CN102662490A - Inputting method having assistant translation function - Google Patents
Inputting method having assistant translation function Download PDFInfo
- Publication number
- CN102662490A CN102662490A CN2012100992898A CN201210099289A CN102662490A CN 102662490 A CN102662490 A CN 102662490A CN 2012100992898 A CN2012100992898 A CN 2012100992898A CN 201210099289 A CN201210099289 A CN 201210099289A CN 102662490 A CN102662490 A CN 102662490A
- Authority
- CN
- China
- Prior art keywords
- translation
- sentence
- vocabulary
- tree
- input
- 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
An inputting method having an assistant translation function is used for giving a suggestion about related contents when translating. The method is characterized by reading a sentence to be translated; recognizing all likely terms in the sentence to be translated; acquiring translation of the terms from a dictionary of the field corresponding to the sentence; acquiring translation of non-term words from a common bilingual electronic dictionary, establishing a high-priority tree, and placing probability of translation result-to-word correspondence in the high-priority tree; simultaneously matching a normal-word tree and the high-priority tree in the input method when the user inputs letters, and comprehensively sequencing the match result according to the probability; clearing a high-priority tree word list of the sentence when completing inputting the sentence; and repeating the above steps until completing inputting all sentences to be translated. The invention integrates word stocks of professional fields to the inputting method, and effectively improves the inputting efficiency of texts in a translation process.
Description
Technical field
The present invention relates to the input method technical field of spoken and written languages, relate in particular in the language translation, provide the quick input method of spoken and written languages of computer-aided translation function.
Background technology
Along with social science and technology and rapid development of economy, the information interchange between the different language has become the important component part in the information exchange, and the thing followed to the demand of various language service, especially translation services also more and more widely.
At present the ubiquitous problem of translation industry is: translate requirements grows with each passing day but to translate the especially high-end translator of strength seriously in short supply, and translation efficient is low, cost is high.The translator mainly contains following obstacle: 1) require a great deal of time and inquire about and translation speciality vocabulary; 2) need improve the Personal Skills through long-term study, be difficult to utilize the knowledge and the data of same trade accumulation; 3) lack practicality and weevil-proof and reduce a large amount of repetitions and simple work in the translation process; 4) lack manual decision and machine intelligence instrument that effectively combine, ability dealing with complicated translation duties.
Because the existence of above-mentioned obstacle; Translate requirements in the face of various fields; The existence of a large amount of specialized vocabularies causes the translator need carry out a large amount of interface switchings through regular meeting, obtains the translation of corresponding specialized vocabulary through various electronic dictionaries or dictionary web sites, and then switches go back to the translation interface; Waste great amount of time, also interrupted translator's thinking simultaneously easily.
Language input method is a necessary instrument in the translation; To the input of the target language input method that places one's entire reliance upon; But input method commonly used at present (for example, can be remembered speech or sentence that the user imported, when the user carries out the input of same words or sentence once more except having certain historical trace function; Provide prompting fast), more help is provided can not for translator's translation.The translator runs into the problem of utilizing input method not import fast targetedly through regular meeting; Need the words of input in the corresponding dictionary of input method, not exist; Perhaps because the relation of frequency of utilization; Come the position after relatively leaning on, still need the translator in the words selectionbar in addition repeatedly page turning just can find the words of needs input, this has produced bigger bottleneck to the raising of translator's translation efficiency undoubtedly.
Granted publication number discloses in a kind of translation process the quick input method based on scene lexicon for the Chinese invention patent of CN100524313C; In the process that the user imports; Read and discern the vocabulary of waiting to translate in the sentence to be translated; From scene lexicon, obtain then with said and wait to translate the vocabulary corresponding entries; This entry is injected in the dictionary that the employed input method of user carries, and the DISPLAY ORDER priority of this entry in the input method input frame is transferred to the position that can be chosen input first, accelerate user's input speed.Wherein, In will from scene lexicon, obtaining the process that corresponding entries is injected into the employed input method dictionary of user; Need at first detect input method, take different modes that the entry of obtaining is injected in the dictionary that the employed input method of user carries according to different input methods.This invention has solved in the existing input method can not carry out targetedly the deficiency of input fast to particular text especially cypher text; Under the situation that does not change user's input method custom; The scheme that adopts dictionary and vocabulary to inject, on the basis of existing input method, through temporary transient in the process of importing in translation, revise the DISPLAY ORDER (word frequency) of words targetedly; The input efficiency of the translation process Chinese version that effectively improves has been realized the quick input in the translation process.
