CN108763359A - A kind of usage mining method, apparatus and electronic equipment with incidence relation - Google Patents
A kind of usage mining method, apparatus and electronic equipment with incidence relation Download PDFInfo
- Publication number
- CN108763359A CN108763359A CN201810469344.5A CN201810469344A CN108763359A CN 108763359 A CN108763359 A CN 108763359A CN 201810469344 A CN201810469344 A CN 201810469344A CN 108763359 A CN108763359 A CN 108763359A
- Authority
- CN
- China
- Prior art keywords
- user
- subordinate
- neighbor
- behavior
- probability
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a kind of usage mining method, apparatus and electronic equipment with incidence relation, the method includes：The user for carrying out specific internet behavior based on same equipment and/or identical internet protocol address in set period of time is determined as the user with neighborhood；The user with subordinate relation in the user with neighborhood is determined based on iterative algorithm；According to the determination of the user with subordinate relation as a result, determining all users with subordinate relation.By using above-mentioned technical proposal, the purpose for excavating the user community with incidence relation may be implemented.
Description
Technical field
The present embodiments relate to computer realm more particularly to a kind of usage mining method with incidence relation, dresses
It sets and electronic equipment.
Background technology
In webcast website, in the prevalence of the cheating of the brushes popularities such as some brush barrages, brush concern.Due to huge profit
The presence of benefit, these cheatings have developed into complete Dark Industry Link item.
Cheating based on platform (such as webcast website) has clique's property mostly, and above-mentioned cheating can also make
At network blockage, the problems such as Platform Server pressure is excessive is broadcast live, therefore in order to reduce the negative shadow that above-mentioned cheating is brought
It rings, being found using rational method has the clique of cheating suspicion significant.
Invention content
The present invention provides a kind of usage mining method, apparatus and electronic equipment with incidence relation, passes through the method
The user community with incidence relation can be excavated.
To achieve the above object, the embodiment of the present invention adopts the following technical scheme that：
In a first aspect, an embodiment of the present invention provides a kind of usage mining method with incidence relation, the method packet
It includes：
Same equipment and/or identical IP (Internet Protocol, Internet protocol) will be based in set period of time
The user that address carries out specific internet behavior is determined as the user with neighborhood；
The user with subordinate relation in the user with neighborhood is determined based on iterative algorithm；
According to the determination of the user with subordinate relation as a result, determining all users with fellowship.
Further, described that the use with subordinate relation in the user with neighborhood is determined based on iterative algorithm
Family, including：
The label for identifying corresponding user is determined for each user；
The subordinate probability between each user and its neighbor user is calculated based on iterative algorithm using the label of each user,
And the subordinate probability between each user and its neighbor user stops iteration when remaining unchanged；
According in user of the subordinate determine the probability with neighborhood between each user and its neighbor user with from
The user of category relationship.
Further, the label using each user be based on iterative algorithm calculate each user and its neighbor user it
Between subordinate probability, including：
The subordinate Probability p between user x and its neighbor user c at the kth iteration is calculated according to following formulak(c,x)：
Wherein, N (x) indicates the set of the neighbor user composition of user x, | N (x) | indicate the number of the neighbor user of user x
Amount, y indicate the neighbor user of user x, pk-1(c, y) indicates that in -1 iteration of kth, the subordinate between user y and user c is general
Rate.
Further, the label using each user be based on iterative algorithm calculate each user and its neighbor user it
Between subordinate probability, further include：
When each iteration, if the subordinate Probability p between user x and its neighbor user ck(c, x) is all higher than equal to reservation threshold
Value then retains when the subordinate Probability p in previous iterationk(c,x)；
If only part pk(c, x), which is more than or equal to, retains threshold value, then the subordinate between user x and its neighbor user c is general
Rate pk(c, x) is normalized；
If all pk(c, x) both less than retains threshold value, then selects user c at random from N (x)0And user x and its are set
Neighbor user c0Between subordinate Probability pk(c0, x) and it is 1, and by pk(c0, x) as user x work as previous iteration result.
Further, in kth time iteration, using following normalization formula between user x and its neighbor user c from
Belong to probability to be normalized：
Wherein, pk' (c, x) indicates the subordinate between user x and its neighbor user c after being normalized in kth time iteration
Probability, N (x) indicate the set of the neighbor user composition of user x, pkUser before (c, x) indicates to normalize at the kth iteration
Subordinate probability between x and its neighbor user c, I (pk(c, x) >=1/v) it is indicative function, if pk(c, x) >=1/v, then I (pk
(c, x) >=1/v) 1 is taken, otherwise take 0,1/v to indicate to retain threshold value.
Further, the determination of the user with subordinate relation described in the basis is as a result, determine with fellowship
All users, including：
With same subscriber will there is the user of subordinate relation to be determined as the user with fellowship.
Further, the specific internet behavior, including：By the way that platform is broadcast live brush barrage row is carried out for identical main broadcaster
For and/or by live streaming platform brush concern behavior is carried out for identical main broadcaster.
