CN102662933A - Distributive intelligent translation method - Google Patents
Distributive intelligent translation method Download PDFInfo
- Publication number
- CN102662933A CN102662933A CN2012100853077A CN201210085307A CN102662933A CN 102662933 A CN102662933 A CN 102662933A CN 2012100853077 A CN2012100853077 A CN 2012100853077A CN 201210085307 A CN201210085307 A CN 201210085307A CN 102662933 A CN102662933 A CN 102662933A
- Authority
- CN
- China
- Prior art keywords
- word
- translation
- data base
- sentence
- original text
- 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 a distributive intelligent translation method which includes the steps of firstly, importing an original text to be translated on a client, analyzing the original text, and dividing the original text into maximum identifiable units (paragraphs); secondly, splitting each paragraph into sentences; thirdly, searching each sentence corresponding to data in a database, directly performing the step 5 if a match for the sentence is searched, and directly performing the step 4 if the match is not searched; fourthly, combining one word, one word adjacent to the former and the Nth word adjacent to the former for each sentence, performing matching search for a plurality of word groups and single minimal word groups, obtained by combining, relative to the data in the local database; fifthly, returning search results to an user interface for the user to select, and recording the original text and a translated text in the local database and a dynamic database for future translation; and sixthly, giving out prompts. By the method providing a professional translation word bank, translating efficiency and translating speed in translation are increased.
Description
Technical field
The present invention relates to interpretation method, be specifically related to a kind of distributed intelligence interpretation method.
Background technology
The CAT software development has formed different techniques principles such as translation, memory, storage to today.With regard to the data memory storehouse; General translation software is in when translation, be with word, speech or the sentence of single delegation and translation as a language to depositing database in, and these data that deposit in are only used when the translation subsequent content as this machine that translation software is installed; When the other one piece of article of translation; If the memory content of above chapter article does not derive, when translation next chapter article, do not import the memory content of above chapter again, be to share the memory content of translating before like this.This method that generates data in this locality, it is just more loaded down with trivial details to operate, and adds data base and can't in the process of a project, be used in real time, makes translation efficiency very low like this.
Raising along with to translation output and quality requirements has higher requirement to translation software, accomplishes accurately, fast the more intelligentized translation software of the translation brief of one piece of article is improved translation output and quality.To a project; Existing translation technology, not high to translation quality, same piece of writing translation article is divided into several parts of translations in a team; The translation of understanding the different interpreters of appearance to translate a word of identical implication in the translation and translating is inconsistent; Translation team is difficult to share the information in the data memory storehouse, and in translation process, not the whole translation information of intelligent memory.On proprietary vocabulary and accuracy thereof, the existing proprietary vocabulary that divides word algorithm to cause can't to mate to occurring in one, thus cause the accuracy of translation article to reduce greatly.
Simultaneously, existing translation software has following several big shortcoming: 1. proprietary lexicon (phrase) lacks and is unprofessional; 2. translation efficiency is low; 3. there is not project collaborative work function, the translation information of whole translation team in can't the Real-Time Sharing translation in the translation process; 4. translation project does not have real-time intelligent and learns whole translation process.
Summary of the invention
The present invention has overcome the deficiency of prior art; A kind of distributed intelligence interpretation method is provided; This method mainly provides the relatively translation lexicon of specialty, has improved translation efficiency and the translation speed in the translation process, has solved the technical matters that exists in the prior art.
The present invention adopts following technical scheme:
A kind of distributed intelligence interpretation method comprises the steps:
Step 1 at the original text that the client importing need be translated, is analyzed original text, and original text is divided into maximum recognition unit---section;
Step 2 splits by sentence each section;
Step 3 is searched each and the data in the database, and above-mentioned database is divided into local data base that is positioned at client and the background data base that is positioned at server end; And the backstage database is divided into static database and dynamic data base, all inquires about local data base during match query earlier, if find the coupling of sentence then directly to get into step 5; If do not find; Look into static database again, static database has been found and has then also directly been got into step 5, under the situation that static database is not found; Inquire about dynamic data base at last if find the coupling of sentence also directly to get into step 5, then do not get into step 4 if do not find;
Step 4 makes up sentence again according to 1 to N word of adjacent words, a plurality of phrases after will making up again and the phrase of single least unit all carry out matched and searched with the data of local data base earlier; If find then progressive rapid 5, if do not find, then with static database in data carry out matched and searched; If find then progressive rapid 5, if do not find, then with dynamic data base in data carry out matched and searched;
Step 5, inquire the result after, return user interface in real time and supply the user to select, original text and the translation with translation records in local data base and the dynamic data base simultaneously, use when translating later on;
Step 6; With the outcome record behind the query composition in the internal memory of client; And sentence the mode of mark at original text, the result who checks out is significantly showed, when needs use; It is selective that click can be presented at user interface with translation here at once, clicks once more that can at once translation to be presented at translation behind the translation capable.
