CN110427611A - Text handling method, device, equipment and storage medium - Google Patents
Text handling method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN110427611A CN110427611A CN201910560145.XA CN201910560145A CN110427611A CN 110427611 A CN110427611 A CN 110427611A CN 201910560145 A CN201910560145 A CN 201910560145A CN 110427611 A CN110427611 A CN 110427611A
- Authority
- CN
- China
- Prior art keywords
- text
- name entity
- recognition result
- semantics recognition
- entity
- 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
- 238000000034 method Methods 0.000 title claims abstract description 72
- 239000012634 fragment Substances 0.000 claims description 36
- 238000004590 computer program Methods 0.000 claims description 26
- 230000011218 segmentation Effects 0.000 claims description 12
- 239000000284 extract Substances 0.000 description 10
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 235000013399 edible fruits Nutrition 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application discloses a kind of text handling method, device, equipment and storage mediums, belong to text-processing field.The described method includes: carrying out semantics recognition to text to be processed, at least one semantics recognition result is obtained；Entity recognition is named to the text, obtains at least one name entity；By at least one semantics recognition result, at least one name entity is associated with this, obtains the association results of the text, the association results of the text are used to characterize each semantics recognition result and name the corresponding relationship between entity.Technical solution provided by the embodiments of the present application solves the problems, such as that the validity for the valuable information being drawn into from text is poor.
Description
Technical field
This application involves text-processing fields, are situated between more particularly to a kind of text handling method, device, equipment and storage
Matter.
Background technique
It under normal conditions, may include much valuable information in text, these valuable information can reflect use
The various intentions at family, alternatively, these valuable information have certain statistical analysis meaning.In view of this, in practical application
It generally requires and extracts these valuable information from text, this is an importance of text-processing.
In the related technology, name entity recognition techniques be can use extracts from text and name entity, and will extract
Name entity is determined as the valuable information extracted from text.Wherein, so-called name entity refers to name, mechanism
Name, place name, number, date, currency and other entities with entitled mark, for example, the relevant technologies can be to " I wants to buy
This text of the house of every square meter 10,003 " is named Entity recognition, to extract name entity " 10,000 from the text
Three ", and the name entity " 10,003 " is determined as the valuable information extracted from text.
However, in many cases, the meaning of the name entity, example can not be accurately reflected according only to name entity itself
Such as, in citing above, the name entity " 10,003 " extracted from text is a number, originally according only to this number
Body can not accurately reflect its meaning, this causes the validity for the valuable information being drawn into from text poor.
Summary of the invention
Based on this, it is necessary to for the problem that the validity for the valuable information being drawn into from text is poor, provide
A kind of text handling method, device, equipment and storage medium.
In a first aspect, a kind of text handling method is provided, this method comprises:
Semantics recognition is carried out to text to be processed, obtains at least one semantics recognition result；
Entity recognition is named to the text, obtains at least one name entity；
By at least one semantics recognition result, at least one name entity is associated with this, obtains the association of the text
As a result, the association results of the text are used to characterize each semantics recognition result and name the corresponding relationship between entity.
Entity recognition is named to the text in one of the embodiments, before obtaining at least one name entity,
This method further include:
Semantic segmentation is carried out to the text according at least one semantics recognition result, obtains at least one text fragments,
Wherein, each text segment is corresponding with a semantics recognition result；
Accordingly, Entity recognition is named to the text, obtains at least one name entity, comprising:
Entity recognition is named to each text segment, obtains the corresponding name entity of each text segment.
By at least one semantics recognition result, at least one name entity is closed with this in one of the embodiments,
Connection, obtains the association results of the text, comprising:
The corresponding name entity of each text segment and the corresponding semantics recognition result of each text segment are carried out
Association, obtains the association results of each text segment；
According to the association results of each text segment, the association results of the text are obtained.
By at least one semantics recognition result, at least one name entity is closed with this in one of the embodiments,
Connection, obtains the association results of the text, comprising:
According to preset name entity rule, each semantics recognition result is obtained at least one name entity from this and is corresponded to
Name entity, wherein the name entity rule, which is used to indicate needed for the corresponding name entity of each semantics recognition result, to be met
Condition；
According to the corresponding name entity of each semantics recognition result, the association results of the text are generated.
The association results of the text include pair of each semantics recognition result and name entity in one of the embodiments,
It should be related to；By at least one semantics recognition result, at least one name entity is associated with this, obtains the association of the text
As a result after, this method further include:
Identical first corresponding relationship of semantics recognition result and the second corresponding relationship are extracted from the association results of the text,
Wherein, which includes the corresponding relationship of target semanteme recognition result and the first name entity, the second corresponding pass
System includes the corresponding relationship of the target semanteme recognition result and the second name entity；
First corresponding relationship and second corresponding relationship are merged, obtain merging corresponding relationship；Wherein, the merging
Corresponding relationship includes the corresponding relationship of the target semanteme recognition result and name entity sets, the name entity sets include this
One name entity and the second name entity.
By at least one semantics recognition result, at least one name entity is closed with this in one of the embodiments,
Connection, after obtaining the association results of the text, this method further include:
It is obtained from database of content items according to the association results of the text and the matched target of the association results of the text
Content item, the database of content items are stored with multiple content items；
The object content item is pushed to the corresponding user of the text.
This method in one of the embodiments, further include: obtain session content, include multiple wait locate in the session content
The text of reason.
Second aspect, provides a kind of text processing apparatus, which includes:
Semantics recognition module obtains at least one semantics recognition result for carrying out semantics recognition to text to be processed；
Entity recognition module is named, for being named Entity recognition to the text, obtains at least one name entity；
Relating module, for by least one semantics recognition result with this at least one name entity be associated, obtain
To the association results of the text, the association results of the text are used to characterize each semantics recognition result and name pair between entity
It should be related to.
