detecting whether keywords appear in each machine translation result, if not, subtracting one score, wherein the initial score of each machine translation engine is 0;
3) Sorting the machine translation engines in ascending order according to the score;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on keyword evaluation;
the method comprises the following steps of:
1) Translating by using each machine translation engine to obtain the translated text to form a test set;
2) Scoring the test set of each machine translation engine by using the trained neural network language model;
3) According to the scoring, sorting the machine translation engines in ascending order;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on the confusion degree evaluation of the language model.
2. The machine translation engine evaluation preference method according to claim 1, wherein the weights of the translation evaluation, keyword evaluation and language model confusion evaluation in the weighted summation are set to 0.4, 0.3, respectively.
3. A machine translation engine evaluation preference system wherein the machine translation engine has a plurality, the system comprising:
the selection module is used for selecting the evaluation word number range of the text to be translated for the uploaded text to be translated;
the initial translation module is used for translating the text within the evaluation word number range of the text to be translated by using each machine translation engine to obtain a corresponding translated text;
the evaluation module is used for evaluating the translated text of each machine translation engine according to the translation, keyword and language model confusion, so as to obtain a plurality of scores;
the calculation module is used for carrying out weighted summation on the scores of the machine translation engines to obtain a weighted sum value of the machine translation engines on the translated text;
the translation module is used for selecting the machine translation engine with the highest weighted sum value to output the whole translation text of the text to be translated;
the translation evaluation method comprises the following steps:
1) Setting an error type and an error level for the text to be translated, wherein the error type has a plurality of sub-error types, and each error level has a corresponding penalty score;
2) Revising each of said translated text, selecting said sub-error type and said error level for each of said revisions, automatically generating said corresponding penalty score;
3) Summing the penalty scores corresponding to the sub-error types which are different in the revision to obtain a score sum value of each translated text;
4) Sorting the machine translation engines in a descending order according to the scores and the values;
5) In the descending order sorting, the sequence of each machine translation engine is the score of the machine translation engine on the translation evaluation;
the keyword evaluation step comprises the following steps:
1) The professional translator gives the keywords which the text to be translated should contain; the translated text and the keywords form an evaluation set together;
2) Detecting whether keywords appear in each machine translation result, if not, subtracting one score, wherein the initial score of each machine translation engine is 0;
3) Sorting the machine translation engines in ascending order according to the score;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on keyword evaluation;
the method comprises the following steps of:
1) Translating by using each machine translation engine to obtain the translated text to form a test set;
2) Scoring the test set of each machine translation engine by using the trained neural network language model;
3) According to the scoring, sorting the machine translation engines in ascending order;
4) In the ascending order, the sequence of each machine translation engine is the score of the machine translation engine on the confusion degree evaluation of the language model.
4. A machine translation engine evaluation preference system according to claim 3 wherein the translation evaluation, keyword evaluation and language model confusion evaluation are weighted at weights of 0.4, 0.3 respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910834491.2A CN110705318B (en) | 2019-09-04 | 2019-09-04 | Machine translation engine evaluation optimization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910834491.2A CN110705318B (en) | 2019-09-04 | 2019-09-04 | Machine translation engine evaluation optimization method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110705318A CN110705318A (en) | 2020-01-17 |
| CN110705318B true CN110705318B (en) | 2023-06-30 |
Family
ID=69194244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910834491.2A Active CN110705318B (en) | 2019-09-04 | 2019-09-04 | Machine translation engine evaluation optimization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110705318B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111680525B (en) * | 2020-06-09 | 2024-03-26 | 语联网（武汉）信息技术有限公司 | Man-machine co-translation method and system based on reverse difference recognition |
| CN111680526B (en) * | 2020-06-09 | 2023-09-08 | 语联网（武汉）信息技术有限公司 | Man-machine interactive translation system and method based on comparison of reverse translation results |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101248432A (en) * | 2005-08-25 | 2008-08-20 | 慕迪灵公司 | Translation quality quantifying apparatus and method |
| CN102799579A (en) * | 2012-07-18 | 2012-11-28 | 西安理工大学 | Statistical machine translation method with error self-diagnosis and self-correction functions |
| CN108428382A (en) * | 2018-02-14 | 2018-08-21 | 广东外语外贸大学 | It is a kind of spoken to repeat methods of marking and system |
| CN109299481A (en) * | 2018-11-15 | 2019-02-01 | 语联网（武汉）信息技术有限公司 | MT engine recommended method, device and electronic equipment |
-
2019
- 2019-09-04 CN CN201910834491.2A patent/CN110705318B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101248432A (en) * | 2005-08-25 | 2008-08-20 | 慕迪灵公司 | Translation quality quantifying apparatus and method |
| CN102799579A (en) * | 2012-07-18 | 2012-11-28 | 西安理工大学 | Statistical machine translation method with error self-diagnosis and self-correction functions |
| CN108428382A (en) * | 2018-02-14 | 2018-08-21 | 广东外语外贸大学 | It is a kind of spoken to repeat methods of marking and system |
| CN109299481A (en) * | 2018-11-15 | 2019-02-01 | 语联网（武汉）信息技术有限公司 | MT engine recommended method, device and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 机器翻译评测技术研究;蒋俊杰;《中国优秀硕士学位论文全文数据库 信息科技辑》;20131215;第6、13、21页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110705318A (en) | 2020-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108363790B (en) | Method, device, equipment and storage medium for evaluating comments | |
| JP3856778B2 (en) | Document classification apparatus and document classification method for multiple languages | |
| CN110580292B (en) | Text label generation method, device and computer readable storage medium | |
| CN109344237B (en) | Information processing method and device for man-machine interaction | |
| US20180329898A1 (en) | Machine translation method for performing translation between languages | |
| US7548847B2 (en) | System for automatically annotating training data for a natural language understanding system | |
| CN110134944A (en) | A kind of reference resolution method based on intensified learning | |
| CN111310440B (en) | Text error correction method, device and system | |
| CN110472256B (en) | Machine translation engine evaluation optimization method and system based on chapters | |
| EP1703419A1 (en) | Translation judgment device, method, and program | |
| CN111666409B (en) | Integrated emotion intelligent classification method for complex comment text based on comprehensive deep capsule network | |
| CN111611810A (en) | Polyphone pronunciation disambiguation device and method | |
| CN110705318B (en) | Machine translation engine evaluation optimization method and system | |
| CN110472257A (en) | A kind of MT engine assessment preferred method and system based on sentence pair | |
| CN109033085A (en) | The segmenting method of Chinese automatic word-cut and Chinese text | |
| CN116719520A (en) | Code generation method and device | |
| CN110688540B (en) | Cheating account screening method, device, equipment and medium | |
| CN104572632A (en) | Method for determining translation direction of word with proper noun translation | |
| CN113822052A (en) | Text error detection method and device, electronic equipment and storage medium | |
| CN111782789A (en) | Intelligent question and answer method and system | |
| CN110377753A (en) | Relation extraction method and device based on relationship trigger word Yu GRU model | |
| CN110750712A (en) | Software security requirement recommendation method based on data driving | |
| CN115587163A (en) | Text classification method and device, electronic equipment and storage medium | |
| CN115563959A (en) | Chinese pinyin spelling error correction-oriented self-supervision pre-training method, system and medium | |
| CN112528003B (en) | Multi-item selection question-answering method based on semantic sorting and knowledge correction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |