CN108228561A - A kind of method that processing text flow is configured - Google Patents
A kind of method that processing text flow is configured Download PDFInfo
- Publication number
- CN108228561A CN108228561A CN201711293595.4A CN201711293595A CN108228561A CN 108228561 A CN108228561 A CN 108228561A CN 201711293595 A CN201711293595 A CN 201711293595A CN 108228561 A CN108228561 A CN 108228561A
- Authority
- CN
- China
- Prior art keywords
- text
- analyzer
- filter
- flow
- text flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- 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
- 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
- G06F16/337—Profile generation, learning or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
- G06F16/90344—Query processing by using string matching techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/253—Grammatical analysis; Style critique
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to tool software development technique field, during the too development for being applied to processing text flow.Main feature of the present invention is to provide a kind of universal method, a filter pretreatment unit, filter unit and filter deployment definition are provided to analyzer, eliminate in input text flow can not analysing content, and the action message that analyzer is supported can be inserted into output text flow, so as to the operating status of control analyzer, the ability of developer's control analyzer state is enhanced.The problem of present invention cannot identify certain character strings in text flow for analyzer, it proposes a kind of according to pattern match text flow and to carry out respective treated method, increase filter preprocessor and filter before analyzer, input text flow is analyzed and processed, expand analyzer analyzes range of text.
Description
Technical field
The invention belongs to tool software development technique fields, are applied to the process of the too development of processing text flow
In.In the case that invention describes one kind, partial character string cannot be analyzed device analysis in text flow is inputted, how to handle
The method that input text flow can be analyzed device analysis using the method that is unable to identification string is deleted or modified, is eliminated
Input text flow in can not analysing content.
Background technology
In computer science, widely apply and be required for handling various types of text flows, text flow in system
Content and form be also not quite similar, it is common processing text flow tool have text flow filter, pattern matcher etc..This
A little tools can be used for changing the text flow for inputing to equipment, so as to fulfill the text stream attribute of needs.Such as text flow filter
Available for handling the input text flow of an analyzer, the i.e. filter as an analyzer.It will be filtered herein with text flow
It device and is illustrated by taking the analyzer as input of the output after text flow filter process as an example, including but not limited to text flow
Filter and analyzer.
Text analyzer can be a program that morphology, grammer or semantic information are extracted from source file.Typical text
Instant analyzer includes a front end, a typically filter, for parsing the source text stream of input.Filter compares input source
Code flow and predefined pattern combine, after some pattern match, it will the action for the pattern definition is performed, for example,
Matched source text can be transmitted to syntactic analysis subsystem as a symbol.Filter can be independent executable journey
Sequence can also be comprised in analyzer, such as compiler front-end, text editor, the text filter of supporting assist edit
Deng.
Text flow filter can be used for the input of modification analyzer upstream, so that analyzer can be handled comprising analysis
Device cannot identify the source text of content.For example, analyzer front end can realize that certain rule is replaced certain analyzers not with space
The character string being understood that.
Text analyzer front-end functionality before is limited, such as the blank of equal length or space can only be used to replace input text
This, but other texts cannot be replaced with.Filter before does not support write-in string attribute information, such as before replacement action
Original text block length information.As a result it is exactly, it is necessary to which definition ignores rule to ignore certain source texts completely.A large amount of source document
This perhaps cannot be analyzed device identification, lead to analysis failure or lose information.There is no length information, can only with be replaced text
The text of equal length goes to replace the content of input text.In order to replace random length, it is necessary to which character string is in character stream
Location information.In addition, text filter before cannot provide the information of analyzer support action, do not support canonical table yet
Up to formula as pattern, and together with the front end that uses of analyzer before and specific analyzer code close-coupled, it is impossible to
Realize the versatility of analyzer front end.
Invention content
Present invention solves the technical problem that it is：It the problem of cannot identifying certain character strings in text flow for analyzer, carries
Go out it is a kind of according to pattern match text flow and carry out respective treated method, increase before analyzer filter preprocessor and
Filter analyzes and processes input text flow, and expand analyzer analyzes range of text.
The text of filtering includes the action of analyzer support and the location information of input text.This enable filter into
It composes a piece of writing this replacement operation, and increases the specific action that analyzer is supported in text flow after filtration.Specific reality in the present invention
Existing method includes defining filter deployment, text flow is inputted according to configuration modification and ultimately generating the action supported comprising analyzer
Filtering text.