The device further includes segmentation module in one of the embodiments,.The segmentation module is used for according to this at least one
Semantics recognition result carries out semantic segmentation to the text, obtains at least one text fragments, wherein each text segment and one
A semantics recognition result is corresponding；Accordingly, the name Entity recognition module, is specifically used for: carrying out to each text segment
Entity recognition is named, the corresponding name entity of each text segment is obtained.
The relating module in one of the embodiments, is specifically used for: by the corresponding name entity of each text segment
Semantics recognition result corresponding with each text segment is associated, and obtains the association results of each text segment；According to
The association results of each text segment, obtain the association results of the text.
The relating module in one of the embodiments, is specifically used for: according to preset name entity rule, from this to
The corresponding name entity of each semantics recognition result is obtained in a few name entity, wherein the name entity rule is for referring to
Show the condition met needed for the corresponding name entity of each semantics recognition result；According to the corresponding life of each semantics recognition result
Name entity, generates the association results of the text.
The association results of the text include pair of each semantics recognition result and name entity in one of the embodiments,
It should be related to；The device further includes extraction module and merging module.
The extraction module, for extracting identical first corresponding relationship of semantics recognition result from the association results of the text
With the second corresponding relationship, wherein first corresponding relationship includes that target semanteme recognition result and the first the corresponding of name entity are closed
System, second corresponding relationship include the corresponding relationship of the target semanteme recognition result and the second name entity；
The merging module obtains merging correspondence for merging first corresponding relationship and second corresponding relationship
Relationship；Wherein, which includes the corresponding relationship of the target semanteme recognition result and name entity sets, the name
Entity sets include the first name entity and the second name entity.
The device further includes that content item obtains module and pushing module in one of the embodiments,.
The content item obtains module, obtains from database of content items for the association results according to the text and the text
The matched object content item of association results, which is stored with multiple content items；
The pushing module, for the object content item to be pushed to the corresponding user of the text.
The device in one of the embodiments, further include:
Acquisition conversation module includes multiple texts to be processed in the session content for obtaining session content.
The third aspect, provides a kind of computer equipment, including memory and processor, which is stored with computer
Program, the computer program realize above-mentioned first aspect any text handling method when being executed by the processor.
Fourth aspect provides a kind of computer readable storage medium, is stored thereon with computer program, which is located
Reason device realizes above-mentioned first aspect any text handling method when executing.
Technical solution bring beneficial effect provided by the embodiments of the present application includes at least:
By carrying out semantics recognition to text to be processed, at least one semantics recognition is obtained as a result, and to be processed
Text is named Entity recognition, obtains at least one name entity, then, extremely with this by least one semantics recognition result
A few name entity is associated, and obtains the association results of the text to be processed, wherein the association results are every for characterizing
Corresponding relationship between a semantics recognition result and name entity, in this manner it is possible to using above-mentioned association results as to be processed
Text in the valuable information that extracts, since association results characterize between each semantics recognition result and name entity
Corresponding relationship, accordingly, it is possible to reflect the meaning of name entity using semantics recognition result, so, so that it may it improves from text
The validity for the valuable information being drawn into this.
Detailed description of the invention
Fig. 1 is a kind of schematic diagram of the implementation environment of text handling method provided by the embodiments of the present application；
Fig. 2 is the flow chart of another text handling method provided by the embodiments of the present application；
Fig. 3 is the flow chart of another text handling method provided by the embodiments of the present application；
Fig. 4 is the flow chart of another text handling method provided by the embodiments of the present application；
Fig. 5 is the flow chart of another text handling method provided by the embodiments of the present application；
Fig. 6 is a kind of block diagram of text processing apparatus provided by the embodiments of the present application；
Fig. 7 is the block diagram of another text processing apparatus provided by the embodiments of the present application；
Fig. 8 is a kind of block diagram of computer equipment provided by the embodiments of the present application.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with attached drawing to the application embodiment party
Formula is described in further detail.
Under normal conditions, it may include much valuable information in text, in practical application, need these are valuable
Information extracted from text, thus using these the valuable information extracted determine user be intended to, alternatively, into
Row statistical analysis.
In a kind of possible application scenarios, user can use instant messaging application and conversate, which includes
It in practical application, can be extracted from the text that the session includes valuable there are many valuable information in text
Information.
In alternatively possible application scenarios, user can in the search box of search engine search text, with
It scans for search engine using the search text as search key, may include many valuable in the search text
Information in practical application, can extract valuable information from the search text.
In another possible application scenarios, it is logical that the voice call function that user can use mobile terminal carries out voice
Words, can be there are many valuable information in the call audio which includes, can be to the conversation voice in practical application
Frequency carries out speech recognition, so that call text is converted by the call audio, it, can after the audio that will converse is converted into call text
To extract valuable information from the call text.
Certainly, above-mentioned application scenarios are only exemplary, and are not intended to limit this application, and in practical application, may be used also
There can be other scenes for needing to extract valuable information from text, the embodiment of the present application just no longer repeats one by one herein
.
In the related technology, name entity recognition techniques be can use extracts from text and name entity, and will extract
Name entity is determined as the valuable information extracted from text, however, in many cases, according only to name entity itself
Can not accurately reflect the meaning of the name entity, this cause the validity for the valuable information being drawn into from text compared with
Difference.
The embodiment of the present application provides a kind of text handling method, and this article treatment method can be improved to be extracted from text
The validity of the valuable information arrived.In this article treatment method, semantics recognition can be carried out to text to be processed, obtained
To at least one semantics recognition as a result, and Entity recognition is named to text to be processed, obtain at least one name entity,
Then, by least one semantics recognition result, at least one name entity is associated with this, obtains the text to be processed
Association results, wherein the association results be used for characterize each semantics recognition result and name entity between corresponding relationship, this
Sample, so that it may using above-mentioned association results as the valuable information extracted from text to be processed, due to association results
It characterizes each semantics recognition result and names the corresponding relationship between entity, accordingly, it is possible to using semantics recognition result come anti-
The meaning of name entity is reflected, so, so that it may improve the validity for the valuable information being drawn into from text
In the following, by being carried out briefly to implementation environment involved by text handling method provided by the embodiments of the present application
It is bright.