But because above-mentioned invention is free on outside the input method; Cause in the middle of the process of its application; Vocabulary and the vocabulary in the input method in the middle of the scene lexicon can not organically combine; No matter the probability of its appearance of vocabulary in the scene lexicon is big or small, appear at the foremost part (even its probability of occurrence is extremely low) of input frame without exception, influenced user's input to a certain extent.And this invention need be injected into the relevant entry in the scene lexicon in the middle of the input method dictionary, need carry out mutual with input method, and the difficulty of application is relatively large.
Summary of the invention
The present invention proposes in order to address the above problem just; Its purpose is; A kind of quick input method with supplementary translation function is provided, and it can not change in translation process under user's the situation of focus, and the professional domain dictionary is integrated in the middle of the input method; Provide various translation promptings according to the input of treating translation of the sentence and user; Make the user can choose vocabulary or sentence that they need import fast, the input efficiency of the translation process Chinese version that effectively improves has been realized the quick input in the translation process.
Quick input method with supplementary translation function of the present invention comprises the steps: to read and treats translation of the sentence; All possible term in the translation of the sentence is treated in identification; The translation of from the dictionary of the pairing field of this sentence, obtaining term; From common bilingual electronic dictionary, obtain remain to be translated the translation of non-term vocabulary; Set up a high priority tree, these translation results and the corresponding probability of vocabulary are put into this high priority tree; When user's input characters, mate central general words tree of input method and high priority tree simultaneously, and carry out carrying out integrated ordered the result of coupling according to its probability; Behind a word EOI, remove this high priority tree word lists; Repeat above-mentioned process, finish until the input of remaining translation of the sentence.
Wherein, when waiting to translate non-term vocabulary translation, the resource of being utilized is common bilingual electronic dictionary; When waiting to translate the term translation; The resource of utilizing is to wait to translate the corresponding field term dictionary in sentence field with current; Like this; The term translation of being obtained is exactly the pairing term translation in this field, and can not obtain the translation (identical term in different field often have different interpretative system) of this term in other field.
According to the present invention; Solved existing input method and various computer-aided translation software can not be intuitively efficiently to the vocabulary deficiency pointed out of technical term in the cypher text of treating in the translation process; In translation process, do not change under user's the situation of focus; The professional domain dictionary is integrated in the middle of the input method, provides various translations promptings, make the user can choose vocabulary or sentence that they need import fast according to the input of treating translation of the sentence and user; The input efficiency of the translation process Chinese version that effectively improves has been realized the quick input in the translation process.
Description of drawings
Fig. 1 is the system construction drawing of the input method with supplementary translation function that the present invention carried.
Fig. 2 is the method flow diagram of the input method with supplementary translation function that the present invention carried.
Fig. 3 is the term prompt facility synoptic diagram that the embodiment of the invention showed.
The input window synoptic diagram that Fig. 4 carries for the embodiment of the invention.
The input window synoptic diagram that Fig. 5 carries for the embodiment of the invention.
The input window synoptic diagram that Fig. 6 carries for the embodiment of the invention.
Embodiment
Below, through coming with reference to accompanying drawing characteristic of the present invention, advantage at length are described that it is clearer that the present invention will become.
Fig. 1 is the system construction drawing of the input method with supplementary translation function carried of the present invention, utilizes this to scheme to explain System Operation of the present invention.As shown in Figure 1, input system 100 of the present invention contains input acquisition module 101, original text read module 102, term identification module 103, dictionary enquiry module 104, candidate's generation module 105 and shows output module 106.
Original text read module 102 reads the original text that the user need translate, and sends it to term identification module 103.
The vocabulary of waiting to translate in the original text that 104 pairs of dictionary enquiry modules send finds corresponding translation in dictionary, and notes.For the term in the original text; According to the realm information of user's appointment, inquire about the relevant glossary in corresponding field, this glossary can be that manual work is compiled or in the middle of the extensive corpus of text in this area, extracted; For other vocabulary; Inquire about common bilingual electronic dictionary, and vocabulary that inquires and probability (probability of vocabulary is in the middle of extensive corpus of text, to add up, and adjusts according to user's use) thereof are deposited in the middle of the high priority tree.