Further, in the determination of the user with subordinate relation according to as a result, determining the institute with fellowship
After having user, further include：
If any user in all users with fellowship is brushed by the way that platform is broadcast live for identical main broadcaster
The number of barrage behavior and/or brush concern behavior is more than first threshold, it is determined that the user is the high individual of cheating suspicion grade；
If with fellowship all users by be broadcast live platform for identical main broadcaster carry out the behavior of brush barrage and/or
The total degree of brush concern behavior is more than second threshold, it is determined that all users are the high clique of cheating suspicion grade.
Further, it is described will in set period of time based on same equipment and/or identical internet protocol address into
Before the user of the specific internet behavior of row is determined as having the user of neighborhood, further include：
Behavior-based control gets acquisition User action log ready, to determine the user for carrying out specific internet behavior；
The network environment information that user uses is obtained for the user for carrying out specific internet behavior, with the determination use
The IP address at family；And/or
The terminal device information that user uses is obtained for the user for carrying out specific internet behavior, with the determination use
The device id that family uses.
Second aspect, an embodiment of the present invention provides a kind of usage mining device with incidence relation, described device packets
It includes：
Neighbor user determining module, for same equipment and/or identical Internet protocol will to be based in set period of time
The user that IP address carries out specific internet behavior is determined as the user with neighborhood；
First user's determining module, for being determined in the user with neighborhood with subordinate based on iterative algorithm
The user of relationship；
Second user determining module, the determination for the user with subordinate relation according to is as a result, determine with group
All users of body relationship.
The third aspect an embodiment of the present invention provides a kind of electronic equipment, including first memory, first processor and is deposited
The computer program that can be run on a memory and on first processor is stored up, the first processor executes the computer journey
The usage mining method with incidence relation as described in above-mentioned first aspect is realized when sequence.
Fourth aspect, an embodiment of the present invention provides a kind of storage medium including computer executable instructions, the meters
Calculation machine executable instruction realizes the use with incidence relation as described in above-mentioned first aspect when being executed by computer processor
Family method for digging.
A kind of usage mining method with incidence relation provided in an embodiment of the present invention, by will be in set period of time
The user that specific internet behavior is carried out based on same equipment and/or identical internet protocol address is determined as with neighborhood
User, and the user with subordinate relation in the user with neighborhood, and then basis are determined based on iterative algorithm
The incidence relation determines the technological means of all users with incidence relation, realizes and excavates the user with incidence relation
The purpose of group.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, institute in being described below to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example without creative efforts, can also be implemented for those of ordinary skill in the art according to the present invention
The content of example and these attached drawings obtain other attached drawings.
Fig. 1 is a kind of usage mining method flow schematic diagram with incidence relation that the embodiment of the present invention one provides；
Fig. 2 is a kind of usage mining method flow schematic diagram with incidence relation provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of customer relationship network diagram provided by Embodiment 2 of the present invention；
Fig. 4 is a kind of usage mining apparatus structure schematic diagram with incidence relation that the embodiment of the present invention three provides；
Fig. 5 is the structural schematic diagram for a kind of electronic equipment that the embodiment of the present invention four provides.
Specific implementation mode
For make present invention solves the technical problem that, the technical solution that uses and the technique effect that reaches it is clearer, below
The technical solution of the embodiment of the present invention will be described in further detail in conjunction with attached drawing, it is clear that described embodiment is only
It is a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those skilled in the art exist
The every other embodiment obtained under the premise of creative work is not made, shall fall within the protection scope of the present invention.
Embodiment one
Fig. 1 is a kind of usage mining method flow schematic diagram with incidence relation that the embodiment of the present invention one provides.This
The usage mining method with incidence relation can be executed by the usage mining device with incidence relation disclosed in embodiment,
Wherein the device can be by software and or hardware realization, and is typically integrated in terminal, such as smart mobile phone or computer etc..Tool
Body is shown in Figure 1, and this method may include steps of：
110, the user of specific internet behavior will be carried out based on same equipment and/or identical IP address in set period of time
It is determined as the user with neighborhood.
Wherein, the specific internet behavior can be the behavior of positive worth promotion, such as online contribution, can also be
The behavior that passive needs resist, such as the behavior of brush barrage is carried out for identical main broadcaster by live streaming platform or passes through live streaming
Platform carries out brush concern behavior for identical main broadcaster.The behavior that usually passive needs resist often brings some negative shadows
It rings, carries out the behavior of brush barrage or by live streaming platform for identical for identical main broadcaster by the way that platform is broadcast live as described above
Main broadcaster carry out brush concern behavior, it will usually cause network blockage, live streaming Platform Server pressure it is excessive the problems such as.Therefore in order to
It reduces brush barrage behavior and either negative effect caused by brush concern behavior or is engaged in certain beneficial to behavior to actively advocate, this
Embodiment discloses a kind of usage mining method with incidence relation, that is, excavates and be engaged in barrage behavior or brush concern behavior
Practise fraud clique, with alerted either take other measures rectified or excavated be engaged in contribution etc. public goods behavior group
It knits, to commend, builds healthy tendency in society etc..