Further be:
Above-mentioned local data base is positioned at each client, and by original text, the translation information of each record interpreter oneself translation, original text, translation that the every translation of interpreter is one all add in the dynamic data base of this local data base and server end.
In the above-mentioned step 4 sentence is made up according to an adjacent word to an adjacent N word again, is meant the sentence that contains N word is made up fractionation as follows:
Step 4-1: with first word and second word, first, second and the 3rd word, first, second, third and the 4th word until first, second ... Make up with N-1 word and to split into multiple result;
Step 4-2: with second word and the 3rd word, second, third with the 4th word, second, third, the 4th and the 5th word, until second, third, the 4th ... Make up with N word and to split into multiple result;
Step 4-3: with the 3rd word and the 4th word, the the 3rd, the 4th and the 5th word, the the the 3rd, the 4th, the 5th and the 6th word is until the 3rd, the 4th ... Make up with N word and to split into multiple result;
And by that analogy until step 4-N-1.
The data that above-mentioned static database deposits in are made up of proprietary vocabulary, and the information recorded sum had comprised vocabulary and sentence when the data in the dynamic data base then were a plurality of client real time translation.
Be to serve as a mark in the above-mentioned step 1, every section words split come through the new line symbol between section and the section.
Be that the end of the sentence punctuation mark is served as a mark in the above-mentioned step 2, every words split come.
Compared with prior art, the invention has the beneficial effects as follows:
1) collaborative work of project is provided, a plurality of clients can realize the collaborative work to same project simultaneously in different physical networks, thereby have saved more resources;
2) method for splitting is scientific and reasonable, and Query Result accurately can be provided;
3) it is scientific and reasonable to search matching process, has improved translation efficiency;
4) in the middle of translation is carried out; The follow-up personnel of words and phrases of first people's translation can use; If the vicious words of words and phrases of translation can be unified to make amendment when project finishes, these various deviations that occurred when just having avoided different people to translate identical content are inconsistent;
5) information recorded also can be examined in the dynamic data base.
Description of drawings
Fig. 1 is a workflow diagram of the present invention;
Fig. 2 is a detailed realization schematic diagram of the present invention.
Embodiment
Below in conjunction with accompanying drawing the present invention is done further elaboration.
A kind of distributed intelligence interpretation method, as shown in Figure 1, comprise the steps:
Step 1 at the original text that the client importing need be translated, is analyzed original text, serves as a mark through the new line symbol between section and the section, and original text is divided into maximum recognition unit---section;
Step 2 serves as a mark the end of the sentence punctuation mark, and each section words are split into every words.
Step 3 is searched each and the data in the database, and above-mentioned database is divided into local data base that is positioned at client and the background data base that is positioned at server end; And the backstage database is divided into static database and dynamic data base, all inquires about local data base during match query earlier, if find the coupling of sentence then directly to get into step 5; If do not find; Look into static database again, static database has been found and has then also directly been got into step 5, under the situation that static database is not found; Inquire about dynamic data base at last if find the coupling of sentence also directly to get into step 5, then do not get into step 4 if do not find;
Step 4 makes up sentence again according to 1 to N word of adjacent words, be meant specifically the sentence that contains N word is made up fractionation as follows:
Step 4-1: with first word and second word, first, second and the 3rd word, first, second, third and the 4th word until first, second ... Make up with N-1 word and to split into multiple result;
Step 4-2: with second word and the 3rd word, second, third with the 4th word, second, third, the 4th and the 5th word, until second, third, the 4th ... Make up with N word and to split into multiple result;
Step 4-3: with the 3rd word and the 4th word, the the 3rd, the 4th and the 5th word, the the the 3rd, the 4th, the 5th and the 6th word is until the 3rd, the 4th ... Make up with N word and to split into multiple result;
And by that analogy until step 4-N-1; The a plurality of phrases after will making up again and the phrase of single least unit all carry out matched and searched with the data of local data base earlier; If find then progressive rapid 5, if do not find, then with static database in data carry out matched and searched; If find then progressive rapid 5, if do not find, then with dynamic data base in data carry out matched and searched;
Step 5, inquire the result after, return user interface in real time and supply the user to select, original text and the translation with translation records in local data base and the dynamic data base simultaneously, use when translating later on;
Step 6; With the outcome record behind the query composition in the internal memory of client; And sentence the mode of mark at original text, the result who checks out is significantly showed, when needs use; It is selective that click can be presented at user interface with translation here at once, clicks once more that can at once translation to be presented at translation behind the translation capable.