It defines a filter deployment to include defining mode and define respective association definition between these patterns, also
There is each pattern to correspond to executable operation.It defines a filter deployment and includes tupe and its respective relationship, so as to be formed
The filter data structure of input text and pattern can be compared, and in input text flow and the base of one of pattern match
The execution of related executable operation is triggered on plinth.It operates the processing procedure performed and comparison procedure can be complete in a procedure
Into can also complete in a separate process.
One filter deployment of definition, which is also possible that, is respectively associated the independent beginning shape of at least one of beginning state
State sets current beginning state to each pattern, and it is one in a large amount of beginning states that this, which currently starts state,.Operation performs
Processing procedure may include handling at least one independent beginning state, it is certain for input text flow so as to complete
Operation.Whether comparison procedure is possibly used for comparing with each pattern in the relevant a large amount of patterns of each beginning state with currently opening
Beginning state is equal.In addition, each starting state, there are one corresponding contexts.
Method in the present invention includes the location information of the character string in analysis input text flow, which can be used for
Determine the difference replaced text and be replaced between text.By the way that source text character string can be replaced with using the location information
The character string of random length.
Method in the present invention, which further includes, compares input text flow and pattern, and in input text flow and some pattern match
On the basis of, the operation of matched text in triggering modification input text flow.Modification input text flow include the use of one it is static
It replaces text and replaces matched text in input text flow, it is also possible to include the use of dynamic replacement text and replace in input text flow
At least one character.Dynamic replacement text can be dynamically determined, for example, can be computed by executable action or
Person is determined by the part in matched text.
Method in the present invention further includes to be defined according to filter deployment and was inserted into the action message that analyzer is supported
In filter output text flow.Analyzer can identify the action message in character stream after receiving filtering text flow and perform corresponding
Action.
The present invention also provides the tool of a filtering text flow.The tool includes filter deployment data structure, one
It is a to input that the filter pretreatment unit of text flow and generation position information, one according to mistake according to filter deployment defined analysis
The filter unit of the location information modification input text flow of filter configuration definition and the output of filter pretreatment unit.Filter
The text flow of pretreatment unit output includes the location information of character string in input text flow.The output text flow of filter unit
The action message that analyzer is supported may be included.
The present invention also provides the medium of a computer-readable storage computer executable program, can to store computer
The method of the filtering input text of execution.This method includes definition and receives a filter deployment, is changed according to filter deployment
Text flow is inputted, so as to generate the text flow after filtering.Text flow after filtering includes the text flow changed by filter, also
The action that analyzer is supported may be included.
The present invention also provides an analysis tool, including a filter deployment data structure, a filter pretreatment
Unit, the unit are used to input the character string location information in text flow according to filter deployment data analysis, and generate and include
The character stream of location information；One location information exported using filter deployment data and filter pretreatment unit is changed defeated
Enter text flow so as to generate the filter unit of text after filtering, it is also possible to be inserted into output text flow analyzer support it is dynamic
Make information.Text after filtering may include the action that analyzer is supported.The analytical equipment also comprising analyzer module, is configured
For handling the text flow after filtering so as to perform corresponding action.
Technical solution
The present invention provides a kind of method for filtering input text flow.
Fig. 1 shows the technical solutions of filtering input text in the present invention.Text flow (101) is inputted by filter (103)
Generation output text flow (104) after processing, in processing procedure, filter (103) can define (102) according to filter deployment and repair
Change input text flow, and be inserted into analyzer (105) energy identification maneuver information wherein；Analyzer (105) receives output text flow
(104) it is analyzed, and identifies that action message therein carries out execution processing.
The first step：
It is analyzer (105) urtext stream to be processed to input text flow (101), and input text flow (101) is to divide
The source code stream or other types of text flow of analysis, wherein comprising the text-string that analyzer (105) cannot identify, therefore need
It to be handled before analyzer (105) is transmitted to, certain character strings in input text flow (101) is deleted or modified；It is original defeated
Enter text flow (101) and output text flow (104) is generated after filter process, output text flow (104) is analyzer (105)
The text flow of recognizable analysis carries out morphological analysis, syntactic analysis, semantic analysis operation according to analyzer (105) itself function
And other operations.
Second step：
Filter (103) is including two parts：Filter pretreatment unit and filter unit；Filter (103) is according to mistake
Filter configuration definition (102) modification input text flow (101), so as to generate the output text flow (104) after filtering, after filtering
Output text flow (104) is the text flow of action message that can be identified comprising analyzer after filtering；
Filter (103) filtering input text flow (101) and the output text flow (104) that obtains that treated；In filter
(103) action message that the location information of character string and analyzer are supported in input text flow (101) can be generated in processing procedure；
The text-string that the corresponding location information of character string in text flow (101) is used to flexibly replace random length is inputted, wherein one
A text-string and a number for representing character string location information in input text flow (101) correspond；Analyzer branch
The subsequently analyzed device of the action message held identifies and performs that the action that analyzer is supported can be configured for control analyzer operation
State, it is to be configured and exported according to actual needs to export analyzer is supported in text flow (104) amount of action and position
In output text flow (104) after to filtering；Output text flow (104) is finally passed to analyzer (105).
