CN107169092A - Intelligent Recognition and the method and system of sensitive content are handled in interaction - Google Patents
Intelligent Recognition and the method and system of sensitive content are handled in interaction Download PDFInfo
- Publication number
- CN107169092A CN107169092A CN201710334441.9A CN201710334441A CN107169092A CN 107169092 A CN107169092 A CN 107169092A CN 201710334441 A CN201710334441 A CN 201710334441A CN 107169092 A CN107169092 A CN 107169092A
- Authority
- CN
- China
- Prior art keywords
- sensitive content
- automatic machine
- search tree
- interaction
- sensitive
- 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
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/374—Thesaurus
-
- 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
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention discloses Intelligent Recognition and the method for handling sensitive content in interaction, including：Default sensitive content is compiled into automatic machine, automatic machine search tree is built, automatic machine search tree is stored to telecommunication network and asked on service platform；Parsing interaction content is obtained interacting character, and interaction character is word for word read in automatic machine search tree by remote service agreement, and the sensitive content in interaction content is obtained by interaction character traversal automatic machine search tree；The number of times of sensitive content appearance is obtained, strategy is handled with reference to the sensitive content pre-set, processing operation corresponding to client executing updates automatic machine search tree；By in the corresponding processing policy store of the automatic machine search tree and sensitive content after renewal to telecommunication network request service platform；When the number of times occurred in interaction character in interaction is more than or equal to sensitive number of times, interaction character is fed back into management port.The present invention is managed collectively to yellow version vocabulary, improves the treatment effeciency of yellow version vocabulary.
Description
Technical field
The present invention relates to the technical field of network interaction management, more particularly, to Intelligent Recognition in a kind of interaction
And handle the method and system of sensitive content.
Background technology
With the development of network technology, various social networking application programs (APP), the release of social platform, social activity chat is gradually
Become the information interaction approach that people commonly use, but personnel are intricate during social activity, chat content be also it is various,
Chat content is more, and to ensure chat quality, social intercourse system needs to shield some uncivil or illegal sensitive vocabulary or hair is wide
Accuse (i.e. social sensitive content, what is also had is called yellow version vocabulary).
At present, in the prior art by the way that chat vocabulary is identified whether with comparison search in default yellow version database
For yellow version vocabulary, if it find that there is yellow version vocabulary, the ID for the person of being used for is shielded.But, if user is using other shapes
Formula exical substitution is present in the yellow version vocabulary in dictionary, " 8 " is such as replaced with to the form of " eight ", with regard to that can bypass shielding, it is impossible to reach
To the purpose of expected yellow version vocabulary shielding.And network words update is so fast, existing system is to emerging sensitivity
The automatic identification ability of content is strong not enough, can not include these emerging sensitive words intelligently, in time in database
Remittance content.
Furthermore, the combination for multiple vocabulary that Chinese and English is combined is with regard to that can obtain a variety of vocabulary implications, for such
As many as combining form, the larger and accuracy by the way of the sensitive vocabulary of existing sensitive lexicon matching contrast inquiry
It is not high, in addition it is also necessary to which that the longer matching comparison time of consumption, execution efficiency is not high, it is impossible to change existing quick comprehensively, exactly
Feel using and propagating for vocabulary.And the complaint process after maloperation shielding user is very complicated, very bad is caused to user
Usage experience.
Therefore it provides a kind of intelligent, degree of accuracy is high, easy to operate and interaction of efficiency high in sensitive content handle
Scheme is this area urgent problem to be solved.
The content of the invention
In view of this, the invention provides Intelligent Recognition in a kind of interaction and handle the method for sensitive content and be
System, solve intelligent can not focus on the technical problem of sensitive content in interaction in the prior art.
In order to solve the above-mentioned technical problem, the present invention proposes in a kind of interaction Intelligent Recognition and handles sensitive content
Method, including：
Default sensitive content is received, the default sensitive content is compiled into automatic machine according to the mechanism of AC automatic machines；Will
The endianness order that the automatic machine is encoded according to default keyword, and will be described automatic according to the endianness order
Mechanism builds up automatic machine search tree, and the automatic machine search tree is stored to telecommunication network request service platform；
The interaction content of client is received, the interaction content is parsed and obtains interacting character, existed by remote service agreement
The interactive character is word for word read in the automatic machine search tree, traveling through the automatic machine search tree by the interactive character obtains
To the sensitive content in the interaction content；
The history intersection record of the client is searched for, the number of times that the sensitive content occurs is obtained, with reference to pre-setting
Sensitive content processing strategy, operation is handled accordingly to the client executing；
Based on the current endianness order, the automatic machine search tree is updated according to the sensitive content；
By the corresponding processing policy store of the automatic machine search tree and the sensitive content after renewal to described long-range
On network request service platform；
The number of times that the interactive character occurs in interaction is counted, is occurred in the interactive character in interaction
Number of times when being more than or equal to sensitive number of times, the interactive character is fed back into management port；
The sensitive content instruction of the management port is received, and is instructed according to the sensitive content by the interactive character more
In the new extremely automatic machine search tree.