105 pairs of candidate's generation modules are by the words tree that input generated and high priorities tree (words tree all is to organize with the pattern that TRIE sets with the high priority tree) merging of user, and the vocabulary in two trees carries out integrated ordered.Whether integrated ordered method is whether be User Defined vocabulary with the corresponding shaping data of each vocabulary if comprising in these data, and whether the length of corresponding input block is the data in the high priority tree, be the node on the whole sentence optimal path, score; Wherein score comprises the term probability, probabilistic language model, User Defined weight.All data sort according to the size of this shaping number.
The result of ordering is exported by demonstration output module 106, supplies the user to select.
According to said structure of the present invention, owing to introduced the information of treating translation of the sentence, glossary is introduced in the middle of the input method, reduced the user inquires about dictionary in translation process work; In the middle of user's input process, user's input is combined with glossary information, can reduce user's the stroke when carrying out translation, thereby accelerate the speed of translation.
Then, the system architecture diagram of the input method of being carried in conjunction with Fig. 1 the present invention with supplementary translation function and the method flow diagram of the input method with supplementary translation function that Fig. 2 the present invention is carried are explained orally operation system of the present invention and method with an embodiment.
Following embodiment only is used to explain the present invention, and obviously the present invention is not limited to this.
Suppose that user's desire uses input method of the present invention to translate " both the adsorption rate and the desorption rate of gold could reach over 98 % under suitable conditions " the words.According to user's operation, original text read module 102 will be located the sentence that he will translate in the text of user's translation, and the words is passed to term identification module 103 (step S01); Term identification module 103 identifications term wherein, such as: " adsorption rate ", " desorption rate ", " conditions " or the like (step S02); Send other all vocabulary in these terms and the sentence to dictionary enquiry module 104 then; The glossary that dictionary enquiry module 104 is relevant with term partial query field; The bilingual dictionary that the popular word partial query is common; Translate speech with what inquiry obtained; Correspondingly be presented at the corresponding below of being translated speech on the system interface with the mode of vertical arrangement respectively like " adsorption rate ", " adsorption efficiency ", " desorption rate ", " suiting ", " operating mode " etc., as shown in Figure 3, system translates speech with all these and is stored in the high priority vocabulary candidate list (step S03); When the user keyed in letter, the input that the input acquisition module obtains the user was delivered in candidate's generation module; Candidate's generation module 105 generates a popular word candidate list according to user's input; And user's input and the vocabulary in the high priority vocabulary candidate list mated; If can find corresponding matched therein; Just the vocabulary in vocabulary that matees in the high priority word lists and the popular word candidate list is carried out integrated orderedly according to parameters such as length, probability, generate a new word lists (step S04); And this word lists given show that output module 106 shows (step S05).In this example; For example when user's input " x "; Candidate's generation module 105 has mated " adsorption rate " and " absorption property " two speech in the high priority word lists; And vocabulary such as the vocabulary D score in these vocabulary and the popular word candidate list, " type ", " property " are carried out integrated ordered, with the ordering the result be shown to the user, as shown in Figure 4.If the user hopes input " adsorption rate " this vocabulary, get as long as key in the whole vocabulary that " 2 " just can this is desirable " adsorption rate ", that is, only need knock two keys and can import this speech, saved the time of importing greatly.Same, if the user hopes input " thermal desorption " this vocabulary, after only needing to key in " r ", key in " 1 " or space bar again, just can accomplish the input of whole vocabulary, as shown in Figure 5.After the user accomplishes input in short; When preparing to translate next; System empties the data in the high priority tree, rebulids its corresponding priorities tree according to next sentence again, so just can guarantee in the translation of carrying out each sentence; Do not have reciprocal influence, guarantee the efficient of translation.As shown in Figure 6; The user is translating second sentence: in " Study on Removing Flue Gas Entrained by Regenerated Catalyst "; When keying in " r " again; " thermal desorption " just do not appear in the middle of the tabulation, and " combustion gas " relevant with this appears in the middle of the tabulation.Through preliminary test, the higher search dog input method of this function ratio existing market occupation rate can be saved the button operation number of times more than at least 1/3, especially can minimizing group speech, operation about 80% the most consuming time in input process such as page turning.
Though embodiment of the present invention has been described as described above, described content is not in order to direct qualification scope of patent protection of the present invention.Obviously to those skilled in the art, do not breaking away within the disclosed spirit and scope, the change or the distortion of on the formal and details of implementing, doing all belong to scope of patent protection of the present invention.Therefore scope of patent protection of the present invention still must with this instructions appending claims the person of being defined be as the criterion.