Third walks：
To form filter (103), need to generate filter deployment definition (102)；Filter deployment defines (102) and includes
For searching for the pattern used, one or more beginning states relevant with each pattern in input text flow (101), defeated
Enter the action message of the operation performed after the success of text flow (101) pattern match and analyzer support；It can perform behaviour after matching
Make to include but not limited to replace one or more of input text flow (101) character with other texts, be replaced with null character string
The text in text flow (101) is inputted, copies matched text to output text flow (104)；Analyzer support action according to
Function that analyzer is supported in itself is different, the processing by filter deployment and filter of the position in text flow export in itself
Journey determines, and the quantity of analyzer support action is also determined in itself by filter deployment and filter；
Filter (103) defines (102) processing input text flow (101) using filter deployment, and search is matched with filter
The text of the pattern match defined in file (102) is put, when finding a matched text, phase is performed by filter (103)
The operation answered；Location information is determined by executable operation, be can perform operation and is realized in input text flow (101) and pattern
The text of timing is replaced；One executable operation can generate one or more marks；One mark is comprising matched text and replaces
The difference of length between exchange of notes sheet, i.e., one executable operation is able to access that and replaces text and matched text, to determine this
The difference of length；
The text of replacement supports dynamic to determine, dynamic determining method includes but not limited to：It is matched text to replace text
Small letter turns the content after capitalization；It is that matched text capitalization turns the content after small letter to replace text；It is by regular expressions to replace text
A part for the matched text of formula expression formula；It is the modified text of matched text either substring to replace text；It replaces
The exchange of notes is originally the text that matched text is calculated in a manner.
4th step：
Output text flow (104) is transmitted to analyzer (105) to analyze, analyzer is included in output text flow (104)
(105) action message that can be identified and perform.
Description of the drawings
Fig. 1 shows the flow charts of filtering input text flow method.
Fig. 2 represents filtering input text flow and output text flow is transmitted to the schematic diagram of analyzer.
Specific embodiment
Fig. 2 represents filtering input text flow in the present invention and exports to the schematic diagram of analyzer.Filter (103) in Fig. 1
It can be divided into filter preprocessor (203) and filter (205) comprising simultaneously quilt here.
It is analyzer urtext stream to be processed to input text flow (201), it is assumed that the sequence string A comprising input text,
The contents such as the B, contents such as C, wherein A, B, C may not be analyzed device identification, it is therefore desirable to text flow be handled, delete or replace
Change the content that analyzer cannot identify.Input text-string A, B, C etc. may each of comprising one or more text words
Symbol.Input text flow (201) is the input of filter pretreatment (203).Filter pretreatment (203) is determined according to filter deployment
Adopted (202) analysis input text flow (201), and export the text flow (204) after analysis.The text flow of filter pretreatment output
(204) comprising location information Pa, Pb, Pc corresponding to character string A, B, C etc. etc., quantity and the character string number phase of location information
Same and one-to-one correspondence.Filter (205) receives the text flow (204) for including location information, and matching is searched in text flow (204)
The text flow of predefined pattern, if it does, then performing an independent predefined operation for matched text.Filter
(205) (202) can also be defined according to filter deployment and generate the action message that analyzer (207) is supported, and by action message
It is inserted into the text flow (206) after filtering.
Filter deployment defines the rule that (202) include one or more filtering input text flow (201).These rules
Comprising Text Mode, including but not limited to regular expression, text or other character string sequences, symbol, these rules by with
In search input text flow (201) and the text flow (204) of filter pretreatment output.In addition, filter deployment defines (202)
In rule include the operation for input text flow (201) to be performed when the success of pattern match, further include analysis
Operation information Action1, Action2 in the text flow (206) that device (207) is supported.These operations include but not limited to defeated
Enter to be inserted into text flow (201) character string location information, matched text in inlet flow is replaced with another text, uses blank
The action message for controlling its own processing procedure that character replaces matched text, analyzer (207) is supported.
