CN108763228A - A kind of natural language secondary treatment system - Google Patents
A kind of natural language secondary treatment system Download PDFInfo
- Publication number
- CN108763228A CN108763228A CN201810490347.7A CN201810490347A CN108763228A CN 108763228 A CN108763228 A CN 108763228A CN 201810490347 A CN201810490347 A CN 201810490347A CN 108763228 A CN108763228 A CN 108763228A
- Authority
- CN
- China
- Prior art keywords
- language
- module
- target
- mark
- source
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/55—Rule-based translation
- G06F40/56—Natural language generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/58—Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
Abstract
The invention discloses a kind of natural language secondary treatment systems, including language in-put module, language processing module, human-computer interaction module and display module.Language in-put module, for inputting pending source language.Language processing module for generating several targets to source language, and is identified target according to database, and several to being identified as the same target.Human-computer interaction module, including language chosen module and language generation module, the language chosen module is used for selected target, and one or more marks are applied in selected target, the language generation module is used for according to mark and database, is Markup Language by selected target processing.Display module, for showing the Markup Language and the source language.The present invention can carry out the source language of input a variety of processing, powerful to the processing function of source language, and multi-faceted processing can be carried out to language, is offered convenience to user.
Description
Technical field
The present invention relates to computer-aided translation fields, and in particular to a kind of natural language secondary treatment system.
Background technology
With social development and scientific and technological progress, computer-aided translation is more and more common.
In terms of natural language processing, people repeat complicated inquiry, retrieve, compare, uniformly, tissue and school
It tests.
Computer-aided translation then needs mass data, and style, the content translated are single, underaction.
Below with《Cha Latusitela is said in this way》One section of German translation specification:
Original text:Als Zarathustra aber allein war,sprach er also zu seinem
Herzen:"Sol lte es dennsein!Dieser alte Hei l ige hat in seinem
Walde noch Nichts davondass Gott todt ist!"-
To the mark of target in terms of natural language processing:Als when as than Zarathustra Cha Latusite
Drawing aber, still war is allein alone, and Sprach says his Also of er zu in this way to his Herzen of seinem
The heart:" Sollte should its this denn of es becausePossible sein is oneself!This alte of Dieser is old
Heilige sage hat have in inner seinem his Walde woods noch also Nichts not have davon from thatIt hears
It hears, ist is in the dust by dass that Gott god todt!"-
Translation:But when Cha Latusitela alone when, he in this way says in mind him:" this should be possible
's!This old sage does not hear that god is in the dust also in his woods!"——
More than, it is seen that in translation process, to the mark of target in terms of natural language processing, target is specially word,
The mark specially corresponding part of the meaning of a word and translation is not necessarily completely unique corresponding.Identical word can correspond to not
Same translation, such as:The translation of Als can be " when working as " according to position and context, " as ", " ratio ";Different lists
Word can also be identical translation;One translation can also be the corresponding phrase etc. for including multiple words.
It can be seen that people, which needs to mitigate the mark of target in terms of assisting being shown in natural language processing, repeats complicated look into
Ask, retrieve, comparing, uniformly, tissue and verification, computer-aided translation then needs the largely target in terms of natural language processing
And mark data, in addition, computer-aided translation is difficult to hold for the style of translation, there is also the content of translation is single, no
Enough flexible problems.
In view of this, being badly in need of a kind of natural language secondary treatment system, problems with is solved:
(1) it in terms of natural language processing, needs repeat complicated inquiry to target and mark, retrieve, compare, unite
One, it organizes and verifies;
(2) in terms of computer-aided translation and natural language processing, a large amount of target and mark data are needed, and translate
Style, content is single, underaction.
Invention content
Existing for existing natural language secondary treatment system, in natural language
In terms of processing, need to target and mark repeat complicated inquiry, retrieve, compare, uniformly, tissue and verification;It is calculating
In terms of machine supplementary translation and natural language processing, a large amount of target and mark data, and the style translated are needed, content is single,
Underaction, the above both sides problem.
In order to solve the above-mentioned technical problem, the technical solution adopted in the present invention is to provide a kind of natural language aid in treatment
System, including:
Language in-put module, for inputting pending source language;
Language processing module, for generating several targets to the source language, and according to database to target into rower
Know, and several to being identified as the same target;
Human-computer interaction module, including language chosen module and language generation module, the language chosen module is for selecting
Target, and one or more marks are applied in selected target, the language generation module is used for according to mark and data
Selected target processing is Markup Language by library;
Display module, for showing the Markup Language and the source language.