Specifically, the present embodiment is for excavating and belong to all users of the same brush barrage or brush concern tissue.Institute
Set period of time is stated for example and can be 9. -10 point of morning May 3, this period is typically certain main broadcaster in live streaming platform live streaming
Save object time.In order to monitor whether there is the behavior for carrying out brush barrage or brush concern for the main broadcaster, when can be by setting
Between section be set as period of main broadcaster's programme televised live.
For example, between 9. -10 point of morning May 3, user a, b, c, d have sent barrage letter by device A to main broadcaster B
Breath, it is determined that user a, b, c, d are the user with neighborhood；User c and user f passes through equipment in this period simultaneously
C has sent barrage information to main broadcaster B, it is determined that user c and user f is the user with neighborhood；In this period simultaneously
User d and user e has sent barrage information by equipment D to main broadcaster B, it is determined that user d and user e is with neighborhood
User.
Further, it is described will in set period of time based on same equipment and/or identical internet protocol address into
Before the user of the specific internet behavior of row is determined as having the user of neighborhood, further include：
Behavior-based control gets acquisition User action log ready, to determine the user for carrying out specific internet behavior；
The network environment information that user uses is obtained for the user for carrying out specific internet behavior, with the determination use
The IP address at family；
And/or the terminal device information that user uses is obtained for the user for carrying out specific internet behavior, with determination
The device id that the user uses.
It is that event, the page (are such as clicked in the place for needing to bury in engineering a little for counting user behavior that the behavior, which is got ready,
Redirect) it is inserted into and buries point code, the internet behavior of user will be recorded in User action log later, by acquiring user behavior
Daily record simultaneously inquires user behavior and can determine the user for carrying out specific internet behavior, described specific to surf the net as which is for example specially
A little users have sent barrage information for main broadcaster A.User is also recorded in User action log simultaneously and carries out internet behavior institute
The network environment information and used terminal device information used.The User action log is in mobile terminal (such as intelligent hand
Machine) it can directly be obtained by a data acquisition interface.
120, the user with subordinate relation in the user with neighborhood is determined based on iterative algorithm.
Wherein, the subordinate relation refers specifically to the special relationship for having between the user of neighborhood, such as described special
Relationship is to belong to the same tissue for exclusively carrying out brush barrage behavior or partner.And then according between the user with neighborhood
Subordinate relation excavate the incidence relation in the user with neighborhood between the user of non-neighborhood, that is, excavate
Go out which user in the user of non-neighborhood belongs to the same tissue for exclusively carrying out brush barrage behavior or partner.
130, according to the determination of the user with subordinate relation as a result, determining all users with fellowship.
For example, if the neighbor user of user a only has user b, i.e. there are absolute subordinate relation between user a and user b, use
There is also subordinate relation between family b and its neighbor user c, and thus can determine has subordinate relation between user a and user c, into
And can determine has fellowship between user a, b, c.
In the present embodiment, the fellowship can specifically refer to belong to it is same exclusively carry out the behavior of brush barrage or brush close
It is tissue or the partner of behavior.
A kind of usage mining method with incidence relation provided in this embodiment, by will be based in set period of time
Same equipment and/or identical IP address carries out the behavior of brush barrage for identical main broadcaster or the user of brush concern behavior is determined as
User with neighborhood, and determine which user belongs to same in the user with neighborhood based on iterative algorithm
It is a to exclusively carry out the tissue of the behavior of brush barrage or brush concern behavior, and then excavate and all belong to same and exclusively carry out brush bullet
The user of curtain behavior or brush concern behavior tissue, realizes and excavates the user community with incidence relation.
Embodiment two
Fig. 2 is a kind of usage mining method flow schematic diagram with incidence relation provided by Embodiment 2 of the present invention.?
On the basis of above-described embodiment, the present embodiment " determines the use with neighborhood to above-mentioned steps 120 based on iterative algorithm
User with subordinate relation in family " is embodied, and gives specific iterative algorithm, and tool is presented by way of example
The iterative process of body.Shown in Fig. 2, this method may include steps of：
210, the user of specific internet behavior will be carried out based on same equipment and/or identical IP address in set period of time
It is determined as the user with neighborhood.
220, the label for identifying corresponding user is determined for each user.
Determine that the purpose of the label for identifying corresponding user is in order to facilitate subsequent iterative algorithm for each user.Often
The initial labels of a user are unique.
230, the subordinate between each user and its neighbor user is calculated based on iterative algorithm using the label of each user
Probability, and when subordinate probability between each user and its neighbor user remains unchanged stop iteration.
Specifically, calculating the subordinate probability between user x and its neighbor user c at the kth iteration according to following formula
pk(c,x)：
Wherein, N (x) indicates that the set of the neighbor user composition of user x, N (x) indicate the quantity of the neighbor user of user x,
Y indicates the neighbor user of user x, pk-1(c, y) is indicated in -1 iteration of kth, the subordinate probability between user y and user c, k
Initial value be 2.