Filtering text (206) is one and includes one or more treated text-string, one or more analyzers
The sequence flows of the action message of support.The character string A ' that filters in text (206) that treated, B ', C ' quantity and input text
The quantity for flowing input text-string A, B, C in (201) etc. is consistent.Rule in (202) is defined according to filter deployment, such as
Former input character string is then changed in the success of fruit pattern match, does not otherwise change former input character string, i.e., single after filter process
Character string is identical with original input character string.The action that the analyzer that the text flow (206) of filter (205) output includes is supported
Action1, Action2 are to be defined defined in (202) in filter deployment, and be that analyzer can be identified and is able to carry out
Action, the content and quantity of action message Action1, Action2 define (202) by filter deployment and determine, quantity with it is defeated
The quantity for entering text-string A, B, C does not need to be identical.
The text-string A ' of each filter pretreatment output, B ', C ' are including but not limited to one kind in following：Input
The corresponding copy of character string A, B, C in text flow (201)；As replacing, character string A, B, C in input text flow (201) are corresponding
Substitute the text of random length being inserted by text flow filter (205)；As replace character string A in input text flow (201),
B, C substitutes the null character string being inserted by text flow filter (205) accordingly.
Pa, Pb, Pc provide location information.Each Pa, Pb, Pc each export text flow (206) comprising a digital representation
Difference between middle character string A ', B ', C ' number of characters and corresponding input text-string A, B, C.In other realizations of the present invention
In mode, support to represent other types of information using Pa, Pb, Pc.It acts Action1, Action2 and analyzer (205) is provided
The action of support will perform corresponding action, the phase after analyzer (205) receives action Action1, Action2
The operating status available for control analyzer should be acted.
The method support of the configuration processing text flow of the present invention is applied to the processing text flow before analysis text flow
Any occasion, that is, support to flow into urtext the modification of line character string, delete and downstream analyzer transmits control information,
During being including but not limited to used in before analyzer, compiler processes text flow.
Claims (1)
- A kind of 1. method that processing text flow is configured, it is characterised in that：Text flow (101) is inputted to handle by filter (103) Generation output text flow (104) afterwards, in processing procedure, it is defeated that filter (103) can define (102) modification according to filter deployment Enter text flow, and be inserted into analyzer (105) energy identification maneuver information wherein；Analyzer (105) receives output text flow (104) It is analyzed, and identifies that action message therein carries out execution processing；The first step：It is analyzer (105) urtext stream to be processed to input text flow (101), and input text flow (101) will be analyzed Source code stream or other types of text flow, wherein the text-string that cannot be identified comprising analyzer (105), it is therefore desirable to It is handled before being transmitted to analyzer (105), certain character strings in input text flow (101) is deleted or modified；It is originally inputted text This stream (101) generates output text flow (104) after filter process, and output text flow (104) is that analyzer (105) can be known The text flow do not analyzed, according to analyzer (105) itself function carry out morphological analysis, syntactic analysis, semantic analysis operation and its It is operated；Second step：Filter (103) is including two parts：Filter pretreatment unit and filter unit；Filter (103) is according to filter Configuration definition (102) modification input text flow (101), so as to generate the output text flow (104) after filtering, the output after filtering Text flow (104) is the text flow of action message that can be identified comprising analyzer after filtering；Filter (103) filtering input text flow (101) and the output text flow (104) that obtains that treated；In filter (103) The action message that the location information of character string and analyzer are supported in input text flow (101) can be generated in processing procedure；Input The corresponding location information of character string is one of literary for flexibly replacing the text-string of random length in text flow (101) This character string and a number for representing character string location information in input text flow (101) correspond；What analyzer was supported The subsequently analyzed device of action message identifies and performs that the action that analyzer is supported can be configured for control analyzer operation shape State, it is to be configured and be output to according to actual needs to export analyzer is supported in text flow (104) amount of action and position In output text flow (104) after filtering；Output text flow (104) is finally passed to analyzer (105)；Third walks：To form filter (103), need to generate filter deployment definition (102)；Filter deployment define (102) comprising for The pattern used, the one or more and relevant beginning state of each pattern are searched in input text flow (101), in input text The action message that the operation and analyzer performed after the success of this stream (101) pattern match is supported；It can perform operation packet after matching It includes but is not limited to replace one or more of input text flow (101) character with other texts, replaced and inputted