Further, wherein, sensitive content processing strategy is：
When the sensitive content occurrence number is less than or equal to setting number of times, the sensitive content is shielded in interactive interface,
And send alert messages to the client；
When the sensitive content occurrence number is more than setting number of times and is less than or equal to setting frequency threshold value, in interactive interface
Middle shielding sensitive content, and by client's end shield preset time；
When the sensitive content occurrence number is more than the setting frequency threshold value, shielded in interactive interface in the sensitivity
Hold, and shield the client.
Further, wherein, the setting number of times, is one to three times；The setting frequency threshold value, is three to five times.
Further, wherein, the endianness order that the automatic machine is encoded according to default keyword, and according to institute
State endianness order and the automechanism is built up into automatic machine search tree, be：
The corresponding extension automatic machine of the automatic machine is obtained according to default automatic machine expanding policy；
The endianness order that the automatic machine and its extension automatic machine are encoded according to default keyword, and according to institute
State endianness order and the automechanism is built up into automatic machine search tree.
On the other hand, the present invention also provides Intelligent Recognition and the system for handling sensitive content in a kind of interaction, including：
Automatic machine search tree creation module, sensitive content acquisition module, sensitive content processing module, automatic machine search tree update module and
Sensitive content handles policy store module；Wherein,
The automatic machine search tree creation module, for being connected with the sensitive content acquisition module, receives default quick
Feel content, the default sensitive content is compiled into automatic machine according to the mechanism of AC automatic machines；By the automatic machine according to default
Keyword coding endianness order, and according to the endianness order by the automechanism build up automatic machine search
Tree, and the automatic machine search tree is stored to telecommunication network request service platform；
The sensitive content acquisition module, for the automatic machine search tree creation module and sensitive content processing module
It is connected, receives the interaction content of client, parses the interaction content and obtain interacting character, by remote service agreement in institute
State and the interactive character is word for word read in automatic machine search tree, traveling through the automatic machine search tree by the interactive character obtains
Sensitive content in the interaction content；
The sensitive content processing module, for the sensitive content acquisition module and automatic machine search tree update module
It is connected, searches for the history intersection record of the client, the number of times that the sensitive content occurs is obtained, with reference to what is pre-set
Sensitive content processing strategy, operation is handled to the client executing accordingly；
The automatic machine search tree update module, for tactful with the sensitive content processing module and sensitive content processing
Memory module is connected, based on the current endianness order, updates the automatic machine according to the sensitive content and searches for
Tree；
The sensitive content handles policy store module, for being connected with the automatic machine search tree update module, will
The corresponding processing policy store of the automatic machine search tree and the sensitive content after renewal asks to take to the telecommunication network
It is engaged on platform；
Further comprise：Default sensitive content update module, for being connected with the automatic machine search tree creation module,
The number of times that the interactive character occurs in interaction is counted, the number of times occurred in the interactive character in interaction is big
When sensitive number of times, the interactive character is fed back into management port；
The sensitive content instruction of the management port is received, and is instructed according to the sensitive content by the interactive character more
In the new extremely automatic machine search tree.
Further, wherein, the sensitive content processing module is：The number of times statistic unit of sensitive content appearance, first
Sensitive content processing unit, the second sensitive content processing unit and the 3rd sensitive content processing unit；Wherein,
The number of times statistic unit that the sensitive content occurs, for being searched for the sensitive content acquisition module, automatic machine
Tree update module and the first sensitive content processing unit are connected, and search for the history intersection record of the client, obtain described
The number of times that sensitive content occurs；
The first sensitive content processing unit, the number of times statistic unit and second for occurring with the sensitive content is quick
Sense content processing unit is connected, when the sensitive content occurrence number is less than or equal to setting number of times, in interactive interface
The sensitive content is shielded, and alert messages are sent to the client；
The second sensitive content processing unit, for the first sensitive content processing unit and the 3rd sensitive content
Processing unit is connected, when the sensitive content occurrence number is more than setting number of times and is less than or equal to setting frequency threshold value,
Shield the sensitive content in interactive interface, and by client's end shield preset time；
The 3rd sensitive content processing unit, it is described quick for being connected with the second sensitive content processing unit
When feeling content occurrence number more than the setting frequency threshold value, the sensitive content is shielded in interactive interface, and shield the visitor
Family end.
Further, wherein, the setting number of times, is one to three times；The setting frequency threshold value, is three to five times.
Further, wherein, the automatic machine search tree creation module is：Automatic machine acquiring unit and automatic machine search
Set creating unit；Wherein,
The automatic machine acquiring unit, for the sensitive content acquisition module and automatic machine search tree creating unit phase
Connection, automatic machine is compiled into according to the mechanism of AC automatic machines by the default sensitive content, and plan is extended according to default automatic machine
Slightly obtain the corresponding extension automatic machine of the automatic machine；
The automatic machine search tree creating unit, for being connected with the automatic machine acquiring unit, by the automatic machine
And its endianness order that extension automatic machine is encoded according to default keyword, and will be described according to the endianness order
Automechanism builds up automatic machine search tree, and the automatic machine search tree is stored to telecommunication network request service platform.
