CN114528858A - Computer-aided correction system and method for machine translation - Google Patents
Computer-aided correction system and method for machine translation Download PDFInfo
- Publication number
- CN114528858A CN114528858A CN202210184626.7A CN202210184626A CN114528858A CN 114528858 A CN114528858 A CN 114528858A CN 202210184626 A CN202210184626 A CN 202210184626A CN 114528858 A CN114528858 A CN 114528858A
- Authority
- CN
- China
- Prior art keywords
- candidate
- words
- candidate word
- word
- translation
- 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
- 238000013519 translation Methods 0.000 title claims abstract description 58
- 238000012937 correction Methods 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000000605 extraction Methods 0.000 claims abstract description 40
- 238000012216 screening Methods 0.000 claims abstract description 31
- 230000014616 translation Effects 0.000 claims description 51
- 238000001914 filtration Methods 0.000 claims description 11
- 239000000284 extract Substances 0.000 description 6
- 238000012545 processing Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06F40/58—Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F40/42—Data-driven translation
- G06F40/49—Data-driven translation using very large corpora, e.g. the web
Abstract
The invention discloses a computer-aided correction system and a computer-aided correction method for machine translation, wherein the system comprises the following steps: the original text candidate word extraction module is used for extracting specific regular words in the original text to serve as a candidate word list; the translation search term extraction module is used for determining a translation correction search term according to the position of a cursor in the translation; the candidate word screening module comprises a memory, is used for screening candidate words matched with the search words from a candidate word list provided by the original text candidate word extraction module according to the search words provided by the translated text search word extraction module, takes the screened candidate words as a candidate word set and stores the candidate words into the memory, and the candidate words in the memory are selectively output; and the pull-down menu generation module is used for receiving the candidate words output from the candidate word screening module and generating option labels based on the received candidate words, so that the user can directly select correct contents from the option labels and input the correct contents into the translation. According to the invention, the efficiency of modifying the translation by the user can be effectively improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a computer-aided correction system and method for machine translation.
Background
The development of the machine translation technology brings great convenience to people, and particularly in the aspect of patent application file translation, the translation efficiency can be greatly improved. However, the machine translation engine has some disadvantages, for example, some information such as number, website address, date, etc. often only need to be presented in the translated text as it is, and in reality, the machine translation engine often cannot recognize the information in the fixed format and can translate the information into other characters by mistake. For documents with strict translation quality requirements (e.g., patent applications), any minor errors need to be corrected. In the current online translation tool, after machine translation is completed, most of translation regions are pure text edit boxes, and original texts may be image files which cannot be directly copied with text contents.
Disclosure of Invention
The invention aims to solve the problem that corresponding contents in the original text cannot be quickly input when the machine translation is corrected in the above scene.
In one aspect, the present invention provides a computer-aided correction system for machine-translated translations, comprising: the original text candidate word extraction module is used for extracting specific regular words in the original text by using a regular expression to serve as a candidate word list; the translation search term extraction module is used for determining a translation correction search term according to the position of the cursor in the translation; the candidate word screening module comprises a memory, is used for screening candidate words matched with the search words from a candidate word list provided by the original text candidate word extraction module according to the search words provided by the translated text search word extraction module, takes the screened candidate words as a candidate word set and stores the candidate words into the memory, and the candidate words in the memory are selectively output; and the pull-down menu generation module is used for receiving the candidate words output from the candidate word screening module and generating option labels based on the received candidate words, so that the user can directly select correct contents from the option labels and input the correct contents into the translation.
Preferably, the translation search term extraction module takes the position of the cursor as an end position, takes the position spaced by one character before the position of the cursor as a start position, and takes the character between the start position and the end position as a search term.
Preferably, if the candidate word set is a non-empty set, the translation search word extraction module adds a character position to the starting position, the candidate word screening module screens out a candidate word matched with the new search word from the candidate word list based on the new search word, replaces the candidate word in the candidate word set with the screened candidate word matched with the new search word, and stores the candidate word matched with the new search word in the memory; and repeating the operation until the candidate word set is an empty set, and outputting the candidate word screened last time stored in the memory to the pull-down menu generating module by the candidate word screening module at the moment.
Here, the specific rule word may include at least one of a date, a number, a letter, and a symbol.
In addition, the original candidate word extraction module may perform a deduplication operation after extracting the specific rule word, so that there is no candidate word that appears repeatedly in the candidate word list.