Claims (5)
1. the input method with supplementary translation function is used for when translating, giving the prompting of user's related content, it is characterized in that this method may further comprise the steps:
Read and treat translation of the sentence;
All possible term in the translation of the sentence is treated in identification;
The translation of from the dictionary of the pairing field of this sentence, obtaining term; From common bilingual electronic dictionary, obtain remain to be translated the translation of non-term vocabulary; Set up a high priority tree, these translation results and the corresponding probability of vocabulary are put into this high priority tree;
When user's input characters, mate central general words tree of input method and high priority tree simultaneously, and carry out carrying out integrated ordered the result of coupling according to its probability;
Behind a word EOI, remove this high priority tree word lists;
Repeat above-mentioned process, finish until the input of remaining translation of the sentence.
2. according to the input method of claim 1, it is characterized in that: general words tree and high priority tree all are to organize with the pattern of TRIE tree.
3. according to the input method of claim 1, it is characterized in that: for treating that the word in the translation of the sentence handles respectively according to the character of word; Promptly for the term of treating in the translation of the sentence; Realm information according to user's appointment; Inquire about the relevant glossary in corresponding field, this glossary is that manual work is compiled or in the middle of the extensive corpus of text in corresponding field, extracted automatically, for the vocabulary outside the term; Inquire about common bilingual electronic dictionary, and the vocabulary that inquires and probability thereof are deposited in the middle of the high priority tree.
4. according to the input method of claim 1, it is characterized in that: the probability of vocabulary is in the middle of extensive corpus of text, to add up, and adjusts according to user's use.
5. according to the input method of claim 1, it is characterized in that: integrated ordered method is whether be User Defined vocabulary with the corresponding shaping data of each vocabulary if comprising in these data; Whether whether the length of corresponding input block is the data in the high priority tree, be the node on the whole sentence optimal path; Score, wherein score comprises term probability, probabilistic language model; The User Defined weight, all data sort according to the size of this shaping number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100992898A CN102662490A (en) | 2012-04-08 | 2012-04-08 | Inputting method having assistant translation function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100992898A CN102662490A (en) | 2012-04-08 | 2012-04-08 | Inputting method having assistant translation function |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102662490A true CN102662490A (en) | 2012-09-12 |
Family
ID=46771994
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012100992898A Pending CN102662490A (en) | 2012-04-08 | 2012-04-08 | Inputting method having assistant translation function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102662490A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104572632A (en) * | 2014-12-25 | 2015-04-29 | 语联网（武汉）信息技术有限公司 | Method for determining translation direction of word with proper noun translation |
| CN104834633A (en) * | 2015-05-29 | 2015-08-12 | 厦门大学 | Cloud translation input method and system |
| CN105069000A (en) * | 2015-08-24 | 2015-11-18 | 中译语通科技（北京）有限公司 | Interactive prediction input method |
| CN105094358A (en) * | 2014-05-20 | 2015-11-25 | 富士通株式会社 | Information processing device and method for inputting target language characters through outer codes |
| CN106528546A (en) * | 2016-10-31 | 2017-03-22 | 用友网络科技股份有限公司 | ERP term machine translation method |
| CN109032377A (en) * | 2018-07-12 | 2018-12-18 | 广州三星通信技术研究有限公司 | The method and apparatus of output input method candidate word for electric terminal |
| CN109241543A (en) * | 2018-09-19 | 2019-01-18 | 传神语联网网络科技股份有限公司 | The preconditioning technique of consistency translationese |
| CN110309517A (en) * | 2019-07-02 | 2019-10-08 | 北京百度网讯科技有限公司 | Expression official documents and correspondence processing method, device, system and storage medium |
| CN113687883A (en) * | 2020-05-18 | 2021-11-23 | 阿里巴巴集团控股有限公司 | Data display method and device, electronic equipment and computer readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034407A (en) * | 2007-03-30 | 2007-09-12 | 传神联合（北京）信息技术有限公司 | Quick input method based on scene lexicon in process of translation |