Above-mentioned local data base is positioned at each client, and by original text, the translation information of each record interpreter oneself translation, original text, translation that the every translation of interpreter is one all add this local data base to.The data that static database deposits in are made up of proprietary vocabulary, and the information recorded sum had comprised vocabulary and sentence when the data in the dynamic data base then were a plurality of client real time translation.
Be described in detail the key problem in technology point of this patent below again: key point comprises: split participle, combination, prompting, memory, shared data library information, local data base, static database, dynamic data base.
Local data base related in this patent is meant: by the original text/translation information of each record interpreter oneself translation, former/translation that the every translation of interpreter is is all with adding this local data base to, when next perhaps later article of translation; After can look into this database earlier; If found, then subsequent action does not need to do again, if do not find; Look into static database again; Static database has been found does not then need differential attitude database again, under the situation that static database is not all found, looks into dynamic data base at last.
Static database is meant: deposit accumulative proprietary vocabulary (non-single individual character or speech), this proprietary vocabulary is through examination and recognition, errorless a kind of information accumulation.Stored 2 field contents in the static database, a row Chinese one row are English, and Chinese and English contrasts one by one, and the implication of Chinese has contrasted English; English conversely implication has also contrasted unique Chinese implication.
Dynamic data base: article of the every translation of interpreter, original text and translation is right as a language, be stored in this dynamic data base.Stored 2 field contents in the database, a row Chinese one row are English, and Chinese and English contrasts one by one, and the implication of Chinese has contrasted English; English conversely implication has also contrasted unique Chinese implication.
This patent is that one section translation of one section original text (translation is for empty) is shown when one piece of article of translation.Original text is used for post analysis to be used, and translation is used to show the information after the corresponding translation.
Come again below to be described in detail the process that realizes translation with the mode of a specific embodiment:
Split
As shown in Figure 2: as after opening one piece of original text, to analyze as follows:
Illustrate, earlier with original text by analyzing, in one piece of original text, be to serve as a mark between section and the section with the new line symbol, can every section talk about to distinguish through this mark in the program and come.Being presented in the system is exactly one section original text, one section translation (being empty); One section original text, one section translation (being empty) Then the section in every a word be with fullstop (..) or exclamation (! ! ) as end mark, through this mark every words difference is come in the program, this process is exactly a split process.After splitting completion; Each elder generation and local data base after being about to split carry out matched and searched; If found then corresponding this follow-up participle, combination, query actions do not need, do not search if find to compare to static database again, if in static state, found content; Then this follow-up participle, combination, query actions do not need accordingly, have improved total system efficient.If do not find again and in dynamic data base, carry out work such as matched and searched.
Divide word combination
Be with the sentence that in database, does not find, according to carrying out participle sentence by sentence in this step.Here be example with Chinese, for example: I am Chinese.Earlier this sentence is split as speech one by one: " I " " I am " " I be in " " I am a China " " I am Chinese " " be " " in being " " being China " " being Chinese " " in " " China " " Chinese " " state " " compatriots " " people ".Split like this and avoided the careless omission of in static lexicon, inquiring about behind the participle, Query Result accurately can be provided.