Compared with prior art, Intelligent Recognition and the method and system of sensitive content are handled in interaction of the invention,
Realize following beneficial effect：
(1) Intelligent Recognition and the method and system of sensitive content are handled in interaction of the present invention, using AC from
Motivation sets up search tree and carries out yellow version lexical search, processing, can comprehensively search for, handle various yellow version vocabulary and its deformation；
The yellow version lexical search service systems of the RPC for calling service can externally be provided by setting up, it is not necessary to build one to each interactive system
Yellow version vocabulary processing module, can also be managed collectively to yellow version vocabulary, improve the treatment effeciency of yellow version vocabulary.
(2) Intelligent Recognition and the method and system of sensitive content are handled in interaction of the present invention, using AC from
Motivation sets up search tree and carries out yellow version lexical search, processing, and sets the sensitive content processing scheme and automatic machine of stagewise to search
Suo Shu and emerging yellow version lexical data more new strategy, it is ensured that the promptness and accuracy of sensitive content processing.
Certainly, implement the present invention any product must not specific needs simultaneously reach all the above technique effect.
By referring to the drawings to the detailed description of the exemplary embodiment of the present invention, further feature of the invention and its
Advantage will be made apparent from.
Brief description of the drawings
The accompanying drawing for being combined in the description and constituting a part for specification shows embodiments of the invention, and even
It is used for the principle for explaining the present invention together with its explanation.
Fig. 1 be described in the embodiment of the present invention 1 in interaction Intelligent Recognition and handle sensitive content method flow
Schematic diagram；
Fig. 2 be described in the embodiment of the present invention 2 in interaction Intelligent Recognition and handle sensitive content method flow
Schematic diagram；
Fig. 3 be described in the embodiment of the present invention 3 in interaction Intelligent Recognition and handle sensitive content system structure
Schematic diagram；
Fig. 4 be described in the embodiment of the present invention 4 in interaction Intelligent Recognition and handle sensitive content system structure
Schematic diagram.
Embodiment
The various exemplary embodiments of the present invention are described in detail now with reference to accompanying drawing.It should be noted that：Unless had in addition
Body illustrates that the part and the positioned opposite of step, numerical expression and numerical value otherwise illustrated in these embodiments does not limit this
The scope of invention.
The description only actually at least one exemplary embodiment is illustrative below, never as to the present invention
And its any limitation applied or used.
It may be not discussed in detail for technology, method and apparatus known to person of ordinary skill in the relevant, but suitable
In the case of, the technology, method and apparatus should be considered as a part for specification.
In shown here and discussion all examples, any occurrence should be construed as merely exemplary, without
It is as limitation.Therefore, other examples of exemplary embodiment can have different values.
It should be noted that：Similar label and letter represents similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined, then it need not be further discussed in subsequent accompanying drawing in individual accompanying drawing.
Embodiment 1
As shown in figure 1, in the interaction described in the present embodiment Intelligent Recognition and handle sensitive content method flow
Schematic diagram, the method for present embodiments providing yellow version sensitive content in a kind of centralized intelligenceization processing interaction, this method bag
Include the steps：
Default sensitive content, is compiled into automatically by step 101, the default sensitive content of reception according to the mechanism of AC automatic machines
Machine；The endianness order that automatic machine is encoded according to default keyword, and built automatic machine according to endianness order
Stored into automatic machine search tree, and by automatic machine search tree to telecommunication network request service platform.
AC automatic machines (Aho-Corasick), are a kind of string matching algorithms based on automatic machine principle, such as Fig. 1 institutes
Show, its basic functional principle is：Feature string (such as virus characteristic storehouse, filtering keys) is compiled into automatic machine first,
Since state 0, content to be matched is word for word read in, when reading in a character every time, checks whether current state has corresponding character
Redirect arrow, if so, then jumping to this redirects corresponding NextState, if not having, jump back to state 0.There are some state quilts
Labeled as matching status, if the match is successful into this status representative.The word that the interaction content of user is encoded with keyword
Symbol form can embody the word combination of the interaction content after carrying out sequential combination.
It will be stored on telecommunication network request service platform and be managed in sensitive word finder, ask to take by telecommunication network
Business platform connects each interactive service platform, advantageously ensures that the uniformity of sensitive vocabulary standard and the uniformity of the network information.
Step 102, the interaction content for receiving client, parsing interaction content obtain interacting character, are assisted by remote service
View word for word reads in interaction character in automatic machine search tree, and traveling through automatic machine search tree by interaction character obtains in interaction content
Sensitive content.
If not obtaining any sensitive content after traversal automatic machine search tree, any interference is not done to the interaction, only obtained
The interaction content of client is recorded, subsequently to use.
