The content of the invention
The present invention provides a kind of corpse account detection method and device, to solve to detect corpse account in the prior art
The problem of time is longer, and detection efficiency is relatively low, and resource consumption is larger.
It is an aspect of the present invention to provide a kind of corpse account detection method, including:
User account set is obtained, and determines that the concern that each user account is mutual in the user account set is closed
System;
According to the concern relation, the attraction between colony that each user account belonged to and each user account is determined
Degree, the attraction degree are used to characterize the concern tightness degree between each user account;
If the attraction degree belonged between each user account of the same group is more than the first predetermined threshold value, and belongs to
The average of attraction degree between each user account of the same group and each user account for belonging to other colonies is less than the
Two predetermined threshold values, it is determined that each user account for belonging to the same group is corpse account.
In method as described above, user account set is being obtained, and determine each user's account in the user account set
After number mutual concern relation, further include:
According to the concern relation, the influence force value of each user account is determined, the influence force value is for characterizing each user
The degree that account is concerned;
Correspondingly, before each user account that the same group is belonged to described in described determine is corpse account, further include:
Whether the influence force value that each user account of the same group is belonged to described in judgement is less than the 3rd predetermined threshold value;
The step of each user account for belonging to the same group is corpse account is determined if so, then performing.
It is described according to the concern relation in method as described above, determine colony that each user account belonged to and
Attraction degree between each user account, including:
Computing is carried out to the concern relation using figure segmentation algorithm, obtains the colony that each user account is belonged to;
Computing is carried out to the concern relation using power guidance algorithm, obtains the attraction degree between each user account.
It is described according to the concern relation in method as described above, determine the influence force value of each user account, including:
Computing is carried out to the concern relation using page rank (Pagerank, abbreviation PR) algorithm, obtains each user's account
Number influence force value.
In method as described above, it is described determine each user account for belonging to the same group for corpse account it
Afterwards, further include:
Attraction degree between the colony that is belonged to according to definite each user account, each user account and definite
Corpse account, determines community relation figure and shows the community relation figure, the community relation figure characterizes each user account phase
Concern tightness degree, the group property and corpse account of each user account between mutually, so that user checks that the community is closed
It is the content that the corpse account shown in figure is issued.
Another aspect of the present invention is to provide a kind of corpse account detection device, including:
Acquisition module, for obtaining user account set, and determines that each user account is mutual in the user account set
Between concern relation;
First determining module, for according to the concern relation, determining colony that each user account belonged to and each using
Attraction degree between the account of family, the attraction degree are used to characterize the concern tightness degree between each user account;
Second determining module, if the attraction degree between each user account for belonging to the same group is more than first
Predetermined threshold value, and belong to the attraction between each user account of the same group and each user account for belonging to other colonies
The average of degree is less than the second predetermined threshold value, it is determined that each user account for belonging to the same group is corpse account.
In device as described above, further include:
3rd determining module, for obtaining user account set in the acquisition module, and determines the user account collection
In conjunction after the mutual concern relation of each user account, according to the concern relation, the influence power of each user account is determined
Value, the influence force value are used to characterize the degree that each user account is concerned;
Correspondingly, further include:
Judgment module, for determining each user account for belonging to the same group for deadlock in second determining module
Whether the influence force value that each user account of the same group is belonged to before corpse account, described in judgement is less than the 3rd predetermined threshold value;
Determine each user account for belonging to the same group for corpse account if so, then performing in second determining module
Step.
In device as described above, first determining module, is specifically used for:
Computing is carried out to the concern relation using figure segmentation algorithm, obtains the colony that each user account is belonged to;
Computing is carried out to the concern relation using power guidance algorithm, obtains the attraction degree between each user account.
In device as described above, the 3rd determining module, is specifically used for:
Computing is carried out to the concern relation using PR algorithms, obtains the influence force value of each user account.
In device as described above, further include:
4th determining module, for determining each user account for belonging to the same group in second determining module
After corpse account, attraction degree between the colony that is belonged to according to definite each user account, each user account with
And definite corpse account, determine community relation figure and show the community relation figure, the community relation figure characterizes each use
The mutual concern tightness degree of family account, the group property and corpse account of each user account, so that user checks institute
State the content that the corpse account shown in community relation figure is issued.
The present invention determines the pass that each user account is mutual in user account set by obtaining user account set
Note relation;According to concern relation, the attraction degree between colony that each user account belonged to and each user account is determined,
Attraction degree is used to characterize the concern tightness degree between each user account;If belong to the same group each user account it
Between attraction degree be more than the first predetermined threshold value, and belong to the same group each user account and belong to other colonies
The average of attraction degree between each user account is less than the second predetermined threshold value, it is determined that belongs to each user of the same group
Account is corpse account.Realize and only gone to detect corpse account according to the concern relation of each user account, detect corpse account
Number time it is shorter, improve detection efficiency, and saved resource.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, the technical solution in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
Part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
All other embodiments obtained without making creative work, belong to the scope of protection of the invention.
Fig. 1 is the flow chart for the corpse account detection method that the embodiment of the present invention one provides, as shown in Figure 1, the present embodiment
Method include:
Step 101, obtain user account set, and determines the concern that each user account is mutual in user account set
Relation.
In the present embodiment, specifically, user account set is obtained first, for example, obtaining some brand official microblogging
Under whole beans vermicelli user account, form a user account set;Or obtain whole user's accounts under some forum
Number, form a user account set.
Then the concern relation that each user account is mutual in user account set is obtained.