Illustrate above-mentioned iterative algorithm：
For example, it is b, c, d to have the user of neighborhood with user a；It is f to have the user of neighborhood with user c；With
It is e that user d, which has the user of neighborhood,；Described a, b, c, d, e, f are the label of the corresponding user of mark.Above-mentioned user a, b,
C, d, e, f are the users that identical internet behavior was carried out in set period of time, according between above-mentioned user a, b, c, d, e, f
Neighborhood, build customer relationship network, customer relationship network diagram shown in Figure 3, each two neighbor user
Between draw a straight line.Be to determine due to the neighbor user of each user, in first time iteration, user x and its
Subordinate probability between neighbor user c：
By taking the neighborhood between above-mentioned user a, b, c, d, e, f as an example, when first time iteration：
Subordinate probability between user a and its neighbor user b, c, d is respectively 1/3,1/3,1/3, is denoted as
Subordinate probability between user b and its neighbor user a is：p1(a, b)=1；
Subordinate probability between user c and its neighbor user a, f is respectively：
Subordinate probability between user d and its neighbor user a, e is respectively：
Subordinate probability between user e and its neighbor user d is：p1(d, e)=1；
Subordinate probability between user f and its neighbor user c is：p1(c, f)=1；
When second of iteration, i.e. k=2 applies mechanically above-mentioned iterative formulaAccording to
The secondary subordinate probability obtained between neighbor user：
Subordinate probability between user a and its neighbor user b, c, d is respectively：
Subordinate probability between user b and its neighbor user a is：p2(a, b)=1；
Subordinate probability between user c and its neighbor user a, f is respectively：
Subordinate probability between user d and its neighbor user a, e is respectively：
Subordinate probability between user e and its neighbor user d is：p2(d, e)=1；
Subordinate probability between user f and its neighbor user c is：p2(c, f)=1.
It can be seen that the subordinate probability between each user and its neighbor user remains unchanged, stop iteration.
In Practical Project realization, it is engaged in identical internet behavior if it is multiple user collectives, each user has more
A neighbor user, in order to excavate with the related non-neighbor user of each user, for user x, in each iteration not
Candidate of the subordinate probability between user x and its all neighbor user as next iteration when can be retained.Therefore, usually by setting
The mode for retaining threshold value is set, the neighbor user of user x is screened, is specified or normalized.The reservation threshold value can
Rationally it is arranged with combining iteration speed, precision and calculating cost.
Further, the label using each user be based on iterative algorithm calculate each user and its neighbor user it
Between subordinate probability, further include：
When each iteration, if the subordinate Probability p between user x and its neighbor user ck(c, x) is all higher than equal to reservation threshold
Value then retains when the subordinate Probability p in previous iterationk(c,x)；
If only part pk(c, x), which is more than or equal to, retains threshold value, then the subordinate between user x and its neighbor user c is general
Rate pk(c, x) is normalized；
If all pk(c, x) both less than retains threshold value, then selects user c at random from N (x)0And user x and its are set
Neighbor user c0Between subordinate Probability pk(c0, x) and it is 1, and by pk(c0, x) as user x work as previous iteration result.
Specifically, in kth time iteration, subordinate of the following normalization formula between user x and its neighbor user c is used
Probability is normalized：
Wherein, pk' (c, x) indicates the subordinate between user x and its neighbor user c after normalizing at the kth iteration
Probability, N (x) indicate the set of the neighbor user composition of user x, pkUser before (c, x) indicates to normalize at the kth iteration
Subordinate probability between x and its neighbor user c, I (pk(c, x) >=1/v) it is indicative function, if pk(c, x) >=1/v, then I (pk
(c, x) >=1/v) 1 is taken, otherwise take 0,1/v to indicate to retain threshold value.
Continue by taking the neighborhood between above-mentioned user a, b, c, d, e, f as an example, setting retain threshold value be 1/2, due to
Subordinate probability in first time iteration between user a and its neighbor user b, c, d is respectively 1/3,1/3,1/3, is respectively less than retained
Therefore threshold value is randomly assigned neighbor user b, c, d of user a, such as from the neighbor user set N (x) of user a
The random subordinate probability selected user c and be arranged between user x and its neighbor user c is 1, and by p1(c a)=1 is used as user
The neighbor user of the iteration result of a first times, user a only has user c；
Due to the subordinate Probability p between user b and its neighbor user a1(a, b)=1 is more than and retains threshold value 1/2, therefore protect
Stay p1(a, b)=1,
Same principle, the subordinate probability between user c and its neighbor user a, f are respectively：
Subordinate probability between user d and its neighbor user a, e is respectively：
Subordinate probability between user e and its neighbor user d is：p1(d, e)=1；
Subordinate probability between user f and its neighbor user c is：p1(c, f)=1；
In second of iteration, continue to apply mechanically above-mentioned iterative formula：
Subordinate probability between user a and its neighbor user c is：p2(c, a)=1
Subordinate probability between user b and its neighbor user a is：p2(c, b)=1, because the neighbor user of user a only has
User c, and the neighbor user of user b only has user a, therefore, between user b and user c is also neighbor user and subordinate is general
Rate is 1；
Subordinate probability between user c and its neighbor user a, f is respectively：It needs to change
Become, since the neighbor user of user a only has user c, the place of user a is directly replaced with into user c.