Fig. 1 is a kind of schematic diagram of implementation environment provided by the embodiments of the present application.As shown in Figure 1, the implementation environment can be with
Including server 101 and terminal 102, wherein server 101 and terminal 102 can be by wired or wirelessly carry out
Communication.
In implementation environment shown in Fig. 1, terminal 102 can obtain text to be processed based on different application scenarios,
After getting text to be processed, which can be sent to server 101 by terminal 102, by servicing
Device 101 executes text handling method provided by the embodiments of the present application to the text to be processed.
It should be pointed out that in some possible implementations, text handling method institute provided by the embodiments of the present application
The implementation environment being related to can only include terminal 102.In the case where implementation environment only includes terminal 102, terminal 102 can be with
Text to be processed is obtained based on different application scenarios, after getting text to be processed, terminal 102 can be waited for this
The text of processing executes text handling method provided by the embodiments of the present application.
It may also be noted that in other possible implementations, text-processing side provided by the embodiments of the present application
Implementation environment involved by method can only include server 101.In the case where implementation environment only includes server 101, clothes
Business device 101 can safeguard a database, can store pending text in the database, which can be
The text got under different application scenarios, server 101 can execute the text to be processed stored in database
Text handling method provided by the embodiments of the present application.
Referring to FIG. 2, it illustrates a kind of flow chart of text handling method provided by the embodiments of the present application, at the text
It (is hereafter collectively referred to as computer to set in the server or terminal that reason method can be applied in implementation environment described above
It is standby).As shown in Fig. 2, this article treatment method may comprise steps of:
Step 201, computer equipment carry out semantics recognition to text to be processed, obtain at least one semantics recognition knot
Fruit.
As described above, which can be the text got in different application scenarios.For example, In
In application scenes, the available session content of computer equipment, wherein the session content can be user and utilize Instant Messenger
News application conversates and the content that generates, the session content be also possible to user using mobile terminal voice call function into
Row voice communication and the content generated, the session content may include multiple texts to be processed, and in step 201, computer is set
It is standby semantics recognition to be carried out to multiple text to be processed respectively.
Semantics recognition refers to the technology identified to the meaning that text includes, passes through semantics recognition, computer equipment
The information such as intention expressed by available text and thought.In step 201, computer equipment can be to text to be processed
This progress semantics recognition, to obtain at least one semantics recognition as a result, at least one semantics recognition result can reflect this
The meaning of text to be processed.For example, text to be processed can be for " I want to buy the house of 100 flat left and right, and it is left often to put down 10,003
It is right just similar ", by carrying out semantics recognition to the text to be processed, available two semantics recognitions are as a result, this two
Semantics recognition result is respectively " intention area " and " intention price ", the two semantics recognition results can reflect text to be processed
This meaning.
Step 202, computer equipment are named Entity recognition to text to be processed, and it is real to obtain at least one name
Body.
Name entity refers to name, mechanism name, place name, number, date, currency and others with entitled mark
Entity.In step 202, computer equipment can be named Entity recognition to text to be processed, to obtain at least one
A name entity.For example, text to be processed can for " I want to buy the house of 100 flat left and right, often put down 10,003 or so just it is poor not
It is more ", by being named Entity recognition to the text to be processed, it is real that two names can be obtained from the text to be processed
Body, this two name entities are respectively " 100 " and " 10,003 ".
At least one semantics recognition result and at least one name entity are associated by step 203, computer equipment, are obtained
To the association results of text to be processed.
Wherein, the association results of the text to be processed are for characterizing between each semantics recognition result and name entity
Corresponding relationship, in one embodiment of the application, it is so-called " association results characterize each semantics recognition result and name entity it
Between corresponding relationship " refer to: association results include each semantics recognition result and name entity corresponding relationship.
For example, text to be processed can for " I want to buy the house of 100 flat left and right, often put down 10,003 or so just it is poor not
It is more ", at least one semantics recognition result of the text to be processed may include " intention area " and " intention price ", be somebody's turn to do wait locate
At least one name entity of the text of reason may include " 100 " and " 10,003 ", then the association results of the text to be processed
It can be with are as follows: intention area=100, intention price=10,003.
In practical applications, when text to be processed includes multiple semantics recognition results, multiple semantics recognition result
It may include identical semantics recognition result.For example, text to be processed can be for " I wants to buy the house of 100 flat left and right, my younger brother
Younger brother wants to buy the house of 150 flat left and right, and two houses often put down all 10,003 or so just almost ", for the text to be processed, warp
Semantics recognition is crossed, available 3 semantics recognitions are as a result, 3 semantics recognition results are respectively " intention area ", " intention face
Product " and " intention price ", wherein include in 3 semantics recognition results identical semantics recognition result " intention area ".
Since when text to be processed includes multiple semantics recognition results, multiple semantics recognition result may include phase
With semantics recognition as a result, therefore, may include semantics recognition knot in the association results of text to be processed obtained in step 203
The identical corresponding relationship of fruit.For example, text to be processed can be for " I want to buy the house of 100 flat left and right, my younger brother wants to buy 150
The house of flat left and right, two houses often put down all 10,003 or so just almost ", the association results of the text to be processed can be with
Are as follows: intention area=100, intention area=150, intention price=10,003, wherein intention area=100 and intention face
Semantics recognition result in=150 the two corresponding relationships of product is identical.