In the above scheme, the target includes word, phrase, sentence, paragraph, carrier and specified target, the mark
Including the meaning of a word, rule, phonetic symbol, pronunciation, phrase, morpheme, sentence, ingredient, word order, retroversion, part of speech, single plural number, tense, displacement,
Flexion, parsing, teaching, source, mandate, grammer, carrier, office collaboration, prompt suggestion, search result, abstract, is specified at derivation
Mark, auxiliary are heard and are read and write.
In the above scheme, the database includes Excel databases, Access databases and SQL database.
In the above scheme, further include custom system, the custom system includes memory module, for having stored user
The target and mark of grasp, the custom system target in the source language for identification, and pass through the Language Processing
Module is not to whether identical target and mark or not with the memory module in the source language.
In the above scheme, further include Style Transfer module, be used for selected target and mark, and according to target and mark
It is selected to generate specified specific objective and specific identifier simultaneously stores, the Style Transfer module be used for according to the specific objective and
Source language processing is the Markup Language by the specific identifier.
In the above scheme, the language chosen module includes input equipment and output equipment, and the input equipment is used for
Selected target, the output equipment is for showing the source language and the Markup Language.
In the above scheme, the input equipment includes mouse, keyboard and touch screen, and the output equipment includes display
Screen.
Compared with prior art, of the invention, a variety of processing, the processing to source language can be carried out to the source language of input
It is powerful, multi-faceted processing can be carried out to language, offered convenience to user.
Specific implementation mode
The present invention provides a kind of natural language secondary treatment system, can effectively solve to assist when people handles natural language
The problem of.In terms of natural language processing, by generating several targets to source language, and target is identified, and according to
Target is generated Markup Language by mark, be effectively reduced the inquiry of repetition complexity, retrieve, compare, uniformly, tissue and verify.This
Outside, compared to existing computer-aided translation, the present invention also efficiently solves it and is needed in terms of natural language processing largely
Target and mark data, and the style translated is difficult to control, the problems such as content of translation is single, underaction.With reference to
Specific implementation mode is described in detail the present invention.
The present invention provides a kind of natural language secondary treatment systems, including language in-put module, language processing module, people
Machine interactive module and display module.Language in-put module, for inputting pending language, which is denoted as source language.Language
It says processing module, for generating several targets to source language, and target is identified according to database, and to the same mesh
Target mark can be several.Human-computer interaction module, including language chosen module and language generation module, language chosen module are used
In selected target, and to the selected target selection to its mark, one or more marks can be applied to as a result, selected
In target, the corresponding mark of the target is shown, will be selected according to the mark and database of the target by language generation module
Target processing be Markup Language, auxiliary generate translation.Display module for instant playback Markup Language, source language and is translated
Text.User inputs source language by language in-put module as a result, then passes through language processing module and human-computer interaction module pair
Target and mark in the source language of input are selected, and be can get Markup Language and translation, can be realized by display module
It is shown while source language and Markup Language and translation, is convenient for user to check as a result, modification, it is intuitive and convenient.
Specifically, above-mentioned target includes word, phrase, sentence, paragraph, carrier and specified target, and mark includes the meaning of a word, rule
Then, phonetic symbol, pronunciation, phrase, morpheme, sentence, ingredient, word order, retroversion, part of speech, single plural number, tense, displacement, flexion, derivation, solution
Analysis, teaching, source, mandate, grammer, carrier, office collaboration, prompt suggestion, search result, abstract, designated identification, auxiliary are heard
And read-write.
Preferably, it by human-computer interaction module, can be selected to the mesh after selected target, in the form of drop-down menu
Target identifies and application range.After selected identity, pass through language generation module, you can according to selected target and mark, generate
Markup Language corresponding with the target and mark, understands original text convenient for user's multi-angle.Such as to some word, phrase, sentence,
The supplementary translation of paragraph, carrier or specified target, can be by identifying the modes such as ingredient, derivation, parsing, so that user is to original text
Carry out deep understanding.
The present invention further includes synchronization module, which is generated target for downloading and uploading off-line data
And mark, and by the target of generation and mark storage to database, thus update the data library.Database includes Excel data
Library, Access databases and SQL database.
The invention also includes custom system, custom system includes memory module, for store user grasped it is original
Language, the source language that custom system can grasp the user of input are identified, and control language processing module is to original
The content that the source language grasped with the user stored in memory module in beginning language repeats does not identify, and controls language
It says that generation module does not carry out supplementary translation to the target, is conducive to be promoted user as a result, to the reading ability of original text, avoids marking
Knowing influences the thinking of user.Specifically, custom system includes memory module, for storing the target and mark that user has grasped
Know, the custom system target in source language for identification, and by language processing module in source language in storage module
Whether identical target and mark or not, hereby it is achieved that above-mentioned function.