| CN101290632A (en) * | 2008-05-30 | 2008-10-22 | 北京搜狗科技发展有限公司 | Input method for user words participating in intelligent word-making and input method system |
| CN102193643A (en) * | 2010-03-15 | 2011-09-21 | 北京搜狗科技发展有限公司 | Word input method and input method system having translation function |
-
2012
- 2012-04-08 CN CN2012100992898A patent/CN102662490A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034407A (en) * | 2007-03-30 | 2007-09-12 | 传神联合（北京）信息技术有限公司 | Quick input method based on scene lexicon in process of translation |
| CN101290632A (en) * | 2008-05-30 | 2008-10-22 | 北京搜狗科技发展有限公司 | Input method for user words participating in intelligent word-making and input method system |
| CN102193643A (en) * | 2010-03-15 | 2011-09-21 | 北京搜狗科技发展有限公司 | Word input method and input method system having translation function |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105094358A (en) * | 2014-05-20 | 2015-11-25 | 富士通株式会社 | Information processing device and method for inputting target language characters through outer codes |
| CN104572632A (en) * | 2014-12-25 | 2015-04-29 | 语联网（武汉）信息技术有限公司 | Method for determining translation direction of word with proper noun translation |
| CN104572632B (en) * | 2014-12-25 | 2017-07-04 | 武汉传神信息技术有限公司 | A kind of method in the translation direction for determining the vocabulary with proper name translation |
| CN104834633A (en) * | 2015-05-29 | 2015-08-12 | 厦门大学 | Cloud translation input method and system |
| CN105069000A (en) * | 2015-08-24 | 2015-11-18 | 中译语通科技（北京）有限公司 | Interactive prediction input method |
| CN106528546A (en) * | 2016-10-31 | 2017-03-22 | 用友网络科技股份有限公司 | ERP term machine translation method |
| CN109032377A (en) * | 2018-07-12 | 2018-12-18 | 广州三星通信技术研究有限公司 | The method and apparatus of output input method candidate word for electric terminal |
| CN109241543A (en) * | 2018-09-19 | 2019-01-18 | 传神语联网网络科技股份有限公司 | The preconditioning technique of consistency translationese |
| CN110309517A (en) * | 2019-07-02 | 2019-10-08 | 北京百度网讯科技有限公司 | Expression official documents and correspondence processing method, device, system and storage medium |
| CN110309517B (en) * | 2019-07-02 | 2023-10-17 | 北京百度网讯科技有限公司 | Expression document processing method, device, system and storage medium |
| CN113687883A (en) * | 2020-05-18 | 2021-11-23 | 阿里巴巴集团控股有限公司 | Data display method and device, electronic equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102662490A (en) | Inputting method having assistant translation function | |
| CN107291783B (en) | Semantic matching method and intelligent equipment | |
| CN102662935A (en) | Interactive machine translation method and machine translation system | |
| US7769804B2 (en) | Server side search with multi-word word wheeling and wildcard expansion | |
| CN106708893A (en) | Error correction method and device for search query term | |
| CN112256860A (en) | Semantic retrieval method, system, equipment and storage medium for customer service conversation content | |
| CN111814465A (en) | Information extraction method and device based on machine learning, computer equipment and medium | |
| US20070164782A1 (en) | Multi-word word wheeling | |
| CN103970798A (en) | Technology for searching and matching data | |
| CN100592385C (en) | Method and system for performing speech recognition on multi-language name | |
| CN114676255A (en) | Text processing method, device, equipment, storage medium and computer program product | |
| CN115080694A (en) | Power industry information analysis method and equipment based on knowledge graph | |
| CN109492083A (en) | A method of more wheel human-computer intellectualizations are realized based on list content | |
| CN103177039A (en) | Data processing method and data processing device | |
| Alipour et al. | Learning bilingual word embedding mappings with similar words in related languages using GAN | |
| US9208134B2 (en) | Methods and systems for tokenizing multilingual textual documents | |
| CN103164398A (en) | Chinese-Uygur language electronic dictionary and automatic translating Chinese-Uygur language method thereof | |
| CN103164397A (en) | Chinese-Kazakh electronic dictionary and automatic translating Chinese- Kazakh method thereof | |
| Xu et al. | Product features mining based on Conditional Random Fields model | |
| CN103164395A (en) | Chinese-Kirgiz language electronic dictionary and automatic translating Chinese-Kirgiz language method thereof | |
| CN101436205A (en) | Method and apparatus for enquiring unique word by explanation | |
| CN104641367B (en) | For formatting formatting module, the system and method for electronic character sequence | |
| CN110688840B (en) | Text conversion method and device | |
| CN101206525A (en) | Disorder input method | |
| Nguyen et al. | Named entity recognition in Vietnamese tweets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120912 |