In another aspect, the present invention provides a computer-aided correction method for machine-translated translation, comprising the following steps: extracting specific regular words in the original text by using a regular expression to serve as a candidate word list; determining a corrected search term of the translated text according to the position of a cursor in the translated text; screening candidate words matched with the search words from the candidate word list according to the search words, taking the screened candidate words as a candidate word set and storing the screened candidate words in a memory, wherein the candidate words in the memory are selectively output; and generating an option label according to the output candidate words, so that the user can directly select correct content from the option label and input the correct content into the translation.
According to the technical scheme of the invention, if a user needs to correct the machine translation, when a first letter, number or symbol is input, the system can pop up a drop-down box, all candidate words comprising the letter, number or symbol are displayed in the drop-down box, and the user can enter or use an up-down key to select a correct candidate word to finish the input; if the number of the displayed candidate words is larger, the user can continue to input subsequent letters, numbers or symbols, and the number of the displayed candidate words is reduced along with the increase of the input digit, so that the user can select the correct candidate words more conveniently.
It should be noted that, according to a preferred embodiment of the present invention, the candidate words displayed in the drop-down box are generated after being initially selected and subjected to one or more selection confirmations, so as to ensure that each displayed candidate word is a candidate word with the smallest range and the highest precision, which is more beneficial for the user to quickly select the correct candidate word, thereby greatly improving the input efficiency.
Drawings
The following drawings are only schematic illustrations and explanations of the present invention, and do not limit the scope of the present invention.
FIG. 1 schematically illustrates a computer-aided correction system for machine translation according to an embodiment of the present invention;
FIG. 2 schematically illustrates a method for computer-aided correction of a machine-translated translation, in accordance with an embodiment of the present invention;
FIG. 3 schematically illustrates a filtering process for outputting displayed candidate words according to a preferred embodiment of the present invention;
FIG. 4 is a screenshot of a drop-down box candidate word display interface in accordance with a preferred embodiment of the present invention;
fig. 5 is an interface screenshot of a user input through a drop-down box candidate word selection according to a preferred embodiment of the present invention.
Description of the reference symbols
1 machine translation computer auxiliary correction system
10 original text candidate word extraction module
20 translation search term extraction module
30 candidate word screening module
31 memory
40 pull down menu generation module.
Detailed Description
Exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
FIG. 1 schematically illustrates a computer-aided correction system 1 for machine translation according to an embodiment of the present invention. As shown in fig. 1, the computer-aided correction system 1 for machine-translated translation includes an original candidate word extracting module 10, a translated text search word extracting module 20, a candidate word screening module 30 and a pull-down menu generating module 40, where the candidate word screening module 30 includes a memory 31 for storing data.
Specifically, the original candidate word extraction module 10 is configured to extract a specific regular word in the original text of the translated document by using a regular expression, and use the extracted regular word as a candidate word list; wherein the specific rule word may include at least one of a date, a number, a letter, and a symbol, for example. The regular expression is a conventional technical means for extracting the target content in the text by a person skilled in the art, and is not described in detail here. In addition, it can be understood that, after the original candidate word extraction module 10 performs the preliminary traversal extraction on the specific rule word in the original text of the translated document, there may be a case that the same candidate word appears repeatedly one or more times in the generated candidate word list. If such a candidate word list is directly used without processing, once a phenomenon of candidate word duplication exists, on one hand, the screening efficiency of the subsequent traversal screening on the candidate word list is affected, and on the other hand, repeated candidate words may appear in a drop-down box which is finally presented to the user for selection. Therefore, as a preferred embodiment, the original candidate word extraction module 10 performs a deduplication operation after extracting the specific rule word, so that there is no candidate word appearing repeatedly in the candidate word list.
The translated text search term extraction module 20 is configured to determine a modified search term of the translated text according to a position of a cursor in the translated text. For example, the translated text search term extraction module 20 may use the position of the cursor as the ending position, the position spaced by one character before the position of the cursor as the starting position, and the characters between the starting position and the ending position as the search term.
The candidate word screening module 30 is configured to screen candidate words matching the search word from the candidate word list provided by the original candidate word extraction module 10 according to the search word provided by the translated text search word extraction module 20, use the screened candidate words as a candidate word set, store the screened candidate words in the memory 31, and selectively output the candidate words in the memory 31. Here, it is obvious that, for a certain search word, the candidate word set generated by candidate word screening module 30 after traversing the candidate word list is either an empty set or a non-empty set, and only when there is no candidate word matching the search word in the candidate word list, the generated candidate word set is the empty set, and otherwise the candidate word set is the non-empty set. For non-empty sets, the shorter the term, the greater the number of candidate words in the set of candidate words in general. In addition, it should be noted that matching a term may be defined as "including the term in the candidate.