with null character string Text in text flow (101) copies matched text to output text flow (104)；The action that analyzer is supported is according to analysis The function that device is supported in itself is different, and the position in text flow is exported is determined by the processing procedure of filter deployment and filter in itself It is fixed, and the quantity of analyzer support action is also determined in itself by filter deployment and filter；Filter (103) defines (102) processing input text flow (101), search and filter deployment text using filter deployment The text of pattern match defined in part (102) when finding a matched text, is performed corresponding by filter (103) Operation；Location information is determined by executable operation, be can perform operation and is realized when inputting text flow (101) with pattern match Text replace；One executable operation can generate one or more marks；One mark is comprising matched text and replaces text The difference of length between this, i.e., one executable operation is able to access that and replaces text and matched text, to determine this length Difference；The text of replacement supports dynamic to determine, dynamic determining method includes but not limited to：It is matched text small letter to replace text Turn the content after capitalization；It is that matched text capitalization turns the content after small letter to replace text；It is by regular expressions formula to replace text A part for the matched text of expression formula；It is the modified text of matched text either substring to replace text；Replace text Originally it is the text that matched text is calculated in a manner；4th step：Output text flow (104) is transmitted to analyzer (105) to analyze, analyzer (105) is included in output text flow (104) The action message that can be identified and perform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711293595.4A CN108228561A (en) | 2017-12-07 | 2017-12-07 | A kind of method that processing text flow is configured |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711293595.4A CN108228561A (en) | 2017-12-07 | 2017-12-07 | A kind of method that processing text flow is configured |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108228561A true CN108228561A (en) | 2018-06-29 |
Family
ID=62654030
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711293595.4A Pending CN108228561A (en) | 2017-12-07 | 2017-12-07 | A kind of method that processing text flow is configured |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108228561A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030037027A1 (en) * | 2001-08-14 | 2003-02-20 | Christian Linhart | Text stream filter |
| CN105917327A (en) * | 2013-12-11 | 2016-08-31 | 触摸式有限公司 | System and method for inputting text into electronic devices |
| CN106168909A (en) * | 2016-06-30 | 2016-11-30 | 北京奇虎科技有限公司 | A kind for the treatment of method and apparatus of daily record |
| CN107391691A (en) * | 2017-07-26 | 2017-11-24 | 成都科来软件有限公司 | The filter method of data in a kind of network analysis |
-
2017
- 2017-12-07 CN CN201711293595.4A patent/CN108228561A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030037027A1 (en) * | 2001-08-14 | 2003-02-20 | Christian Linhart | Text stream filter |
| CN105917327A (en) * | 2013-12-11 | 2016-08-31 | 触摸式有限公司 | System and method for inputting text into electronic devices |
| CN106168909A (en) * | 2016-06-30 | 2016-11-30 | 北京奇虎科技有限公司 | A kind for the treatment of method and apparatus of daily record |
| CN107391691A (en) * | 2017-07-26 | 2017-11-24 | 成都科来软件有限公司 | The filter method of data in a kind of network analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hulden | Foma: a finite-state compiler and library | |
| Chang et al. | TokensRegex: Defining cascaded regular expressions over tokens | |
| Grefenstette | Tokenization | |
| CN106843840B (en) | Source code version evolution annotation multiplexing method based on similarity analysis | |
| US7779049B1 (en) | Source level optimization of regular expressions | |
| JP4007562B2 (en) | Programming support method and apparatus | |
| CN105389195B (en) | A kind of static analysis tools improved method replaced based on code with regular expression | |
| CN103049458B (en) | A kind of method and system revising user thesaurus | |
| CN101398758A (en) | Detection method of code copy | |
| US20240201984A1 (en) | Deep learning-based java program internal annotation generation method and syste | |
| CN107992476B (en) | Corpus generation method and system for sentence-level biological relation network extraction | |
| Thomas et al. | Not all links are equal: Exploiting dependency types for the extraction of protein-protein interactions from text | |
| CN108241658A (en) | A kind of logging mode finds method and system | |
| CN110221836A (en) | A kind of lexical analysis tool | |
| Hall et al. | Problem oriented software engineering: A design-theoretic framework for software engineering | |
| US20060143555A1 (en) | Apparatus and method for extracting information from a formatted document | |
| EP3422181B1 (en) | Method and device for generating code assistance information | |
| CN108228561A (en) | A kind of method that processing text flow is configured | |
| CN102982029B (en) | A kind of search need recognition methods and device | |
| Reger | Better Proof Output for Vampire. | |
| Aksoy et al. | MATAWS: A multimodal approach for automatic WS semantic annotation | |
| Sochor et al. | Fuzzing-Based Grammar Inference | |
| Hardie | Automated part-of-speech analysis of Urdu: conceptual and technical issues | |
| Jiao et al. | Research on Procedure Pattern Recognition Method Based on Conditional Constraints | |
| Carvalho et al. | Unfuzzying fuzzy parsing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180629 |
|
| RJ01 | Rejection of invention patent application after publication |