In view of the above, in order to guarantee the accuracy for the valuable information extracted from text to be processed with
And integrality, optionally, it is identical that computer equipment can extract semantics recognition result from the association results of text to be processed
The first corresponding relationship and the second corresponding relationship, wherein first corresponding relationship include target semanteme recognition result and first life
The corresponding relationship of name entity, second corresponding relationship include the corresponding relationship of target semanteme recognition result and the second name entity,
Then, computer equipment can merge the first corresponding relationship and the second corresponding relationship, obtain merging corresponding relationship,
In, which includes the corresponding relationship of target semanteme recognition result and name entity sets, the name entity sets
Including the first name entity and the second name entity.
For example, if the association results of text to be processed can be with are as follows: intention area=100, intention area=150, intention
Price=10,003, then computer equipment can extract the identical corresponding relationship of semantics recognition result from the association results,
The identical corresponding relationship of semantics recognition result is intention area=100 and intention area=150, and then, computer equipment can
To merge the identical corresponding relationship of semantics recognition result, obtain merging corresponding relationship, which can be with
Are as follows: intention area={ 100,150 }.
In practical application, if the text to be processed in the embodiment of the present application is in multiple texts that session content includes
At one, optionally, the association results for each text that the available session content of computer equipment includes, and out of session
The identical corresponding relationship of semantics recognition result is extracted in the association results for each text that appearance includes, then, computer equipment can
To merge the identical corresponding relationship of semantics recognition result according to similar mode described above, obtain merging correspondence
Relationship, to guarantee the accuracy and integrality for the valuable information extracted.
For example, text A to be processed is one in multiple texts that session content includes, wherein session content includes
Multiple texts be respectively text A, text B and text C, then computer equipment can obtain text A, text B and text C respectively
Association results, and the identical corresponding relationship of semantics recognition result, then, computer are extracted from the association results got
Equipment can merge the identical corresponding relationship of semantics recognition result according to similar mode described above, be closed
And corresponding relationship.
In text handling method provided by the embodiments of the present application, by carrying out semantics recognition to text to be processed, obtain
To at least one semantics recognition as a result, and Entity recognition is named to text to be processed, obtain at least one name entity,
Then, by least one semantics recognition result, at least one name entity is associated with this, obtains the text to be processed
Association results, wherein the association results be used for characterize each semantics recognition result and name entity between corresponding relationship, this
Sample, so that it may using above-mentioned association results as the valuable information extracted from text to be processed, due to association results
It characterizes each semantics recognition result and names the corresponding relationship between entity, accordingly, it is possible to using semantics recognition result come anti-
The meaning of name entity is reflected, so, so that it may improve the validity for the valuable information being drawn into from text.
On the basis of embodiment described above, the embodiment of the present application provides two kinds at least one semantics recognition knot
The method that fruit and at least one name entity are associated.In the following, the embodiment of the present application will carry out letter to the first correlating method
Illustrate.
As shown in figure 3, the first is by least one semantics recognition result and at least one side for being associated of name entity
Method may comprise steps of:
Step 301, computer equipment carry out semantic segmentation to text to be processed according at least one semantics recognition result,
Obtain at least one text fragments.
Wherein, each text fragments are corresponding with a semantics recognition result.
For example, text to be processed is " I want to buy the house of 100 flat left and right, often puts down 10,003 or so just almost ", it should
At least one semantics recognition result of text to be processed may include " intention area " and " intention price ", in step 301,
Semantic segmentation can be carried out to text to be processed according to the semantics recognition result " intention area " and " intention price ", obtain two
A text fragments, two text fragments are respectively as follows: " I wants to buy the house of 100 flat left and right " and " it is just poor often to put down 10,003 or so
Seldom ", wherein text fragments " I wants to buy the house of 100 flat left and right " and semantics recognition result " intention area " corresponding, text
Segment " often putting down 10,003 or so just almost " is corresponding with semantics recognition result " intention price ".
Step 302, computer equipment are named Entity recognition to each text fragments, and it is corresponding to obtain each text fragments
Name entity.
Semantic segmentation is being carried out to text to be processed according at least one semantics recognition result, is obtaining at least one text
After segment, computer equipment can execute step 202, in other words, step 302 according to the technical process of step 302
Technical process is an optional specific implementation of step 202.
Such as the example above, in step 302, computer equipment can respectively to " I wants to buy the house of 100 flat left and right " and
" often putting down 10,003 or so just almost " the two text fragments are named Entity recognition respectively, for " I wants to buy a 100 flat left sides
This text fragments of right house ", computer equipment are named Entity recognition to it, and obtained name entity is " 100 ", right
In " often putting down 10,003 or so just almost " this text fragments, computer equipment is named Entity recognition to it, obtains
Naming entity is " 10,003 ".
Step 303, computer equipment are by the corresponding name entity of each text fragments and the corresponding language of each text fragments
Adopted recognition result is associated, and obtains the association results of each text fragments.
Entity recognition is being named to each text fragments, after obtaining the corresponding name entity of each text fragments,
Computer equipment can execute step 203 according to the technical process of step 303 and step 304, in other words, step 303 and
The technical process of step 304 is an optional specific implementation of step 203.
Such as the example above, " I wants to buy the house of 100 flat left and right " the corresponding semantics recognition result of this text fragments is " meaning
To area ", corresponding name entity is " 100 ", " often putting down 10,003 or so just almost " corresponding semanteme of this text fragments
Recognition result is " intention price ", and corresponding name entity is " 10,003 ", and in step 303, computer equipment " can will anticipate
To area " and " 100 " be associated, obtain the association results of " I wants to buy the house of 100 flat left and right " this text fragments, calculate
" intention price " and " as soon as 10,003 " can be associated by machine equipment, obtain " often putting down 10,003 or so almost " this text
The association results of segment.
Step 304, computer equipment obtain the association knot of text to be processed according to the association results of each text fragments
Fruit.