The history intersection record of step 103, search client, obtains the number of times of sensitive content appearance, with reference to pre-setting
Sensitive content processing strategy, it is corresponding to client executing processing operation.
Alternatively, intersection record of the search client within nearest a period of time, this can be 1 month to 1 for a period of time
In year., may if be not any limitation as to the period that recording interactive is recorded because netspeak update speed is quickly
Larger error is caused to sensitive content processing.
Step 104, based on current endianness order, automatic machine search tree is updated according to sensitive content.
The basic automatic machine character that sensitive content is searched for after the sensitive content that this search is obtained is likely to become, utilizes friendship
The sensitive content searched out during mutually updates in itself can preferably improve automatic machine search tree.
Step 105, by the corresponding processing policy store of the automatic machine search tree and sensitive content after renewal to telecommunication network
Ask on service platform.
Can be corresponding sensitive vocabulary or phase on other interaction platforms by storing the corresponding processing strategy of sensitive content
Foundation is provided like the processing of sensitive vocabulary, it might even be possible to which the sensitive content processing strategy of each interaction platform of intelligent comprehensive is obtained
The processing standard of the sensitive content.
The number of times that step 106, statistics interaction character occur in interaction, occurs in interaction character in interaction
Number of times when being more than or equal to sensitive number of times, interaction character is fed back into management port；The sensitive content for receiving management port refers to
Order, and interaction character is updated in automatic machine search tree according to sensitive content instruction.
Embodiment 2
As shown in Fig. 2 in the interaction described in the present embodiment Intelligent Recognition and handle sensitive content method flow
Schematic diagram, the present embodiment is a kind of preferred embodiment provided on the basis of above-described embodiment 1, and this method comprises the following steps：
Default sensitive content, is compiled into automatically by step 201, the default sensitive content of reception according to the mechanism of AC automatic machines
Machine, the corresponding extension automatic machine of automatic machine is obtained according to default automatic machine expanding policy.
Alternatively, automatic machine expanding policy can be：Chinese, phonetic, initials in chinese pinyin, English, the English of vocabulary
Initial, alternative word etc..
Step 202, the endianness order for encoding automatic machine and its extension automatic machine according to default keyword, and root
Automechanism is built up into automatic machine search tree according to endianness order, and automatic machine search tree is stored to telecommunication network request clothes
It is engaged on platform.
Step 203, the interaction content for receiving client, parsing interaction content obtain interacting character, are assisted by remote service
View word for word reads in interaction character in automatic machine search tree, and traveling through automatic machine search tree by interaction character obtains in interaction content
Sensitive content.
The history intersection record of step 204, search client, obtains the number of times of sensitive content appearance, with reference to pre-setting
Sensitive content processing strategy, it is corresponding to client executing processing operation.
When step 205, sensitive content occurrence number are less than or equal to setting number of times, shielded in interactive interface in the sensitivity
Hold, and alert messages are sent to client；Sensitive content occurrence number is more than setting number of times and less than or equal to setting number of times threshold
During value, the sensitive content is shielded in interactive interface, and by client's end shield preset time；Sensitive content occurrence number, which is more than, to be set
When determining frequency threshold value, the sensitive content is shielded in interactive interface, and shield client.
Alternatively, number of times is set, is further one to three times；Frequency threshold value is set, is further three to five times.
Step 206, based on current endianness order, automatic machine search tree is updated according to sensitive content.
Step 207, by the corresponding processing policy store of the automatic machine search tree and sensitive content after renewal to telecommunication network
Ask on service platform.
The number of times that step 208, statistics interaction character occur in interaction, occurs in interaction character in interaction
Number of times when being more than or equal to sensitive number of times, interaction character is fed back into management port；The sensitive content for receiving management port refers to
Order, and interaction character is updated in automatic machine search tree according to sensitive content instruction.
Embodiment 3
As shown in figure 3, in the interaction described in the present embodiment Intelligent Recognition and handle sensitive content system structure
Schematic diagram, system described in the present embodiment is used to implement described in above-described embodiment in interaction Intelligent Recognition and handled in sensitivity
The method of appearance, the system includes：Automatic machine search tree creation module 301, sensitive content acquisition module 302, sensitive content processing
Module 303, automatic machine search tree update module 304 and sensitive content processing policy store module 305.
Wherein, automatic machine search tree creation module 301 is connected with sensitive content acquisition module 302, default for receiving
Sensitive content, automatic machine is compiled into according to the mechanism of AC automatic machines by default sensitive content；By automatic machine according to default key
The endianness order of word coding, and automechanism is built up by automatic machine search tree according to endianness order, and by automatic machine
Search tree is stored to telecommunication network request service platform.
Sensitive content acquisition module 302 is connected with automatic machine search tree creation module 301 and sensitive content processing module 303
Connect, the interaction content for receiving client, parsing interaction content obtains interacting character, by remote service agreement in automatic machine
Interaction character is word for word read in search tree, is traveled through by interaction character in the sensitivity that automatic machine search tree is obtained in interaction content
Hold.