Same principle, the subordinate probability between user d and its neighbor user a, e are respectively：
Subordinate probability between user e and its neighbor user d is：p2(d, e)=1；
Subordinate probability between user f and its neighbor user c is：p2(c, f)=1.
As it can be seen that the subordinate probability between each user and its neighbor user is all higher than equal to threshold value 1/2, thus each user with
Subordinate probability between its neighbor user will remain unchanged, identical as the result of first time iteration, iteration stopping.
Have in the user 240, according to the subordinate determine the probability between each user and its neighbor user with neighborhood
There is the user of subordinate relation.
There are subordinate probability between user a and user c it can be seen from above-mentioned iteration result, therefore user a and user c
Between have subordinate relation；There are subordinate probability between user b and user c, therefore between user b and user c there is subordinate to close
System；There are subordinate probability between user c and user a, f, therefore have subordinate relation between user c and user a and user f；
There are subordinate probability between user d and user c, e, therefore have subordinate relation between user d and user c and user e；User
There are subordinate probability between e and user d, therefore have subordinate relation between user e and user d；Have between user f and user c
There is subordinate relation.
250, according to the determination of the user with subordinate relation as a result, determining all users with fellowship.
Further, the determination of the user with subordinate relation described in the basis is as a result, determine with fellowship
All users, including：
With same subscriber will there is the user of subordinate relation to be determined as the user with fellowship.
User a, b, c, d, f have subordinate relation with user c it can be seen from above-mentioned iteration result, therefore can be true
It may be a group to determine user a, b, c, d, f.
Further, in the determination of the user with subordinate relation according to as a result, determining the institute with fellowship
After having user, further include：
If any user in all users (such as user a, b, c, d, f) with fellowship is by being broadcast live platform needle
The number that the behavior of brush barrage and/or brush concern behavior are carried out to identical main broadcaster is more than first threshold, it is determined that the user is to make
The high individual of disadvantage suspicion grade；
If with fellowship all users by be broadcast live platform for identical main broadcaster carry out the behavior of brush barrage and/or
The total degree of brush concern behavior is more than second threshold, it is determined that all users are the high clique of cheating suspicion grade.
A kind of usage mining method with incidence relation provided in this embodiment, by the way that reservation is arranged in an iterative process
Threshold value is screened with multiple neighbor users to user x, is specified or normalized, has multiple neighbours for user x
When user, the quick excavation with incidence relation user is realized.
Embodiment three
Fig. 4 is a kind of usage mining apparatus structure schematic diagram with incidence relation that the embodiment of the present invention three provides.Ginseng
As shown in Figure 4, described device includes：Neighbor user determining module 410, first user's determining module 420 and second user determine
Module 430；
Wherein, neighbor user determining module 410, for will in set period of time based on same equipment and/or it is identical mutually
The user that networking protocol IP address carries out specific internet behavior is determined as the user with neighborhood；
First user's determining module 420, for determining in the user with neighborhood have based on iterative algorithm
The user of subordinate relation；
Second user determining module 430, the determination for the user with subordinate relation according to is as a result, determination has
All users of fellowship.
Usage mining device provided in this embodiment with incidence relation, it is same by that will be based in set period of time
Equipment and/or identical IP address carries out the behavior of brush barrage for identical main broadcaster or the user of brush concern behavior is determined as having
The user of neighborhood, and determine which user belongs to same special in the user with neighborhood based on iterative algorithm
Door carries out the tissue of the behavior of brush barrage or brush concern behavior, and then excavates and all belong to same and exclusively carry out brush barrage row
For or brush concern behavior tissue user, realize and excavate the user community with incidence relation.
Embodiment three
Fig. 5 is the structural schematic diagram for a kind of electronic equipment that the embodiment of the present invention three provides.As shown in figure 5, the electronics is set
It is standby to include：It first processor 670, first memory 671 and is stored on first memory 671 and can be in first processor 670
The computer program of upper operation；Wherein, the quantity of first processor 670 can be one or more, at one first in Fig. 5
For reason device 670；First processor 670 realizes that having as described in above-described embodiment one is closed when executing the computer program
The usage mining method of connection relationship.As shown in figure 5, the electronic equipment can also include that the first input unit 672 and first is defeated
Go out device 673.First processor 670, first memory 671, the first input unit 672 and the first output device 673 can lead to
It crosses bus or other modes connects, in Fig. 5 for being connected by bus.
First memory 671 is used as a kind of computer readable storage medium, can be used for storing software program, computer can be held
Line program and module, if usage mining device/module with incidence relation in the embodiment of the present invention is (for example, with association
Neighbor user determining module 410 and first user's determining module 420 in the usage mining device of relationship etc.).First processor
670 are stored in software program, instruction and module in first memory 671 by operation, to execute each of electronic equipment
Kind application of function and data processing, that is, realize the above-mentioned usage mining method with incidence relation.