As an option: candidate word filtering module 30 may directly traverse the candidate word list with the shortest search word (e.g., a character), take the filtered candidate words as a candidate word set, and store the filtered candidate words in memory 31. At this time, the generated candidate word set (as long as it is a non-empty set) certainly includes the correct candidate word required by the user, and the candidate word screening module 30 may output the corresponding candidate word stored in the memory 31 to the pull-down menu generating module 40 for use in subsequently generating the user option tag. Although this approach can ensure that the correct candidate word is ultimately presented to the user in the selected drop-down box, it has the following disadvantages: the number of candidate words in the candidate word set is also relatively largest, i.e. the candidate words present in the drop-down box are relatively largest.
As a preferred option, in order to reduce the number of candidate words in the drop-down box for the user to select as much as possible, in other words, in order to provide the candidate words for the user to select as accurately as possible, the translated text search word extraction module 20 first extracts a shortest search word (for example, a character), the candidate word screening module 30 traverses the candidate word list using the shortest search word, if the generated candidate word set is a non-empty set, the translated text search word extraction module 20 adds a character position to the start position, the candidate word screening module 30 screens out candidate words matching the new search word from the candidate word list based on the new search word, replaces the candidate words in the candidate word set with the screened candidate words matching the new search word, and stores the candidate words matching the new search word in the memory 31; this operation is repeated until the candidate word set is an empty set, at which time the candidate word filtering module 30 outputs the last filtered candidate word stored in the memory 31 (i.e., the last filtered candidate word before the candidate word set is the empty set) to the pull-down menu generating module 40.
The pull-down menu generating module 40 receives the candidate words output from the candidate word screening module 30, and generates pull-down box option labels based on the received candidate words, so that the user can directly select correct contents from the option labels and input the correct contents into the translation.
FIG. 2 schematically illustrates a method for computer-aided correction of a machine-translated translation, according to an embodiment of the present invention. Referring to fig. 2, the computer-aided correction method for machine translation includes: step S10, extracting specific regular words in the original text by using a regular expression to serve as a candidate word list; step S20, determining a translation correction search term according to the position of the translation cursor; step S30, selecting candidate words matched with the search words from the candidate word list according to the search words, taking the selected candidate words as a candidate word set and storing the selected candidate words in a memory, wherein the candidate words in the memory are selectively output; and step S40, generating option labels according to the output candidate words, so that the user can directly select correct contents from the option labels and input the correct contents into the translated text.
The screening process of the candidate words output to the pull-down menu generating module 40 according to a preferred embodiment of the present invention is described in detail below with reference to fig. 3.
In step S21, the translated term extraction module 20 sets the position of the cursor as the end position and sets the position one character before the position of the cursor as the start position. In step S22, the translated text search term extraction module 20 extracts characters between the start position and the end position as a search term. In step S31, the candidate word filtering module 30 traverses the candidate word list using the search word extracted in step S22, filters out candidate words matching the search word, generates a candidate word set, and stores the filtered candidate words in the memory 31. In step S32, the candidate word screening module 30 determines whether the generated candidate word set is an empty set. If the candidate word set is not an empty set, the translated sentence search term extraction module 20 adds a character position to the start position in advance in step S23, and then the translated sentence search term extraction module 20 extracts a character between the new start position and the new end position as a search term in step S22. Next, in step S31, the candidate word filtering module 30 re-filters the candidate word matching the new search word from the candidate word list based on the new search word, replaces the original candidate word in the candidate word set with the filtered candidate word matching the new search word, and stores the candidate word matching the new search word in the memory 31. Continuing next at step S32, the candidate word filtering module 30 determines whether the generated set of candidate words is an empty set. As can be seen from fig. 3, this loop operation is repeated until the candidate word set is an empty set, and then to step S33, the candidate word that is filtered last time and stored in the memory 31 is output to the pull-down menu generating module 40 for use in generating the option label.
It is to be noted that although in the above exemplary embodiment, the search word is iterated in a loop starting with one character, those skilled in the art can understand that under the teaching of the present invention: the search term can also start iteration from two characters, three characters or more characters, and only the judgment mode of the iteration result and the subsequent corresponding processing flow need to be correspondingly adjusted, so that detailed description is omitted. In addition, although the candidate word filtering module is described as including a memory for convenience of description in the exemplary embodiment of the present invention, the memory may obviously be disposed outside the candidate word filtering module, which should not be construed as a limitation to the present invention.
Next, in order to facilitate a clearer understanding of a preferred embodiment of the present invention, a description will be given, with reference to fig. 4 and 5, by taking an example in which the content "background (x, 0.0067x2-0.6034x +79.729, -0.0067x2-0.3966x + 20.271)" in the text of the translated document japanese is not correctly presented in the text of chinese, and thus, a modification is required. The user has deleted the wrong content in the translation, the content presented in the current translation is the "coordinate (x,", the user is inputting the correct content according to the original text, and the assumption is just input to the "coordinate (x, 0.") of the content.