Sensitive content processing module 303 is connected with sensitive content acquisition module 302 and automatic machine search tree update module 304
Connect, for the history intersection record of search client, the number of times of sensitive content appearance is obtained, with reference to the sensitive content pre-set
Processing strategy, processing operation corresponding to client executing.
Automatic machine search tree update module 304 and sensitive content processing module 303 and sensitive content processing policy store mould
Block 305 is connected, for based on current endianness order, automatic machine search tree to be updated according to sensitive content.
Sensitive content processing policy store module 305 is connected with automatic machine search tree update module 304, for that will update
On the corresponding processing policy store of automatic machine search tree and sensitive content afterwards to telecommunication network request service platform.
Said system further comprises：Default sensitive content update module 306, for automatic machine search tree creation module
301 are connected, the number of times that statistics interaction character occurs in interaction, the number of times occurred in interaction character in interaction
During more than or equal to sensitive number of times, interaction character is fed back into management port；Receive the sensitive content instruction of management port, and root
Interaction character is updated in automatic machine search tree according to sensitive content instruction.
Embodiment 4
As shown in figure 4, in the interaction described in the present embodiment Intelligent Recognition and handle sensitive content system structure
Schematic diagram, the present embodiment is a kind of preferred embodiment provided on the basis of above-described embodiment 3, and the system includes：Automatic machine is searched
Suo Shu creation modules 401, sensitive content acquisition module 402, sensitive content processing module 403, automatic machine search tree update module
404 and sensitive content processing policy store module 405.
Wherein, automatic machine search tree creation module 401 is further：Automatic machine acquiring unit 411 and automatic machine search tree
Creating unit 412.
Automatic machine acquiring unit 411, for sensitive content acquisition module 402 and automatic machine search tree creating unit 412
It is connected, default sensitive content is compiled into automatic machine according to the mechanism of AC automatic machines, according to default automatic machine expanding policy
Obtain the corresponding extension automatic machine of automatic machine.
Automatic machine search tree creating unit 412, for being connected with automatic machine acquiring unit 411, by automatic machine and its expansion
The endianness order that exhibition automatic machine is encoded according to default keyword, and built up automechanism certainly according to endianness order
Motivation search tree, and automatic machine search tree is stored to telecommunication network request service platform.
Sensitive content acquisition module 402 is connected with automatic machine search tree creation module 401 and sensitive content processing module 403
Connect, the interaction content for receiving client, parsing interaction content obtains interacting character, by remote service agreement in automatic machine
Interaction character is word for word read in search tree, is traveled through by interaction character in the sensitivity that automatic machine search tree is obtained in interaction content
Hold.
Sensitive content processing module 403 is connected with sensitive content acquisition module 402 and automatic machine search tree update module 404
Connect, for the history intersection record of search client, the number of times of sensitive content appearance is obtained, with reference to the sensitive content pre-set
Processing strategy, processing operation corresponding to client executing.
Sensitive content processing module 403, further for：The number of times statistic unit 431, first that sensitive content occurs is sensitive interior
Hold processing unit 432, the second sensitive content processing unit 433 and the 3rd sensitive content processing unit 434.
Wherein, the number of times statistic unit 431 that sensitive content occurs, for being searched with sensitive content acquisition module 402, automatic machine
The sensitive content processing unit 432 of Suo Shu update modules 404 and first is connected, and the history intersection record of search client is obtained
The number of times that sensitive content occurs.
First sensitive content processing unit 432, the number of times statistic unit 431 and second for occurring with sensitive content is sensitive
Content processing unit 433 is connected, and when sensitive content occurrence number is less than or equal to setting number of times, is shielded in interactive interface
The sensitive content, and send alert messages to client.
Second sensitive content processing unit 433, for at the first sensitive content processing unit 432 and the 3rd sensitive content
Reason unit 434 is connected, when sensitive content occurrence number is more than setting number of times and is less than or equal to setting frequency threshold value, in interaction
Shield the sensitive content in interface, and by client's end shield preset time.
3rd sensitive content processing unit 434, for being connected with the second sensitive content processing unit 433, sensitive content
When occurrence number is more than setting frequency threshold value, the sensitive content is shielded in interactive interface, and shield client.
Alternatively, number of times is set, is further one to three times；Frequency threshold value is set, is further three to five times.
Automatic machine search tree update module 404 and sensitive content processing module 403 and sensitive content processing policy store mould
Block 405 is connected, for based on current endianness order, automatic machine search tree to be updated according to sensitive content.
Sensitive content processing policy store module 405 is connected with automatic machine search tree update module 404, for that will update
On the corresponding processing policy store of automatic machine search tree and sensitive content afterwards to telecommunication network request service platform.
Said system further comprises：Default sensitive content update module 406, for automatic machine search tree creation module
401 are connected, the number of times that statistics interaction character occurs in interaction, the number of times occurred in interaction character in interaction
During more than or equal to sensitive number of times, interaction character is fed back into management port；Receive the sensitive content instruction of management port, and root
Interaction character is updated in automatic machine search tree according to sensitive content instruction.