The present invention further includes Style Transfer module, is Markup Language for handling source language according to specific mode.Tool
Body, target and mark of the Style Transfer module for being generated to source language are selected, and by selecting, generate specified spy
It sets the goal and specific identifier, Style Transfer module stores the specific objective and specific identifier, Style Transfer module is stored according to it
Specific objective and specific identifier by source language processing be Markup Language, thus to obtain specified Markup Language.Such as selected mesh
It marks " they ", selectes the mark " plural number " of the target, store specified target and mark, and the specified target and mark as a result,
It can be several to know, that is, Style Transfer module may specify the mark to several targets, in use, referring in advance automatically according to it
Fixed specific objective and specific identifier generate Markup Language.Specifically, the mesh generated in Style Transfer module identification source language
Mark, and is compared with the specific objective of storage, by target identical with specific objective in source language according to specific objective and
The target selected in source language processing is Markup Language by specific identifier, this Markup Language is due to according to specific objective and spy
Calibration, which is known, to be generated, and is had specific style, is achieved in Style Transfer.The present invention can be right without selected all target and mark
Source language carries out multi-faceted processing, offers convenience to user.
The invention also includes identification module, corresponding target is identified for showing, convenient for user to selected target and
Mark is compared, and changes and checks.
In the present invention, language chosen module includes input equipment and output equipment, and input equipment is used for selected target, output
Equipment is for showing source language and Markup Language.Preferably, input equipment includes mouse, keyboard and touch screen, output equipment
Including display screen.
Compared with prior art, of the invention, a variety of processing, the processing to source language can be carried out to the source language of input
It is powerful, multi-faceted processing can be carried out to language, offered convenience to user.
The invention is not limited in above-mentioned preferred forms, and anyone should learn that is made under the inspiration of the present invention
Structure change, the technical schemes that are same or similar to the present invention are each fallen within protection scope of the present invention.
Claims (7)
1. a kind of natural language secondary treatment system, which is characterized in that including:
Language in-put module, for inputting pending source language;
Language processing module for generating several targets to the source language, and is identified target according to database, and
It is several to being identified as the same target;
Human-computer interaction module, including language chosen module and language generation module, the language chosen module are used for selected target,
And one or more marks are applied in selected target, the language generation module is used for according to mark and database, will
Selected target processing is Markup Language;
Display module, for showing the Markup Language and the source language.
2. a kind of natural language secondary treatment system according to claim 1, which is characterized in that the target includes single
Word, phrase, sentence, paragraph, carrier and specified target, the mark includes the meaning of a word, rule, phonetic symbol, pronunciation, phrase, morpheme, sentence
Son, ingredient, word order, retroversion, part of speech, single plural number, tense, displacement, flexion, derivation, parsing, teaching, source, mandate, grammer, load
Body, office collaboration, prompt suggestion, search result, abstract, designated identification, auxiliary are heard and are read and write.
3. a kind of natural language secondary treatment system according to claim 1, which is characterized in that the database includes
Excel databases, Access databases and SQL database.
4. a kind of natural language secondary treatment system according to claim 1, which is characterized in that further include custom system,
The custom system includes memory module, and for storing the target and mark that user has grasped, the custom system is for knowing
Target in the not described source language, and by the language processing module in the source language in the memory module
Identical target and mark or not.
5. a kind of natural language secondary treatment system according to claim 1, which is characterized in that further include Style Transfer mould
Block is used for selected target and mark, and according to the specified specific objective of the selected generation of target and mark and specific identifier and deposits
Storage, the Style Transfer module are used to according to the specific objective and the specific identifier be described by source language processing
Markup Language.
6. a kind of natural language secondary treatment system according to claim 1, which is characterized in that the language chosen module
Including input equipment and output equipment, the input equipment is used for selected target, and the output equipment is described original for showing
Language and the Markup Language.