First memory 671 can include mainly storing program area and storage data field, wherein storing program area can store behaviour
Make the application program needed for system, at least one function；Storage data field can be stored uses created data according to terminal
Deng.In addition, first memory 671 may include high-speed random access memory, can also include nonvolatile memory, such as
At least one disk memory, flush memory device or other non-volatile solid state memory parts.In some instances, it first deposits
Reservoir 671 can further comprise that the memory remotely located relative to first processor 670, these remote memories can pass through
Network connection is to electronic equipment/storage medium.The example of above-mentioned network includes but not limited to internet, intranet, local
Net, mobile radio communication and combinations thereof.
First input unit 672 can be used for receiving the number or character information of input, and generate the use with electronic equipment
Family is arranged and the related key signals input of function control.First output device 673 may include that display screen etc. shows equipment.
Embodiment five
The embodiment of the present invention five also provides a kind of storage medium including computer executable instructions, and the computer can be held
Row instruction by computer processor when being executed for executing a kind of usage mining method with incidence relation, this method packet
It includes：
Specific internet behavior will be carried out based on same equipment and/or identical internet protocol address in set period of time
User be determined as the user with neighborhood；
The user with subordinate relation in the user with neighborhood is determined based on iterative algorithm；
According to the determination of the user with subordinate relation as a result, determining all users with fellowship.
Certainly, a kind of storage medium including computer executable instructions that the embodiment of the present invention is provided, computer
The operation of method that executable instruction is not limited to the described above, can also be performed that any embodiment of the present invention provided has association
The usage mining relevant operation of relationship.
By the description above with respect to embodiment, it is apparent to those skilled in the art that, the present invention
It can be realized by software and required common hardware, naturally it is also possible to which by hardware realization, but the former is more in many cases
Good embodiment.Based on this understanding, technical scheme of the present invention substantially in other words contributes to the prior art
Part can be expressed in the form of software products, which can be stored in computer readable storage medium
In, such as the floppy disk of computer, read-only memory (Read-Only Memory, ROM), random access memory (Random
Access Memory, RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are with so that a computer is set
Standby (can be personal computer, storage medium or the network equipment etc.) executes described in each embodiment of the present invention.
Note that above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The present invention is not limited to specific embodiments described here, can carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out to the present invention by above example
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
May include other more equivalent embodiments, and the scope of the present invention is determined by scope of the appended claims.
Claims (12)
1. a kind of usage mining method with incidence relation, which is characterized in that including：
The use of specific internet behavior will be carried out based on same equipment and/or identical internet protocol address in set period of time
Family is determined as the user with neighborhood；
The user with subordinate relation in the user with neighborhood is determined based on iterative algorithm；
According to the determination of the user with subordinate relation as a result, determining all users with fellowship.
2. according to the method described in claim 1, it is characterized in that, described have neighborhood based on iterative algorithm determination is described
User in subordinate relation user, including：
The label for identifying corresponding user is determined for each user；
The subordinate probability between each user and its neighbor user is calculated based on iterative algorithm using the label of each user, and
Stop iteration when each the subordinate probability between user and its neighbor user remains unchanged；
It is closed with subordinate according in user of the subordinate determine the probability with neighborhood between each user and its neighbor user
The user of system.
3. according to the method described in claim 2, it is characterized in that, the label using each user is based on iterative algorithm meter
The subordinate probability between each user and its neighbor user is calculated, including：
The subordinate Probability p between user x and its neighbor user c at the kth iteration is calculated according to following formulak(c,x)：
Wherein, N (x) indicates the set of the neighbor user composition of user x, | N (x) | indicate the quantity of the neighbor user of user x, y
Indicate the neighbor user of user x, pk-1(c, y) is indicated in -1 iteration of kth, the subordinate probability between user y and user c.
4. according to the method described in claim 3, it is characterized in that, the label using each user is based on iterative algorithm meter
The subordinate probability between each user and its neighbor user is calculated, further includes：
When each iteration, if the subordinate Probability p between user x and its neighbor user ck(c, x) is all higher than equal to threshold value is retained, then
Retain when the subordinate Probability p in previous iterationk(c,x)；
If only part pk(c, x), which is more than or equal to, retains threshold value, then the subordinate Probability p between user x and its neighbor user ck
(c, x) is normalized；
If all pk(c, x) both less than retains threshold value, then selects user c at random from N (x)0And user x and its neighbour are set
User c0Between subordinate Probability pk(c0, x) and it is 1, and by pk(c0, x) as user x work as previous iteration result.