The original candidate word extraction module 10 may extract numbers using the regular expression "/\ d +/g" ("\ d" represents a number, "+" represents more than 1, "g" represents finding all), and the extraction results are "0, 0067, 2, 0, 6034, 79, 729, 0, 0067, 2, 0, 3966, 20, 271". The original candidate word extraction module 10 may extract dates and numbers by using the regular expression "/\ d + _ d +/g" ("\ d" represents a number, "+" represents more than 1, "" represents an arbitrary character, "g" represents finding all), and the extraction results are "0.0067 x2, 0.6034, 79.729, 0.0067x2, 0.3966, 20.271". The extracted results are merged and then subjected to a deduplication (repeated data removal) operation, and the obtained results are used as a candidate word list "0, 0067, 2, 6034, 79, 729, 3966, 20, 271, 0.0067x2, 0.6034, 79.729, 0.3966, 20.271".
The translated text search term extraction module 20 reads the position of a cursor, the cursor is positioned after the number "0" just input by the user, the position is taken as an end position, the position of a character before the position is taken as a start position, and the character between the start position and the end position is taken as a search term.
Candidate word filtering module 30 filters candidate words that match the term "through the candidate word list" 0, 0067, 2, 6034, 79, 729, 3966, 20, 271, 0.0067x2, 0.6034, 79.729, 0.3966, 20.271 "as" 0.0067x2 "," 0.6034 ", and" 0.3966 ". These several screened candidate words constitute the candidate word set "0.0067 x2, 0.6034, 0.3966" and are stored in the memory 31.
At this time, since the candidate word set is a non-empty set (i.e. there are matching candidate words), the candidate word filtering module 30 does not directly output the candidate words "0.0067 x2, 0.6034, 0.3966" in the memory 31 to the pull-down menu generating module 40. Instead, the translation search term extraction module 20 adds a character position to the start position, and extracts a character "0" between the new start position and the new end position as a search term. Next, the candidate word screening module 30 performs re-screening from the candidate word list based on the new search word "0." no matching candidate word exists, that is, the matching candidate word set is an empty set, and then outputs the last screening result "0.0067 x2, 0.6034, 0.3966" stored in the memory 31 to the pull-down menu generating module 40, and generates an option tag for the user to select, as shown in fig. 4.
At this time, the user can directly select the correct candidate word "0.0067 x 2" in the drop-down box by using the up and down keys on the mouse or keyboard, and then input the word into the translation, as shown in fig. 5.
The computer-aided correction system for machine translation according to the present invention may be implemented as specific hardware on a device or as software or firmware installed on a device, etc.
In the embodiments provided in the present invention, it should be understood that the disclosed method and system can be implemented in other ways. The above-described embodiments are merely illustrative, and for example, the division of the modules into only one logical functional division may be implemented in practice in another way, and for example, multiple units or components may be combined or integrated into another device, or some features may be omitted, or not implemented.
In addition, each functional unit in the embodiments provided in the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Finally, it should be noted that: the above-mentioned embodiments are merely exemplary embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; and that such modifications, changes, or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention. Are intended to be covered by the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope defined by the claims.
Claims (10)
1. A computer-aided correction system (1) for machine-translated translations, comprising:
the original text candidate word extraction module (10) is used for extracting specific regular words in the original text by using a regular expression to serve as a candidate word list;
the translation search term extraction module (20) is used for determining a translation correction search term according to the position of a translation cursor;
a candidate word screening module (30) comprising a memory (31) and used for screening candidate words matched with the search words from a candidate word list provided by the original text candidate word extraction module (10) according to the search words provided by the translated text search word extraction module (20), taking the screened candidate words as a candidate word set and storing the screened candidate words into the memory (31), wherein the candidate words in the memory (31) are selectively output; and
and the pull-down menu generating module (40) receives the candidate words output from the candidate word screening module (30) and generates option labels based on the received candidate words, so that the user can directly select correct contents from the option labels and input the correct contents into the translation.
2. The computer-aided correction system (1) of machine-translated translations according to claim 1, wherein said specific rule words comprise at least one of dates, numbers, letters, and symbols.
3. The computer-aided correction system (1) of machine-translated translation according to claim 1, characterized in that the original candidate word extraction module (10) performs a deduplication operation after extracting a specific regular word, so that there are no candidate words that appear repeatedly in the candidate word list.