7. a kind of natural language secondary treatment system according to claim 6, which is characterized in that the input equipment includes
Mouse, keyboard and touch screen, the output equipment include display screen.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810490347.7A CN108763228A (en) | 2018-05-21 | 2018-05-21 | A kind of natural language secondary treatment system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810490347.7A CN108763228A (en) | 2018-05-21 | 2018-05-21 | A kind of natural language secondary treatment system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108763228A true CN108763228A (en) | 2018-11-06 |
Family
ID=64007651
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810490347.7A Pending CN108763228A (en) | 2018-05-21 | 2018-05-21 | A kind of natural language secondary treatment system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108763228A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109408626A (en) * | 2018-11-09 | 2019-03-01 | 苏州思必驰信息科技有限公司 | The method and device that natural language is handled |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5826219A (en) * | 1995-01-12 | 1998-10-20 | Sharp Kabushiki Kaisha | Machine translation apparatus |
CN1308748A (en) * | 1998-05-04 | 2001-08-15 | 特雷道斯股份有限公司 | Machine-assisted translation tools |
CN1487449A (en) * | 2002-08-22 | 2004-04-07 | ��ʽ���綫֥ | Machine translation apparatus and method |
US20100223047A1 (en) * | 2009-03-02 | 2010-09-02 | Sdl Plc | Computer-assisted natural language translation |
CN102541843A (en) * | 2010-12-22 | 2012-07-04 | 陈本东 | Device and method for improving mechanical translation quality |
CN102708097A (en) * | 2012-04-27 | 2012-10-03 | 曾立人 | Online computer translation method and online computer translation system |
CN104679735A (en) * | 2013-11-30 | 2015-06-03 | 赵会军 | Pragmatic machine translation method |
CN107391500A (en) * | 2017-08-21 | 2017-11-24 | 阿里巴巴集团控股有限公司 | Text interpretation method, device and equipment |
-
2018
- 2018-05-21 CN CN201810490347.7A patent/CN108763228A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5826219A (en) * | 1995-01-12 | 1998-10-20 | Sharp Kabushiki Kaisha | Machine translation apparatus |
CN1308748A (en) * | 1998-05-04 | 2001-08-15 | 特雷道斯股份有限公司 | Machine-assisted translation tools |
CN1487449A (en) * | 2002-08-22 | 2004-04-07 | ��ʽ���綫֥ | Machine translation apparatus and method |
US20100223047A1 (en) * | 2009-03-02 | 2010-09-02 | Sdl Plc | Computer-assisted natural language translation |
CN101826072A (en) * | 2009-03-02 | 2010-09-08 | Sdl有限公司 | Computer assisted natural language translation |
CN102541843A (en) * | 2010-12-22 | 2012-07-04 | 陈本东 | Device and method for improving mechanical translation quality |
CN102708097A (en) * | 2012-04-27 | 2012-10-03 | 曾立人 | Online computer translation method and online computer translation system |
CN104679735A (en) * | 2013-11-30 | 2015-06-03 | 赵会军 | Pragmatic machine translation method |
CN107391500A (en) * | 2017-08-21 | 2017-11-24 | 阿里巴巴集团控股有限公司 | Text interpretation method, device and equipment |
Non-Patent Citations (1)
Title |
---|
陈旷: "计算机辅助翻译的核心模块和技术论析", 《江汉大学学报(自然科学版)》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109408626A (en) * | 2018-11-09 | 2019-03-01 | 苏州思必驰信息科技有限公司 | The method and device that natural language is handled |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Fevry et al. | Unsupervised sentence compression using denoising auto-encoders | |
McCarthy | A prosodic theory of nonconcatenative morphology | |
Bethard et al. | Semeval-2015 task 6: Clinical tempeval | |
Rastogi et al. | Weighting finite-state transductions with neural context | |
Boudelaa et al. | Aralex: A lexical database for modern standard Arabic | |
KR101726667B1 (en) | Grammar compiling methods, semantic parsing methods, devices, computer storage media, and apparatuses | |
US5640575A (en) | Method and apparatus of translation based on patterns | |
CN107025217A (en) | The synonymous literary generation method of conversion, device, program and machine translation system | |
CN110390003A (en) | Question and answer processing method and system, computer equipment and readable medium based on medical treatment | |
Crane | Generating and parsing classical Greek | |
US20110184723A1 (en) | Phonetic suggestion engine | |
KR20160060821A (en) | Question answering system and method for structured knowledge-base using deep natrural language question analysis | |
JPH1011447A (en) | Translation method and system based upon pattern | |
Park et al. | Neural spelling correction: translating incorrect sentences to correct sentences for multimedia | |
Chetail et al. | The internal structure of chaos: Letter category determines visual word perceptual units | |
CN104239289B (en) | Syllabification method and syllabification equipment | |
CN108763228A (en) | A kind of natural language secondary treatment system | |
Gold et al. | When linearity prevails over hierarchy in syntax | |
Wankerl et al. | An Analysis of Perplexity to Reveal the Effects of Alzheimer's Disease on Language | |
Mishra | Simulating the pāṇinian system of sanskrit grammar | |
Bier et al. | Adaptive math-to-speech interface | |
Marchand et al. | Evaluating automatic syllabification algorithms for English | |
CN107301172A (en) | Data processing method and storage medium | |
Singh | Multilingualism, sociolinguistics and theories of linguistic form: Some unfinished reflections | |
Busemann | Surface transformations during the generation of written German sentences |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181106 |