5. according to the method described in claim 4, it is characterized in that, in kth time iteration, using following normalization formula to
Subordinate probability between family x and its neighbor user c is normalized：
Wherein, pk' (c, x) indicates the subordinate probability between user x and its neighbor user c, N after being normalized in kth time iteration
(x) set of the neighbor user composition of user x, p are indicatedk(c, x) indicate at the kth iteration normalize before user x and its
Subordinate probability between neighbor user c, I (pk(c, x) >=1/v) it is indicative function, if pk(c, x) >=1/v, then I (pk(c,x)≥
1 1/v) is taken, 0,1/v is otherwise taken to indicate to retain threshold value.
6. according to the method described in any one of claim 2-5, which is characterized in that with subordinate relation described in the basis
The determination of user as a result, determine all users with fellowship, including：
With same subscriber will there is the user of subordinate relation to be determined as the user with fellowship.
7. method according to any one of claims 1-5, which is characterized in that the specific internet behavior, including：Pass through
Live streaming platform carries out the behavior of brush barrage for identical main broadcaster and/or carries out brush concern for identical main broadcaster by the way that platform is broadcast live
Behavior.
8. the method according to the description of claim 7 is characterized in that determining knot according to the user with subordinate relation
Fruit, determine with fellowship all users after, further include：
If any user in all users with fellowship carries out brush barrage by the way that platform is broadcast live for identical main broadcaster
The number of behavior and/or brush concern behavior is more than first threshold, it is determined that the user is the high individual of cheating suspicion grade；
If all users with fellowship are by being broadcast live, platform carries out the behavior of brush barrage for identical main broadcaster and/or brush closes
The total degree of note behavior is more than second threshold, it is determined that all users are the high clique of cheating suspicion grade.
9. method according to any one of claims 1-5, which is characterized in that described to be based on together in set period of time
The user that one equipment and/or identical internet protocol address carry out specific internet behavior is determined as the user with neighborhood
Before, further include：
Behavior-based control gets acquisition User action log ready, to determine the user for carrying out specific internet behavior；
The network environment information that user uses is obtained for the user for carrying out specific internet behavior, with the determination user's
IP address；And/or
The terminal device information that user uses is obtained for the user for carrying out specific internet behavior, the user makes with determination
Device id.
10. a kind of usage mining device with incidence relation, which is characterized in that described device includes：
Neighbor user determining module, being used for will be in set period of time based on same equipment and/or identical Internet protocol IP
The user that location carries out specific internet behavior is determined as the user with neighborhood；
First user's determining module, for being determined in the user with neighborhood with subordinate relation based on iterative algorithm
User；
Second user determining module, for being closed with subordinate according to the determination of the user with subordinate relation as a result, determining
All users of system.
11. a kind of electronic equipment, including first memory, first processor and storage are on a memory and can be in first processor
The computer program of upper operation, which is characterized in that realized when the first processor executes the computer program as right is wanted
Seek the usage mining method with incidence relation described in any one of 1-9.
12. a kind of storage medium including computer executable instructions, the computer executable instructions are by computer disposal
The usage mining method with incidence relation as claimed in any one of claims 1-9 wherein is realized when device executes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810469344.5A CN108763359A (en) | 2018-05-16 | 2018-05-16 | A kind of usage mining method, apparatus and electronic equipment with incidence relation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810469344.5A CN108763359A (en) | 2018-05-16 | 2018-05-16 | A kind of usage mining method, apparatus and electronic equipment with incidence relation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108763359A true CN108763359A (en) | 2018-11-06 |
Family
ID=64007950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810469344.5A Pending CN108763359A (en) | 2018-05-16 | 2018-05-16 | A kind of usage mining method, apparatus and electronic equipment with incidence relation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108763359A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522342A (en) * | 2018-11-30 | 2019-03-26 | 北京百度网讯科技有限公司 | Police affairs management method, device, equipment and storage medium |
| CN109714636A (en) * | 2018-12-21 | 2019-05-03 | 武汉瓯越网视有限公司 | A kind of user identification method, device, equipment and medium |
| CN110428001A (en) * | 2019-07-31 | 2019-11-08 | 腾讯科技（深圳）有限公司 | A kind of community mining method, apparatus, server and storage medium |
| CN110502697A (en) * | 2019-08-26 | 2019-11-26 | 武汉斗鱼网络科技有限公司 | A kind of target user's recognition methods, device and electronic equipment |