Inquiry
Speech (getting rid of single word or word) behind above minute word combination is searched in local data base earlier; Local data base does not find; Then from static data, inquire about again; For example, with " I am " " I be in " " I am a China " " I am Chinese " " in being " " being China " " being Chinese " " China " " Chinese " " compatriots ".If found corresponding content, then the English with correspondence directly is prompted to user's use, has improved work efficiency, does not need manual translation once more.And general dictionary software can only be accomplished the translation to word or group speech commonly used, and can not accomplish translation to whole a word, perhaps more can not accomplish accurate translation to the proprietary vocabulary that occurs in the middle of a word.
Memory
In whole translation process; If in local data, static database or dynamic data base, found occurrence; Then show original text/translation information that this has been arranged in database; If there is not the corresponding matched item, content that then can each original text/translation is right records local data base and dynamic data base (then not remembering only if dynamic data base has had the content identical with original text) by sentence, the method for memory also be according to original text fullstop (..) or exclamation (! ! ), corresponding the fullstop of translation (..) or exclamation (! ! ), every translation is accomplished one section can be with the former/translation after dividing out by sentence to writing dynamic data base.The dynamic data base of this local data base and server end uses when supplying other article of later stage translation.If run into the content that has in the local data base, then directly provide the result, do not need human translation need not remove to read server end yet, improved work efficiency greatly.And along with the increase of translation article, this dynamic data base can be more and more abundanter.
Prompting
With the outcome record behind the query composition in internal memory; And sentence the mode that underlines/change font color at original text; The result who checks out is significantly showed; When needs use, call the reminding module of independent research, click can be presented at user interface at once with translation here and supply the user to select.The user clicks once more that can at once translation to be presented at translation behind the translation capable.
Shared data
This method provides the collaborative work of project, and a plurality of clients can realize the collaborative work to same project simultaneously in different physical networks.Thereby saved more resources.In the middle of translation was carried out, the follow-up personnel of words and phrases of first people's translation can use, if the vicious words of words and phrases of translation can be unified to make amendment when project finishes, these various deviations that occurred when just having avoided different people to translate identical content are inconsistent.Information recorded also can be examined at this in the dynamic data base.
Claims (6)
1. a distributed intelligence interpretation method is characterized in that: comprise the steps:
Step 1 at the original text that the client importing need be translated, is analyzed original text, and original text is divided into maximum recognition unit---section;
Step 2 splits by sentence each section;
Step 3 is searched each and the data in the database, and described database is divided into local data base that is positioned at client and the background data base that is positioned at server end; And the backstage database is divided into static database and dynamic data base, all inquires about local data base during match query earlier, if find the coupling of sentence then directly to get into step 5; If do not find; Look into static database again, static database has been found and has then also directly been got into step 5, under the situation that static database is not found; Inquire about dynamic data base at last if find the coupling of sentence also directly to get into step 5, then do not get into step 4 if do not find;
Step 4 makes up sentence again according to an adjacent word to an adjacent N word, a plurality of phrases after will making up again and the phrase of single least unit all carry out matched and searched with the data of local data base earlier; If find then progressive rapid 5, if do not find, then with static database in data carry out matched and searched; If find then progressive rapid 5, if do not find, then with dynamic data base in data carry out matched and searched;
Step 5, inquire the result after, return user interface in real time and supply the user to select, original text and the translation with translation records in local data base and the dynamic data base simultaneously, use when translating later on;
Step 6; With the outcome record behind the query composition in the internal memory of client; And sentence the mode of mark at original text, the result who checks out is significantly showed, when needs use; It is selective that click can be presented at user interface with translation here at once, clicks once more that can at once translation to be presented at translation behind the translation capable.
2. distributed intelligence interpretation method according to claim 1; It is characterized in that: described local data base is positioned at each client; By original text, the translation information of each record interpreter oneself translation, the dynamic data base that the original text that the every translation of interpreter is, translation all add this local data base and server end to.
3. distributed intelligence interpretation method according to claim 1 is characterized in that: in the described step 4 sentence is made up according to an adjacent word to an adjacent N word again, be meant the sentence that contains N word is made up fractionation as follows:
Step 4-1: with first word and second word, first, second and the 3rd word, first, second, third and the 4th word until first, second ... Make up with N-1 word and to split into multiple result;
Step 4-2: with second word and the 3rd word, second, third with the 4th word, second, third, the 4th and the 5th word, until second, third, the 4th ... Make up with N word and to split into multiple result;
Step 4-3: with the 3rd word and the 4th word, the the 3rd, the 4th and the 5th word, the the the 3rd, the 4th, the 5th and the 6th word is until the 3rd, the 4th ... Make up with N word and to split into multiple result;
And by that analogy until step 4-N-1.