4. The computer-aided correction system (1) for machine-translated translation according to claim 1, wherein the translation search term extraction module (20) takes the position of the cursor as the ending position, the position spaced by one character before the cursor as the starting position, and the characters between the starting position and the ending position as the search term.
5. The computer-aided correction system (1) for machine-translated translation according to claim 4, wherein if the candidate word set is a non-empty set, the translation search word extraction module (20) adds a character position to the start position, the candidate word screening module (30) screens out a candidate word matching the new search word from the candidate word list based on the new search word, replaces the candidate word in the candidate word set with the screened candidate word matching the new search word, and stores the candidate word matching the new search word in the memory (31); this operation is repeated until the candidate word set is an empty set, at which time the candidate word filtering module (30) outputs the last filtered candidate word stored in the memory (31) to the pull-down menu generating module (40).
6. A method for computer-aided correction of a machine-translated translation, comprising the steps of:
extracting specific regular words in the original text by using a regular expression to serve as a candidate word list;
determining a corrected search term of the translated text according to the position of a cursor in the translated text;
screening candidate words matched with the search words from the candidate word list according to the search words, taking the screened candidate words as a candidate word set and storing the screened candidate words in a memory, wherein the candidate words in the memory are selectively output; and
and generating an option label according to the output candidate word, so that the user can directly select correct content from the option label and input the correct content into the translation.
7. The method of claim 6, wherein the specific rule word comprises at least one of a date, a number, a letter, and a symbol.
8. The method of claim 6, wherein the specific regular word is extracted and then subjected to a deduplication operation, such that there are no candidate words in the candidate word list that occur repeatedly.
9. The method of claim 6, wherein the position of the cursor is used as an end position, the position which is one character before the position of the cursor is used as a start position, and characters between the start position and the end position are used as search words.
10. The method of claim 9, wherein if the set of candidate words is a non-empty set, adding a character position to the starting position, screening candidate words matching the new search word from the candidate word list based on the new search word, replacing candidate words in the candidate word set with the screened candidate words matching the new search word, and storing the candidate words matching the new search word in the memory; and repeating the operation until the candidate word set is an empty set, and outputting the candidate word which is stored in the memory and screened last time for generating the option label.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210184626.7A CN114528858A (en) | 2022-02-28 | 2022-02-28 | Computer-aided correction system and method for machine translation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210184626.7A CN114528858A (en) | 2022-02-28 | 2022-02-28 | Computer-aided correction system and method for machine translation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114528858A true CN114528858A (en) | 2022-05-24 |
Family
ID=81623966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210184626.7A Pending CN114528858A (en) | 2022-02-28 | 2022-02-28 | Computer-aided correction system and method for machine translation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114528858A (en) |
-
2022
- 2022-02-28 CN CN202210184626.7A patent/CN114528858A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4311365B2 (en) | Document processing apparatus and program | |
| JP5372110B2 (en) | Information output device, information output method, and computer program | |
| WO2003065245A1 (en) | Translating method, translated sentence outputting method, recording medium, program, and computer device | |
| JP2011150466A (en) | Device, program and method for recognizing character string | |
| CN103049458A (en) | Method and system for revising user word bank | |
| US10896292B1 (en) | OCR error correction | |
| JP2011141749A (en) | Apparatus and method for generating document image and computer program | |
| JP7040227B2 (en) | Information processing programs, information processing methods, and information processing equipment | |
| Sturgeon | Large-scale Optical Character Recognition of pre-modern Chinese texts | |
| JP5353325B2 (en) | Document data generation apparatus and document data generation method | |
| CN114528858A (en) | Computer-aided correction system and method for machine translation | |
| JP7027757B2 (en) | Information processing equipment and information processing programs | |
| CN111310457B (en) | Word mismatching recognition method and device, electronic equipment and storage medium | |
| JP7268389B2 (en) | Information processing device and program | |
| US20110033114A1 (en) | Image processing apparatus and computer readable medium | |
| JPH0696288A (en) | Character recognizing device and machine translation device | |
| JP3398729B2 (en) | Automatic keyword extraction device and automatic keyword extraction method | |
| JP3103179B2 (en) | Document creation device and document creation method | |
| JPH0883280A (en) | Document processor | |
| JPH09153034A (en) | Document preparing device and method therefor | |
| JP5628485B2 (en) | Translation support system and method and program thereof | |
| JP7342518B2 (en) | Image processing device and image processing program | |
| JP2004078531A (en) | Character recognition device, character recognition method, and program for executing method in computer | |
| JP2013068983A (en) | Information processing device and information processing program | |
| JP2002245470A (en) | Language specifying device, translating device, and language specifying method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20220524 |
|
| WD01 | Invention patent application deemed withdrawn after publication |