Intelligent Recognition and handle the method for sensitive content by above-described embodiment, in interaction of the invention and be
System, has reached following beneficial effect：
(1) Intelligent Recognition and the method and system of sensitive content are handled in interaction of the present invention, using AC from
Motivation sets up search tree and carries out yellow version lexical search, processing, can comprehensively search for, handle various yellow version vocabulary and its deformation；
The yellow version lexical search service systems of the RPC for calling service can externally be provided by setting up, it is not necessary to build one to each interactive system
Yellow version vocabulary processing module, can also be managed collectively to yellow version vocabulary, improve the treatment effeciency of yellow version vocabulary.
(2) Intelligent Recognition and the method and system of sensitive content are handled in interaction of the present invention, using AC from
Motivation sets up search tree and carries out yellow version lexical search, processing, and sets the sensitive content processing scheme and automatic machine of stagewise to search
Suo Shu and emerging yellow version lexical data more new strategy, it is ensured that the promptness and accuracy of sensitive content processing.
It should be understood by those skilled in the art that, embodiments of the invention can be provided as method, device or computer program
Product.Therefore, the present invention can be using the reality in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Apply the form of example.Moreover, the present invention can be used in one or more computers for wherein including computer usable program code
The computer program production that usable storage medium is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
Although some specific embodiments of the present invention are described in detail by example, the skill of this area
Art personnel are it should be understood that example above is merely to illustrate, the scope being not intended to be limiting of the invention.The skill of this area
Art personnel to above example it should be understood that can modify without departing from the scope and spirit of the present invention.This hair
Bright scope is defined by the following claims.
Claims (8)
1. Intelligent Recognition and the method for handling sensitive content in a kind of interaction, it is characterised in that including：
Default sensitive content is received, the default sensitive content is compiled into automatic machine according to the mechanism of AC automatic machines；Will be described
The endianness order that automatic machine is encoded according to default keyword, and according to the endianness order by the automechanism
Automatic machine search tree is built up, and the automatic machine search tree is stored to telecommunication network request service platform；
The interaction content of client is received, the interaction content is parsed and obtains interacting character, by remote service agreement described
The interactive character is word for word read in automatic machine search tree, traveling through the automatic machine search tree by the interactive character obtains institute
State the sensitive content in interaction content；
The history intersection record of the client is searched for, the number of times that the sensitive content occurs is obtained, it is quick with reference to what is pre-set
Feel contents processing strategy, operation is handled accordingly to the client executing；
Based on the current endianness order, the automatic machine search tree is updated according to the sensitive content；
By the corresponding processing policy store of the automatic machine search tree and the sensitive content after renewal to the telecommunication network
Ask on service platform；
The number of times that the interactive character occurs in interaction is counted, time occurred in the interactive character in interaction
When number is more than or equal to sensitive number of times, the interactive character is fed back into management port；
The sensitive content instruction of the management port is received, and is updated to the interactive character according to sensitive content instruction
In the automatic machine search tree.
2. Intelligent Recognition and the method for handling sensitive content in interaction according to claim 1, it is characterised in that institute
State sensitive content processing strategy, further for：
When the sensitive content occurrence number is less than or equal to setting number of times, the sensitive content is shielded in interactive interface, and to
The client sends alert messages；
When the sensitive content occurrence number is more than setting number of times and is less than or equal to setting frequency threshold value, shield in interactive interface
Cover the sensitive content, and by client's end shield preset time；
When the sensitive content occurrence number is more than the setting frequency threshold value, the sensitive content is shielded in interactive interface, and
Shield the client.
3. Intelligent Recognition and the method for handling sensitive content in interaction according to claim 2, it is characterised in that institute
Setting number of times is stated, is further one to three times；The setting frequency threshold value, is further three to five times.
4. Intelligent Recognition and the method for handling sensitive content in interaction according to claim 1, it is characterised in that will
The endianness order that the automatic machine is encoded according to default keyword, and will be described automatic according to the endianness order
Mechanism builds up automatic machine search tree, further for：
The corresponding extension automatic machine of the automatic machine is obtained according to default automatic machine expanding policy；
The endianness order that the automatic machine and its extension automatic machine are encoded according to default keyword, and according to the word
The automechanism is built up automatic machine search tree by symbol order arrangement.