Such as the example above, in step 304, computer equipment can according to " I wants to buy the house of 100 flat left and right " this
The association results of this text fragments of the association results of text fragments and " often putting down 10,003 or so just almost " obtain to be processed
Text " I want to buy the house of 100 flat left and right, often puts down 10,003 or so just almost " association results.
In the following, the embodiment of the present application will be briefly described second of correlating method.As shown in figure 4, second near
The method that a few semantics recognition result is associated at least one name entity may comprise steps of:
Step 401, computer equipment obtain each according to preset name entity rule from least one name entity
The corresponding name entity of semantics recognition result.
In second of correlating method, computer equipment can be executed according to the technical process of step 401 and step 402
Step 203, in other words, the technical process of step 401 and step 402 is an optional specific implementation of step 203.
" preset name entity rule " in step 401 is used to indicate the corresponding name entity of each semantics recognition result
The condition of required satisfaction.
In practical application, the corresponding name entity of different semantics recognition results can generally meet different conditions, for example,
Under normal conditions, the corresponding name entity of " intention area " this semantics recognition result needs to meet " integer no more than 500 "
This condition, the corresponding name entity of " intention price " this semantics recognition result need meet " number greater than 7000 " this
Condition.
In view of this, in the embodiment of the present application, technical staff can preset name entity rule, the name entity
Rule can indicate the condition met needed for the corresponding name entity of each semantics recognition result.Computer equipment is to be processed
Text carry out semantics recognition and name identification identification, obtain at least one semantics recognition result and at least one name entity it
Afterwards, the condition met needed for the corresponding name entity of each semantics recognition result can be obtained according to the name entity rule, and
Afterwards, computer equipment can be selected at least one name entity from this meets the corresponding name of each semantics recognition result in fact
The name entity of the condition met needed for body.
Such as the example above, computer equipment carries out semantics recognition and name Entity recognition to text to be processed, obtains
Semantics recognition result is " intention area " and " intention price ", and obtained name entity is " 100 " and " 10,003 ", wherein " meaning
To area " the corresponding name entity of this semantics recognition result need meet " integer no more than 500 " this condition, due to life
Name entity " 100 " meets above-mentioned condition in name entity " 100 " and " 10,003 ", therefore, " intention area " this semantics recognition
As a result corresponding name entity is " 100 ", and the corresponding name entity of " intention price " this semantics recognition result needs to meet " big
In 7000 number " this condition, since name entity " 10,003 " meets above-mentioned item in name entity " 100 " and " 10,003 "
Part, therefore, the corresponding name entity of " intention price " this semantics recognition result are " 10,003 ".
Step 402, computer equipment generate text to be processed according to the corresponding name entity of each semantics recognition result
Association results.
Such as the example above, computer equipment obtains the corresponding name entity of " intention area " this semantics recognition result and is
" 100 ", the corresponding name entity of " intention price " this semantics recognition result are " 10,003 ", and in step 402, computer is set
The standby association results that text to be processed can be generated according to above-mentioned two corresponding relationship.
Referring to FIG. 5, it illustrates the flow chart of another text handling method provided by the embodiments of the present application, the text
The server or terminal that processing method can be applied in implementation environment described above (that is to say computer as described herein
Equipment) in.As shown in figure 5, on the basis of embodiment described above, after step 203, this article treatment method can be with
The following steps are included:
Step 501, computer equipment obtained from database of content items according to the association results of text to be processed with to
The matched object content item of the association results of the text of processing.
Wherein, which is stored with multiple content items.
Under normal conditions, extracting valuable information from text to be processed, (in the embodiment of the present application, this has
The information of value is the association results of text to be processed) after, computer equipment can be executed according to the valuable information
Operation correspondingly.
In one possible implementation, computer equipment can push content to user according to the valuable information
, optionally, which can be advertisement etc..
In order to precisely push content item to user, computer equipment can be according to the association results of text to be processed from interior
Hold the matched object content item of association results obtained in association database with text to be processed, for example, computer equipment can be with
House is obtained from database of content items according to the association results of " intention area=100 " and " intention price=10,003 " to go out
Sell advertisement, should the area in house to be vended corresponding to House to let advertisement can be 100 flat left and right, price can be often to put down one
10003 or so, this way it is secured that the accurate push of advertisement.
Object content item is pushed to the corresponding user of text to be processed by step 502, computer equipment.
In the embodiment of the present application, which it is to be processed can be pushed to this by internet by computer equipment
The corresponding user of text.
So-called " the corresponding user of text to be processed " what is referred to can be text to be processed from mobile terminal institute
Corresponding user.
Referring to FIG. 6, it illustrates a kind of block diagram of text processing apparatus 600 provided by the embodiments of the present application, the text
Processing unit 600 can be configured in server or terminal.As shown in fig. 6, text processing unit 600 may include: language
Adopted identification module 601, name Entity recognition module 602 and relating module 603.
Wherein, the semantics recognition module 601 obtains at least one language for carrying out semantics recognition to text to be processed
Adopted recognition result.
It is real to obtain at least one name for being named Entity recognition to the text for the name Entity recognition module 602
Body.
The relating module 603, at least one name entity to close with this by least one semantics recognition result
Connection, obtains the association results of the text, the association results of the text for characterize each semantics recognition result and name entity it
Between corresponding relationship.
In one embodiment of the application, which is specifically used for: according to preset name entity rule,
The corresponding name entity of each semantics recognition result is obtained at least one name entity from this, wherein the name entity rule
It is used to indicate the condition met needed for the corresponding name entity of each semantics recognition result；According to each semantics recognition result pair
The name entity answered, generates the association results of the text.
With reference to Fig. 7, the embodiment of the present application also provides another text processing apparatus 700, text processing unit 700 is removed
Outside each module for including including text processing apparatus 600, optionally, text processing unit 700 can also include: segmentation mould
Block 604, extraction module 605, merging module 606, content item obtain module 607, pushing module 608 and acquisition conversation module 609.