4. distributed intelligence interpretation method according to claim 2 is characterized in that: the data that described static database deposits in are made up of proprietary vocabulary, information recorded sum when the data in the dynamic data base then are a plurality of client real time translation.
5. distributed intelligence interpretation method according to claim 1 is characterized in that: be to serve as a mark through the new line symbol between section and the section in the described step 1, every section words split come.
6. distributed intelligence interpretation method according to claim 1 is characterized in that: be that the end of the sentence punctuation mark is served as a mark in the described step 2, every words split come.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100853077A CN102662933A (en) | 2012-03-28 | 2012-03-28 | Distributive intelligent translation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100853077A CN102662933A (en) | 2012-03-28 | 2012-03-28 | Distributive intelligent translation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102662933A true CN102662933A (en) | 2012-09-12 |
Family
ID=46772428
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012100853077A Pending CN102662933A (en) | 2012-03-28 | 2012-03-28 | Distributive intelligent translation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102662933A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103838718A (en) * | 2014-02-24 | 2014-06-04 | 武汉传神信息技术有限公司 | Translation system and translation method |
| CN103838716A (en) * | 2012-11-27 | 2014-06-04 | 英业达科技有限公司 | System and method for splitting target data to server and client for translation |
| CN103885942A (en) * | 2014-03-18 | 2014-06-25 | 成都优译信息技术有限公司 | Rapid translation device and method |
| CN104346324A (en) * | 2013-07-26 | 2015-02-11 | 英业达科技有限公司 | Word and sentence translation system and method thereof |
| CN104462072A (en) * | 2014-11-21 | 2015-03-25 | 中国科学院自动化研究所 | Input method and device oriented at computer-assisting translation |
| WO2015063536A1 (en) * | 2013-10-28 | 2015-05-07 | Translation Management Systems Ltd. | Networked language translation system and method |
| CN105989428A (en) * | 2015-02-02 | 2016-10-05 | 成都优译信息技术有限公司 | Method for automatically distributing manuscript |
| CN106294186A (en) * | 2016-08-30 | 2017-01-04 | 深圳市悲画软件自动化技术有限公司 | Intelligence software automated testing method |
| CN106844354A (en) * | 2017-01-11 | 2017-06-13 | 中国科学院合肥物质科学研究院 | A kind of webpage takes word Chinese interpretation method and its device |
| CN106933811A (en) * | 2017-02-14 | 2017-07-07 | 南京南瑞继保电气有限公司 | A kind of entry automatic generation method and device |
| CN108595422A (en) * | 2018-04-13 | 2018-09-28 | 卓望信息技术（北京）有限公司 | A method of the bad multimedia message of filtering |
| CN109783826A (en) * | 2019-01-15 | 2019-05-21 | 四川译讯信息科技有限公司 | A kind of document automatic translating method |
| CN110211570A (en) * | 2019-05-20 | 2019-09-06 | 北京百度网讯科技有限公司 | Simultaneous interpretation processing method, device and equipment |
| CN112164403A (en) * | 2020-09-27 | 2021-01-01 | 江苏四象软件有限公司 | Natural language processing system based on artificial intelligence |
-
2012
- 2012-03-28 CN CN2012100853077A patent/CN102662933A/en active Pending
Non-Patent Citations (2)
| Title |
|---|
| 梁三云: "机器翻译与计算机辅助翻译比较分析", 《外语电化教学》 * |
| 罗桂琼等: "基于反序词典的中文分词技术研究", 《计算机技术与发展》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103838716A (en) * | 2012-11-27 | 2014-06-04 | 英业达科技有限公司 | System and method for splitting target data to server and client for translation |
| CN104346324B (en) * | 2013-07-26 | 2017-06-20 | 英业达科技有限公司 | Words and phrases translation system and its method |
| CN104346324A (en) * | 2013-07-26 | 2015-02-11 | 英业达科技有限公司 | Word and sentence translation system and method thereof |
| WO2015063536A1 (en) * | 2013-10-28 | 2015-05-07 | Translation Management Systems Ltd. | Networked language translation system and method |
| CN103838718A (en) * | 2014-02-24 | 2014-06-04 | 武汉传神信息技术有限公司 | Translation system and translation method |