5. Intelligent Recognition and the system for handling sensitive content in a kind of interaction, it is characterised in that including：Automatic machine search tree
Creation module, sensitive content acquisition module, sensitive content processing module, automatic machine search tree update module and sensitive content processing
Policy store module；Wherein,
The automatic machine search tree creation module, for being connected with the sensitive content acquisition module, receives default sensitive interior
Hold, the default sensitive content is compiled into automatic machine according to the mechanism of AC automatic machines；The automatic machine is closed according to default
The endianness order of key word coding, and the automechanism is built up by automatic machine search tree according to the endianness order,
And store the automatic machine search tree to telecommunication network request service platform；
The sensitive content acquisition module, for being connected with the automatic machine search tree creation module and sensitive content processing module
Connect, receive the interaction content of client, parse the interaction content and obtain interacting character, by remote service agreement it is described from
The interactive character is word for word read in motivation search tree, traveling through the automatic machine search tree by the interactive character obtains described
Sensitive content in interaction content；
The sensitive content processing module, for being connected with the sensitive content acquisition module and automatic machine search tree update module
Connect, search for the history intersection record of the client, the number of times that the sensitive content occurs is obtained, with reference to the sensitivity pre-set
Contents processing strategy, operation is handled to the client executing accordingly；
The automatic machine search tree update module, for handling policy store with the sensitive content processing module and sensitive content
Module is connected, and based on the current endianness order, the automatic machine search tree is updated according to the sensitive content；
The sensitive content handles policy store module, for being connected with the automatic machine search tree update module, will update
The corresponding processing policy store of the automatic machine search tree and the sensitive content afterwards is flat to telecommunication network request service
On platform；
Further comprise：Default sensitive content update module, for being connected with the automatic machine search tree creation module, statistics
The number of times that the interactive character occurs in interaction, the number of times occurred in the interactive character in interaction be more than or
During equal to sensitive number of times, the interactive character is fed back into management port；
The sensitive content instruction of the management port is received, and is updated to the interactive character according to sensitive content instruction
In the automatic machine search tree.
6. Intelligent Recognition and the system for handling sensitive content in interaction according to claim 5, it is characterised in that institute
State sensitive content processing module, further for：Sensitive content occur number of times statistic unit, the first sensitive content processing unit,
Second sensitive content processing unit and the 3rd sensitive content processing unit；Wherein,
The number of times statistic unit that the sensitive content occurs, for the sensitive content acquisition module, automatic machine search tree more
New module and the first sensitive content processing unit are connected, and search for the history intersection record of the client, obtain the sensitivity
The number of times that content occurs；
The first sensitive content processing unit, it is sensitive interior for the number of times statistic unit occurred with the sensitive content and second
Hold processing unit to be connected, when the sensitive content occurrence number is less than or equal to setting number of times, shielded in interactive interface
The sensitive content, and send alert messages to the client；
The second sensitive content processing unit, for being handled with the first sensitive content processing unit and the 3rd sensitive content
Unit is connected, when the sensitive content occurrence number is more than setting number of times and is less than or equal to setting frequency threshold value, in interaction
Shield the sensitive content in interface, and by client's end shield preset time；
The 3rd sensitive content processing unit, it is described sensitive interior for being connected with the second sensitive content processing unit
When holding occurrence number more than the setting frequency threshold value, the sensitive content is shielded in interactive interface, and shield the client.
7. Intelligent Recognition and the system for handling sensitive content in interaction according to claim 6, it is characterised in that institute
Setting number of times is stated, is further one to three times；The setting frequency threshold value, is further three to five times.
8. Intelligent Recognition and the system for handling sensitive content in interaction according to claim 5, it is characterised in that institute
State automatic machine search tree creation module, further for：Automatic machine acquiring unit and automatic machine search tree creating unit；Wherein,
The automatic machine acquiring unit, for being connected with the sensitive content acquisition module and automatic machine search tree creating unit
Connect, the default sensitive content is compiled into automatic machine according to the mechanism of AC automatic machines, according to default automatic machine expanding policy
Obtain the corresponding extension automatic machine of the automatic machine；
The automatic machine search tree creating unit, for being connected with the automatic machine acquiring unit, by the automatic machine and its
The endianness order that extension automatic machine is encoded according to default keyword, and will be described automatic according to the endianness order