| CN111010584A (en) * | 2019-12-05 | 2020-04-14 | 网易（杭州）网络有限公司 | Live broadcast data processing method and device, server and storage medium |
| CN112800286A (en) * | 2021-04-08 | 2021-05-14 | 北京轻松筹信息技术有限公司 | User relationship chain construction method and device and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103327075A (en) * | 2013-05-27 | 2013-09-25 | 电子科技大学 | Distributed mass organization realizing method based on label interaction |
| CN103729475A (en) * | 2014-01-24 | 2014-04-16 | 福州大学 | Multi-label propagation discovery method of overlapping communities in social network |
| CN106485559A (en) * | 2015-08-19 | 2017-03-08 | 阿里巴巴集团控股有限公司 | Cheating recognition methods and device for on-line shop |
| CN106708844A (en) * | 2015-11-12 | 2017-05-24 | 阿里巴巴集团控股有限公司 | User group partitioning method and device |
| CN106951524A (en) * | 2017-03-21 | 2017-07-14 | 哈尔滨工程大学 | Overlapping community discovery method based on node influence power |
| CN107093090A (en) * | 2016-10-25 | 2017-08-25 | 北京小度信息科技有限公司 | Abnormal user recognition methods and device |
-
2018
- 2018-05-16 CN CN201810469344.5A patent/CN108763359A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103327075A (en) * | 2013-05-27 | 2013-09-25 | 电子科技大学 | Distributed mass organization realizing method based on label interaction |
| CN103729475A (en) * | 2014-01-24 | 2014-04-16 | 福州大学 | Multi-label propagation discovery method of overlapping communities in social network |
| CN106485559A (en) * | 2015-08-19 | 2017-03-08 | 阿里巴巴集团控股有限公司 | Cheating recognition methods and device for on-line shop |
| CN106708844A (en) * | 2015-11-12 | 2017-05-24 | 阿里巴巴集团控股有限公司 | User group partitioning method and device |
| CN107093090A (en) * | 2016-10-25 | 2017-08-25 | 北京小度信息科技有限公司 | Abnormal user recognition methods and device |
| CN106951524A (en) * | 2017-03-21 | 2017-07-14 | 哈尔滨工程大学 | Overlapping community discovery method based on node influence power |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522342A (en) * | 2018-11-30 | 2019-03-26 | 北京百度网讯科技有限公司 | Police affairs management method, device, equipment and storage medium |
| CN109714636A (en) * | 2018-12-21 | 2019-05-03 | 武汉瓯越网视有限公司 | A kind of user identification method, device, equipment and medium |
| CN109714636B (en) * | 2018-12-21 | 2021-04-23 | 武汉瓯越网视有限公司 | User identification method, device, equipment and medium |
| CN110428001A (en) * | 2019-07-31 | 2019-11-08 | 腾讯科技（深圳）有限公司 | A kind of community mining method, apparatus, server and storage medium |
| CN110428001B (en) * | 2019-07-31 | 2024-04-05 | 腾讯科技（深圳）有限公司 | Community mining method and device, server and storage medium |
| CN110502697A (en) * | 2019-08-26 | 2019-11-26 | 武汉斗鱼网络科技有限公司 | A kind of target user's recognition methods, device and electronic equipment |
| CN110502697B (en) * | 2019-08-26 | 2022-06-21 | 武汉斗鱼网络科技有限公司 | Target user identification method and device and electronic equipment |
| CN111010584A (en) * | 2019-12-05 | 2020-04-14 | 网易（杭州）网络有限公司 | Live broadcast data processing method and device, server and storage medium |
| CN112800286A (en) * | 2021-04-08 | 2021-05-14 | 北京轻松筹信息技术有限公司 | User relationship chain construction method and device and electronic equipment |
| CN112800286B (en) * | 2021-04-08 | 2021-07-23 | 北京轻松筹信息技术有限公司 | User relationship chain construction method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108763359A (en) | A kind of usage mining method, apparatus and electronic equipment with incidence relation | |
| CN109255632A (en) | A kind of user community recognition methods, device, equipment and medium | |
| CN106156941B (en) | A kind of user credit scoring optimization method and device | |
| CN101937447B (en) | Alarm association rule mining method, and rule mining engine and system | |
| CN110162519A (en) | Data clearing method | |
| EP4198775A1 (en) | Abnormal user auditing method and apparatus, electronic device, and storage medium | |
| CN107102941A (en) | The generation method and device of a kind of test case | |
| CN105869035A (en) | Mobile user credit evaluation method and apparatus | |
| CN104836696B (en) | A kind of detection method and device of IP address | |
| CN112860951B (en) | Method and system for identifying target account | |
| Leno et al. | Discovering process maps from event streams | |
| CN104965876A (en) | Method and apparatus for carrying out mining on work units of users on basis of position information | |
| Schliebs et al. | China’s inauthentic UK Twitter diplomacy: a coordinated network amplifying PRC diplomats | |
| Peng et al. | Visual abstraction of dynamic network via improved multi-class blue noise sampling | |
| CN105354249A (en) | Multi-account relevance method and device, and electronic equipment | |
| CN107481039A (en) | A kind of event-handling method and terminal device | |
| CN107547523A (en) | Message processing method, device, the network equipment and machinable medium | |
| CN107563204A (en) | A kind of privacy leakage methods of risk assessment of anonymous data | |
| CN110096646A (en) | The generation of category related information and its video pushing method and relevant device | |
| CN104156458B (en) | The extracting method and device of a kind of information | |
| CN109002856A (en) | A kind of traffic characteristic automatic generation method and system based on random walk | |
| Lam et al. | Internet diffusion in Vietnam | |
| CN113360895A (en) | Station group detection method and device and electronic equipment | |
| CN110851684B (en) | Social topic influence recognition method and device based on ternary association graph | |
| CN112667961A (en) | Method and system for identifying advertisement bullet screen publisher |
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: 20181106 |