| CN103885942A (en) * | 2014-03-18 | 2014-06-25 | 成都优译信息技术有限公司 | Rapid translation device and method |
| CN103885942B (en) * | 2014-03-18 | 2017-09-05 | 成都优译信息技术股份有限公司 | A kind of rapid translation device and method |
| CN104462072B (en) * | 2014-11-21 | 2017-09-26 | 中国科学院自动化研究所 | The input method and device of computer-oriented supplementary translation |
| CN104462072A (en) * | 2014-11-21 | 2015-03-25 | 中国科学院自动化研究所 | Input method and device oriented at computer-assisting translation |
| CN105989428A (en) * | 2015-02-02 | 2016-10-05 | 成都优译信息技术有限公司 | Method for automatically distributing manuscript |
| CN106294186A (en) * | 2016-08-30 | 2017-01-04 | 深圳市悲画软件自动化技术有限公司 | Intelligence software automated testing method |
| CN106844354A (en) * | 2017-01-11 | 2017-06-13 | 中国科学院合肥物质科学研究院 | A kind of webpage takes word Chinese interpretation method and its device |
| CN106933811A (en) * | 2017-02-14 | 2017-07-07 | 南京南瑞继保电气有限公司 | A kind of entry automatic generation method and device |
| CN108595422A (en) * | 2018-04-13 | 2018-09-28 | 卓望信息技术（北京）有限公司 | A method of the bad multimedia message of filtering |
| CN109783826A (en) * | 2019-01-15 | 2019-05-21 | 四川译讯信息科技有限公司 | A kind of document automatic translating method |
| CN109783826B (en) * | 2019-01-15 | 2023-11-21 | 四川译讯信息科技有限公司 | Automatic document translation method |
| CN110211570A (en) * | 2019-05-20 | 2019-09-06 | 北京百度网讯科技有限公司 | Simultaneous interpretation processing method, device and equipment |
| CN110211570B (en) * | 2019-05-20 | 2021-06-25 | 北京百度网讯科技有限公司 | Simultaneous interpretation processing method, device and equipment |
| CN112164403A (en) * | 2020-09-27 | 2021-01-01 | 江苏四象软件有限公司 | Natural language processing system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102662933A (en) | Distributive intelligent translation method | |
| US20210407499A1 (en) | Automatically generating conference minutes | |
| US20190163691A1 (en) | Intent Based Dynamic Generation of Personalized Content from Dynamic Sources | |
| KR102310650B1 (en) | Coherent question answering in search results | |
| CN104252533B (en) | Searching method and searcher | |
| US7606700B2 (en) | Adaptive task framework | |
| CN103268313B (en) | A kind of semantic analytic method of natural language and device | |
| US20130060769A1 (en) | System and method for identifying social media interactions | |
| KR20080066946A (en) | Adaptive task framework | |
| CA2807494C (en) | Method and system for integrating web-based systems with local document processing applications | |
| WO2010000208A1 (en) | Resource locator suggestions from input character sequence | |
| CN112445894A (en) | Business intelligent system based on artificial intelligence and analysis method thereof | |
| CN110929007A (en) | Electric power marketing knowledge system platform and application method | |
| CN101425086A (en) | Dictionary enquiry method and dictionary enquiry system based on network | |
| Neustein et al. | Mobile speech and advanced natural language solutions | |
| Kim et al. | Compact lexicon selection with spectral methods | |
| CN103235833B (en) | Answer search method and device by the aid of statistical machine translation | |
| Al-Sultany et al. | Enriching tweets for topic modeling via linking to the wikipedia | |
| CN113779961B (en) | Conventional sentence pattern extraction method of natural language text and electronic device | |
| Felicetti | Teaching archaeology to machines: Extracting semantic knowledge from free text excavation reports | |
| Dai | Construction of English and American literature corpus based on machine learning algorithm | |
| Kupiyalova et al. | Semantic search using Natural Language Processing | |
| Oramas et al. | Automatic creation of knowledge graphs from digital musical document libraries | |
| Ren | An interview of cloud translation platform | |
| CN113987146B (en) | Dedicated intelligent question-answering system of electric power intranet |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120912 |