Mechanism builds up automatic machine search tree, and the automatic machine search tree is stored to telecommunication network request service platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710334441.9A CN107169092A (en) | 2017-05-12 | 2017-05-12 | Intelligent Recognition and the method and system of sensitive content are handled in interaction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710334441.9A CN107169092A (en) | 2017-05-12 | 2017-05-12 | Intelligent Recognition and the method and system of sensitive content are handled in interaction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107169092A true CN107169092A (en) | 2017-09-15 |
Family
ID=59815958
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710334441.9A Pending CN107169092A (en) | 2017-05-12 | 2017-05-12 | Intelligent Recognition and the method and system of sensitive content are handled in interaction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107169092A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108970123A (en) * | 2018-07-16 | 2018-12-11 | 网易（杭州）网络有限公司 | The sending method of interference information and device, electronic equipment in game |
| CN111177518A (en) * | 2019-12-18 | 2020-05-19 | 深圳市任子行科技开发有限公司 | Webpage purification method, system and computer readable storage medium |
| CN111683049A (en) * | 2020-05-08 | 2020-09-18 | 江苏涵秋网络科技有限公司 | Sensitive content interception system for network security |
| WO2021135103A1 (en) * | 2020-05-29 | 2021-07-08 | 平安科技（深圳）有限公司 | Method and apparatus for semantic analysis, computer device, and storage medium |
| CN114048102A (en) * | 2021-11-18 | 2022-02-15 | 广州银汉科技有限公司 | Chat intelligent analysis monitoring system based on big data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150432A (en) * | 2013-03-07 | 2013-06-12 | 宁波成电泰克电子信息技术发展有限公司 | Method for internet public opinion analysis |
| CN105468684A (en) * | 2015-11-17 | 2016-04-06 | 贵阳朗玛信息技术股份有限公司 | Sensitive word filtering system and communication method thereof |
| CN105893626A (en) * | 2016-05-10 | 2016-08-24 | 中广核工程有限公司 | Index library creation method used for nuclear power engineering and index system adopting index library creation method |
| CN106445998A (en) * | 2016-05-26 | 2017-02-22 | 达而观信息科技（上海）有限公司 | Text content auditing method and system based on sensitive word |
| CN106529294A (en) * | 2016-11-15 | 2017-03-22 | 广东华仝九方科技有限公司 | Method for determining and filtering mobile phone viruses |
-
2017
- 2017-05-12 CN CN201710334441.9A patent/CN107169092A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150432A (en) * | 2013-03-07 | 2013-06-12 | 宁波成电泰克电子信息技术发展有限公司 | Method for internet public opinion analysis |
| CN105468684A (en) * | 2015-11-17 | 2016-04-06 | 贵阳朗玛信息技术股份有限公司 | Sensitive word filtering system and communication method thereof |
| CN105893626A (en) * | 2016-05-10 | 2016-08-24 | 中广核工程有限公司 | Index library creation method used for nuclear power engineering and index system adopting index library creation method |
| CN106445998A (en) * | 2016-05-26 | 2017-02-22 | 达而观信息科技（上海）有限公司 | Text content auditing method and system based on sensitive word |
| CN106529294A (en) * | 2016-11-15 | 2017-03-22 | 广东华仝九方科技有限公司 | Method for determining and filtering mobile phone viruses |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108970123A (en) * | 2018-07-16 | 2018-12-11 | 网易（杭州）网络有限公司 | The sending method of interference information and device, electronic equipment in game |
| CN111177518A (en) * | 2019-12-18 | 2020-05-19 | 深圳市任子行科技开发有限公司 | Webpage purification method, system and computer readable storage medium |
| CN111683049A (en) * | 2020-05-08 | 2020-09-18 | 江苏涵秋网络科技有限公司 | Sensitive content interception system for network security |
| WO2021135103A1 (en) * | 2020-05-29 | 2021-07-08 | 平安科技（深圳）有限公司 | Method and apparatus for semantic analysis, computer device, and storage medium |
| CN114048102A (en) * | 2021-11-18 | 2022-02-15 | 广州银汉科技有限公司 | Chat intelligent analysis monitoring system based on big data |
| CN114048102B (en) * | 2021-11-18 | 2022-07-22 | 广州银汉科技有限公司 | Chat intelligent analysis monitoring system based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107169092A (en) | Intelligent Recognition and the method and system of sensitive content are handled in interaction | |
| AU2019200437B2 (en) | A method to build an enterprise-specific knowledge graph | |
| US10169471B2 (en) | Generating and executing query language statements from natural language | |
| CA3087534C (en) | System and method for information extraction with character level features | |
| CN104412265B (en) | Update for promoting the search of application searches to index | |
| US10951492B2 (en) | System and a method for automatic conversion of monolithic services to micro-services | |
| WO2020108063A1 (en) | Feature word determining method, apparatus, and server | |
| CN112262390A (en) | Regular expression generation based on positive and negative pattern matching examples | |
| CN106528613B (en) | Intelligent answer method and device | |
| CN102647414B (en) | Protocol analysis method, protocol analysis device and protocol analysis system | |
| CN102567455A (en) | Method and system of managing documents using weighted prevalence data for statements | |
| CN109697231A (en) | A kind of display methods, system, storage medium and the processor of case document | |
| CN113282762A (en) | Knowledge graph construction method and device, electronic equipment and storage medium | |
| CN109753517A (en) | A kind of method, apparatus, computer storage medium and the terminal of information inquiry | |
| CN108304363A (en) | Public sentiment public relations method and system | |
| CN109933331A (en) | Data transfer device and associated component between a kind of client-server | |
| CN108388547A (en) | Character string parsing method, apparatus, equipment and computer readable storage medium | |
| US9898467B1 (en) | System for data normalization | |
| CN107273546A (en) | Counterfeit application detection method and system | |
| CN110489740A (en) | Semantic analytic method and Related product | |
| CN109683727A (en) | A kind of data processing method and device | |
| US8090750B2 (en) | Prompting of an end user with commands | |
| CN113515630A (en) | Triple generating and checking method and device, electronic equipment and storage medium | |
| CN113742529A (en) | Multi-table front-end processing method and device | |
| CN115774797A (en) | Video content retrieval method, device, equipment and computer readable 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170915 |