Wherein, segmentation module 604, for carrying out semantic point to the text according at least one semantics recognition result
It cuts, obtains at least one text fragments, wherein each text segment is corresponding with a semantics recognition result.
Accordingly, the name Entity recognition module 602, is specifically used for: being named entity to each text segment and knows
Not, the corresponding name entity of each text segment is obtained.
The relating module 603, is specifically used for: by the corresponding name entity of each text segment and each text segment
Corresponding semantics recognition result is associated, and obtains the association results of each text segment；According to each text segment
Association results obtain the association results of the text.
The association results of the text include pair of each semantics recognition result and name entity in one of the embodiments,
It should be related to.
The extraction module 605 is corresponded to for extracting semantics recognition result identical first from the association results of the text
Relationship and the second corresponding relationship, wherein first corresponding relationship includes pair of target semanteme recognition result and the first name entity
It should be related to, which includes the corresponding relationship of the target semanteme recognition result and the second name entity.
The merging module 606 obtains merging pair for merging to first corresponding relationship and second corresponding relationship
It should be related to；Wherein, which includes the corresponding relationship of the target semanteme recognition result and name entity sets, the life
Name entity sets include the first name entity and the second name entity.
The content item obtains module 607, obtains and is somebody's turn to do from database of content items for the association results according to the text
The matched object content item of the association results of text, the database of content items are stored with multiple content items.
The pushing module 608, for the object content item to be pushed to the corresponding user of the text.
The acquisition conversation module 609 includes multiple texts to be processed in the session content for obtaining session content.
Above method embodiment, realization principle and skill may be implemented in text processing apparatus provided by the embodiments of the present application
Art effect is similar, and details are not described herein.
Specific about text processing apparatus limits the restriction that may refer to above for text handling method, herein not
It repeats again.Modules in above-mentioned text processing apparatus can be realized fully or partially through software, hardware and combinations thereof.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, the corresponding operation of the above modules is executed in order to which processor calls.
In one embodiment of the application, a kind of computer equipment is provided, which can be server
Or terminal, internal structure chart can be as shown in Figure 8.The computer equipment include by system bus connect processor and
Memory.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system and computer program.This is interior
Memory provides environment for the operation of operating system and computer program in non-volatile memory medium.The computer program quilt
To realize a kind of text handling method when processor executes.
It will be understood by those skilled in the art that structure shown in Fig. 8, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment of the application, a kind of computer equipment is provided, which can be server
Or terminal, the computer equipment include memory and processor, are stored with computer program in memory, which executes
It is performed the steps of when computer program
Semantics recognition is carried out to text to be processed, obtains at least one semantics recognition result；The text is named
Entity recognition obtains at least one name entity；By at least one semantics recognition result and this at least one name entity into
Row association, obtains the association results of the text, and the association results of the text are real for characterizing each semantics recognition result and name
Corresponding relationship between body.
In one embodiment of the application, processor is also performed the steps of according to this extremely when executing computer program
A few semantics recognition result carries out semantic segmentation to the text, obtains at least one text fragments, wherein each this article this film
Duan Yuyi semantics recognition result is corresponding；Entity recognition is named to each text segment, obtains each this article this film
The corresponding name entity of section.
In one embodiment of the application, processor also performs the steps of when executing computer program to be somebody's turn to do each
The corresponding name entity of text fragments and the corresponding semantics recognition result of each text segment are associated, and obtain each this article
The association results of this segment；According to the association results of each text segment, the association results of the text are obtained.
In one embodiment of the application, processor is also performed the steps of when executing computer program according to default
Name entity rule, from this at least one name entity in obtain the corresponding name entity of each semantics recognition result, wherein
The name entity rule is used to indicate the condition met needed for the corresponding name entity of each semantics recognition result；It is each according to this
The corresponding name entity of semantics recognition result, generates the association results of the text.
In one embodiment of the application, the association results of the text include each semantics recognition result and name entity
Corresponding relationship；In one embodiment of the application, processor is also performed the steps of when executing computer program from this article
Identical first corresponding relationship of semantics recognition result and the second corresponding relationship are extracted in this association results, wherein this first pair
It should be related to the corresponding relationship including target semanteme recognition result and the first name entity, which includes the target language
The corresponding relationship of adopted recognition result and the second name entity；First corresponding relationship and second corresponding relationship are merged,
It obtains merging corresponding relationship；Wherein, which includes pair of the target semanteme recognition result and name entity sets
It should be related to, which includes the first name entity and the second name entity.
In one embodiment of the application, processor is also performed the steps of when executing computer program according to this article
This association results obtain and the matched object content item of the association results of the text, the content item number from database of content items
Multiple content items are contained according to inventory；The object content item is pushed to the corresponding user of the text.
In one embodiment of the application, processor also performs the steps of acquisition session when executing computer program
Content includes multiple texts to be processed in the session content.
Computer equipment provided by the embodiments of the present application, implementing principle and technical effect and above method embodiment class
Seemingly, details are not described herein.
In one embodiment of the application, a kind of computer readable storage medium is provided, computer is stored thereon with
Program performs the steps of when computer program is executed by processor
Semantics recognition is carried out to text to be processed, obtains at least one semantics recognition result；The text is named
Entity recognition obtains at least one name entity；By at least one semantics recognition result and this at least one name entity into
Row association, obtains the association results of the text, and the association results of the text are real for characterizing each semantics recognition result and name
Corresponding relationship between body.
Basis is also performed the steps of in one embodiment of the application, when computer program is executed by processor should
At least one semantics recognition result carries out semantic segmentation to the text, obtains at least one text fragments, wherein each text
Segment is corresponding with a semantics recognition result；Entity recognition is named to each text segment, obtains each text
The corresponding name entity of segment.
In one embodiment of the application, also performing the steps of when computer program is executed by processor will be each
The corresponding name entity of text segment and the corresponding semantics recognition result of each text segment are associated, and obtain each be somebody's turn to do
The association results of text fragments；According to the association results of each text segment, the association results of the text are obtained.
It is also performed the steps of in one embodiment of the application, when computer program is executed by processor according to pre-
If name entity rule, from this at least one name entity in obtain the corresponding name entity of each semantics recognition result,
In, which is used to indicate the condition met needed for the corresponding name entity of each semantics recognition result；According to this
The corresponding name entity of each semantics recognition result, generates the association results of the text.
The association results of the text include pair of each semantics recognition result and name entity in one of the embodiments,
It should be related to；In one embodiment of the application, also perform the steps of when computer program is executed by processor from the text
Association results in extract identical first corresponding relationship of semantics recognition result and the second corresponding relationship, wherein this is first corresponding
Relationship includes the corresponding relationship of target semanteme recognition result and the first name entity, which includes target semanteme
The corresponding relationship of recognition result and the second name entity；First corresponding relationship and second corresponding relationship are merged, obtained
To merging corresponding relationship；Wherein, which includes that the target semanteme recognition result is corresponding with name entity sets
Relationship, the name entity sets include the first name entity and the second name entity.
Basis is also performed the steps of in one embodiment of the application, when computer program is executed by processor should
The association results of text obtain and the matched object content item of the association results of the text, the content item from database of content items
Database purchase has multiple content items；The object content item is pushed to the corresponding user of the text.
In one embodiment of the application, is also performed the steps of when computer program is executed by processor and obtain meeting
Content is talked about, includes multiple texts to be processed in the session content.
Computer readable storage medium provided in this embodiment, implementing principle and technical effect and above method embodiment
Similar, details are not described herein.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
The limitation to claim therefore cannot be interpreted as.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of text handling method, which is characterized in that the described method includes:
Semantics recognition is carried out to text to be processed, obtains at least one semantics recognition result；
Entity recognition is named to the text, obtains at least one name entity；
At least one described semantics recognition result and at least one described name entity are associated, the pass of the text is obtained
Connection is as a result, the association results of the text are used to characterize each semantics recognition result and name the corresponding relationship between entity.
2. being obtained the method according to claim 1, wherein described be named Entity recognition to the text
Before at least one name entity, the method also includes:
Semantic segmentation is carried out to the text according at least one described semantics recognition result, obtains at least one text fragments,
Wherein, each text fragments are corresponding with a semantics recognition result；
Accordingly, described that Entity recognition is named to the text, obtain at least one name entity, comprising:
Entity recognition is named to each text fragments, obtains the corresponding name entity of each text fragments.
3. according to the method described in claim 2, it is characterized in that, it is described will at least one described semantics recognition result with it is described
At least one name entity is associated, and obtains the association results of the text, comprising:
Each corresponding name entity of text fragments and the corresponding semantics recognition result of each text fragments are carried out
Association, obtains the association results of each text fragments；
According to the association results of each text fragments, the association results of the text are obtained.
4. the method according to claim 1, wherein it is described will at least one described semantics recognition result with it is described
At least one name entity is associated, and obtains the association results of the text, comprising:
According to preset name entity rule, it is corresponding from least one described name entity to obtain each semantics recognition result
Name entity, wherein the name entity rule, which is used to indicate needed for the corresponding name entity of each semantics recognition result, to be met
Condition；
According to the corresponding name entity of each semantics recognition result, the association results of the text are generated.
5. method according to any one of claims 1 to 4, which is characterized in that the association results of the text include each
The corresponding relationship of semantics recognition result and name entity；It is described will at least one described semantics recognition result and it is described at least one
Name entity is associated, after obtaining the association results of the text, the method also includes:
Identical first corresponding relationship of semantics recognition result and the second corresponding relationship are extracted from the association results of the text,
In, first corresponding relationship includes the corresponding relationship of target semanteme recognition result and the first name entity, and described second is corresponding
Relationship includes the corresponding relationship of the target semanteme recognition result and the second name entity；
First corresponding relationship and second corresponding relationship are merged, obtain merging corresponding relationship；Wherein, the conjunction
And corresponding relationship includes the corresponding relationship of the target semanteme recognition result and name entity sets, the name entity sets packet
Include the first name entity and the second name entity.
6. method according to any one of claims 1 to 4, which is characterized in that described by least one described semantics recognition
As a result it is associated at least one described name entity, after obtaining the association results of the text, the method also includes:
It is obtained from database of content items according to the association results of the text and the matched target of the association results of the text
Content item, the database of content items are stored with multiple content items；
The object content item is pushed to the corresponding user of the text.
7. method according to any one of claims 1 to 4, which is characterized in that the method also includes:
Session content is obtained, includes multiple texts to be processed in the session content.
8. a kind of text processing apparatus, which is characterized in that described device includes:
Semantics recognition module obtains at least one semantics recognition result for carrying out semantics recognition to text to be processed；
Entity recognition module is named, for being named Entity recognition to the text, obtains at least one name entity；
Relating module is obtained at least one described semantics recognition result and at least one described name entity to be associated
To the association results of the text, the association results of the text are for characterizing between each semantics recognition result and name entity
Corresponding relationship.
9. a kind of computer equipment, which is characterized in that including memory and processor, the memory is stored with computer journey
Sequence, which is characterized in that the text as described in claim 1 to 7 is any is realized when the computer program is executed by the processor
Treatment method.
10. a kind of computer readable storage medium, which is characterized in that be stored thereon with computer program, which is characterized in that the journey
The text handling method as described in claim 1 to 7 is any is realized when sequence is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910560145.XA CN110427611A (en) | 2019-06-26 | 2019-06-26 | Text handling method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910560145.XA CN110427611A (en) | 2019-06-26 | 2019-06-26 | Text handling method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110427611A true CN110427611A (en) | 2019-11-08 |
Family
ID=68408713
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910560145.XA Pending CN110427611A (en) | 2019-06-26 | 2019-06-26 | Text handling method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110427611A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021151353A1 (en) * | 2020-10-20 | 2021-08-05 | 平安科技（深圳）有限公司 | Medical entity relationship extraction method and apparatus, and computer device and readable storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246745A (en) * | 2013-05-22 | 2013-08-14 | 中国工商银行股份有限公司 | Device and method for processing data based on data warehouse |
| US20160092550A1 (en) * | 2014-09-30 | 2016-03-31 | Yahoo!, Inc. | Automated search intent discovery |
| CN106407178A (en) * | 2016-08-25 | 2017-02-15 | 中国科学院计算技术研究所 | Session abstract generation method and device |
| US20170097933A1 (en) * | 2015-10-05 | 2017-04-06 | Yahoo! Inc. | Methods, systems and techniques for ranking blended content retrieved from multiple disparate content sources |
| US20170249956A1 (en) * | 2016-02-29 | 2017-08-31 | International Business Machines Corporation | Inferring User Intentions Based on User Conversation Data and Spatio-Temporal Data |
| CN108170859A (en) * | 2018-01-22 | 2018-06-15 | 北京百度网讯科技有限公司 | Method, apparatus, storage medium and the terminal device of speech polling |
| CN108304466A (en) * | 2017-12-27 | 2018-07-20 | 中国银联股份有限公司 | A kind of user view recognition methods and user view identifying system |
| CN109063048A (en) * | 2018-07-18 | 2018-12-21 | 哈尔滨工业大学 | A kind of matched data cleaning method of knowledge based library figure and device |
| CN109461039A (en) * | 2018-08-28 | 2019-03-12 | 厦门快商通信息技术有限公司 | A kind of text handling method and intelligent customer service method |
-
2019
- 2019-06-26 CN CN201910560145.XA patent/CN110427611A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246745A (en) * | 2013-05-22 | 2013-08-14 | 中国工商银行股份有限公司 | Device and method for processing data based on data warehouse |
| US20160092550A1 (en) * | 2014-09-30 | 2016-03-31 | Yahoo!, Inc. | Automated search intent discovery |
| US20170097933A1 (en) * | 2015-10-05 | 2017-04-06 | Yahoo! Inc. | Methods, systems and techniques for ranking blended content retrieved from multiple disparate content sources |
| US20170249956A1 (en) * | 2016-02-29 | 2017-08-31 | International Business Machines Corporation | Inferring User Intentions Based on User Conversation Data and Spatio-Temporal Data |
| CN106407178A (en) * | 2016-08-25 | 2017-02-15 | 中国科学院计算技术研究所 | Session abstract generation method and device |
| CN108304466A (en) * | 2017-12-27 | 2018-07-20 | 中国银联股份有限公司 | A kind of user view recognition methods and user view identifying system |
| CN108170859A (en) * | 2018-01-22 | 2018-06-15 | 北京百度网讯科技有限公司 | Method, apparatus, storage medium and the terminal device of speech polling |
| CN109063048A (en) * | 2018-07-18 | 2018-12-21 | 哈尔滨工业大学 | A kind of matched data cleaning method of knowledge based library figure and device |
| CN109461039A (en) * | 2018-08-28 | 2019-03-12 | 厦门快商通信息技术有限公司 | A kind of text handling method and intelligent customer service method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021151353A1 (en) * | 2020-10-20 | 2021-08-05 | 平安科技（深圳）有限公司 | Medical entity relationship extraction method and apparatus, and computer device and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109543516A (en) | Signing intention judgment method, device, computer equipment and storage medium | |
| CN111325037B (en) | Text intention recognition method and device, computer equipment and storage medium | |
| CN109729383B (en) | Double-recording video quality detection method and device, computer equipment and storage medium | |
| CN109816399A (en) | Complain management method, device, computer equipment and the storage medium of part | |
| CN111984779B (en) | Dialogue text analysis method, device, equipment and readable medium | |
| CN109710402A (en) | Method, apparatus, computer equipment and the storage medium of process resource acquisition request | |
| KR20210088680A (en) | Video cutting method, apparatus, computer equipment and storage medium | |
| CN110457679B (en) | User portrait construction method, device, computer equipment and storage medium | |
| CN109815489A (en) | Collection information generating method, device, computer equipment and storage medium | |
| CN110032724B (en) | Method and device for recognizing user intention | |
| US20220131975A1 (en) | Method And Apparatus For Predicting Customer Satisfaction From A Conversation | |
| CN109947971B (en) | Image retrieval method, image retrieval device, electronic equipment and storage medium | |
| CN112015747B (en) | Data uploading method and device | |
| CN109831677B (en) | Video desensitization method, device, computer equipment and storage medium | |
| CN109766474A (en) | Inquest signal auditing method, device, computer equipment and storage medium | |
| CN113435862B (en) | Bill processing method and device based on mailbox | |
| CN111382570B (en) | Text entity recognition method, device, computer equipment and storage medium | |
| US8862609B2 (en) | Expanding high level queries | |
| CN110427611A (en) | Text handling method, device, equipment and storage medium | |
| CN114282019A (en) | Target multimedia data searching method and device, computer equipment and storage medium | |
| WO2022021304A1 (en) | Method for identifying highlight clip in video on basis of bullet screen, and terminal and storage medium | |
| US11373057B2 (en) | Artificial intelligence driven image retrieval | |
| CN111368061B (en) | Short text filtering method, device, medium and computer equipment | |
| CN111382569B (en) | Method and device for identifying entity in dialogue corpus and computer equipment | |
| CN109543177A (en) | Message data processing method, device, computer equipment and storage medium |
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: 20191108 |