Step 102, according to concern relation, determine the suction between colony that each user account belonged to and each user account
Gravitation degree, attraction degree are used to characterize the concern tightness degree between each user account.
In the present embodiment, specifically, according to the mutual concern relation of each user account, to all user accounts
It is grouped, determines the colony that each user account is belonged to, colony's mark is assigned to each user account.And root
According to the mutual concern relation of each user account, the concern tightness degree between each user account is calculated, so as to
Obtain the attraction degree between each user account.
If step 103, the attraction degree belonged between each user account of the same group are more than the first predetermined threshold value,
And belong to the equal of the attraction degree between each user account of the same group and each user account for belonging to other colonies
Value is less than the second predetermined threshold value, it is determined that each user account for belonging to the same group is corpse account.
In the present embodiment, if specifically, the attraction degree between belonging to each user account of the same group is all big
In the first predetermined threshold value, while belong between each user account of the same group and each user account for belonging to other colonies
The average of attraction degree be less than the second predetermined threshold value, it may be determined that going out these each user accounts for belonging to the same group is
Corpse account.
The present embodiment is by according to the mutual concern relation of each user account in user account set, obtaining each user
Attraction degree between colony that account is belonged to and each user account, belong to the same group each user account it
Between attraction degree be more than the first predetermined threshold value, and belong to the same group each user account and belong to other colonies
In the case that the average of attraction degree between each user account is less than the second predetermined threshold value, determine to belong to the same group
Each user account is corpse account.Realize and only gone to detect corpse account according to the concern relation of each user account, detect
The time of corpse account is shorter, improves detection efficiency, and saved resource.
Fig. 2 is the flow chart of corpse account detection method provided by Embodiment 2 of the present invention, on the basis of embodiment one,
As shown in Fig. 2, the method for the present embodiment includes:
After step 101, further include:
Step 201, according to concern relation, determine the influence force value of each user account, influence force value and be used to characterize each user
The degree that account is concerned;
Correspondingly, in step 103, before determining that each user account for belonging to the same group is corpse account, also
Including:Whether the influence force value for each user account for judging to belong to the same group is less than the 3rd predetermined threshold value;If so, then perform
The step of each user account for determining to belong to the same group in step 103 is corpse account.
In the present embodiment, specifically, according to the mutual concern relation of each user account, each user account is determined
The degree being concerned, so as to obtain the influence force value of each user account.
Then, if the attraction degree between belonging to each user account of the same group in step 103 is both greater than
One predetermined threshold value, while belong to the suction between each user account of the same group and each user account for belonging to other colonies
The average of gravitation degree is less than the second predetermined threshold value, meanwhile, if belonging to the influence force value of each user account of the same group all
Less than the 3rd predetermined threshold value, then it is corpse account that can determine that these belong to each user account of the same group.
The present embodiment by determine each user account influence force value, according to belong to the same group each user account it
Between attraction degree be more than the first predetermined threshold value, and belong to the same group each user account and belong to other colonies
The average of attraction degree between each user account is less than the second predetermined threshold value, and belongs to each user's account of the same group
Number influence force value be both less than the 3rd predetermined threshold value in the case of, it is corpse account to determine each user account for belonging to the same group
Number.Realizing and only gone to detect corpse account according to the concern relation of each user account, the time of detection corpse account is shorter,
Detection efficiency is improved, and has saved resource, while improves the accuracy rate of detection corpse account.
Further, on the basis of above-described embodiment, the embodiment of step 102, including:
Computing is carried out to concern relation using figure segmentation algorithm, obtains the colony that each user account is belonged to;
Computing is carried out to concern relation using power guidance algorithm, obtains the attraction degree between each user account.
In the present embodiment, specifically, according to the mutual correlation of each user account, figure segmentation algorithm is passed through
(Louvain Modularity, abbreviation LM) algorithm correlation mutual to each user account carries out computing, to user
Account carries out cutting, can obtain the colony that each user account is belonged to.According to the mutual correlation of each user account,
Computing is carried out by the power guidance algorithm correlation mutual to each user account, can be to user account cutting after
User data is calculated, and can obtain making a concerted effort for attraction between each user account and repulsive force, it is each use that this, which is made a concerted effort,
Attraction degree between the account of family, attraction degree indicate the concern tightness degree between each user account.
Further, on the basis of above-described embodiment, the embodiment of step 201, including:
Computing is carried out to concern relation using PR algorithms algorithm, obtains the influence force value of each user account.
In the present embodiment, specifically, PR algorithms can be used to map the concern relation of each user account
Computing is realized on reduction (Map-Reduce, abbreviation MR) programming model, obtains the social influence power of each user account, i.e., each user
The influence force value of account.
Further, on the basis of above-described embodiment, determine to belong to each user account of the same group in step 103
After corpse account, further include:
Attraction degree between the colony that is belonged to according to definite each user account, each user account and definite
Corpse account, determines community relation figure and shows community relation figure, and it is mutual that community relation figure characterizes each user account
Tightness degree, the group property and corpse account of each user account are paid close attention to, so that user checks in community relation figure what is shown
The content that corpse account is issued.
In the present embodiment, specifically, determine each user account for belonging to the same group for corpse account it
Afterwards, attraction degree, the shadow of each user account between the colony that is belonged to according to definite each user account, each user account
Force value and definite corpse account are rung, determines community relation figure.The size of the influence force value of each user account, it is determined that community
Each size of node of each user account is represented in graph of a relation.The colony that each user account is belonged to, it is determined that community relation figure
The color of the middle each node for representing each user account, the color for belonging to the user account of the same group are identical.Each user account
Between attraction degree, it is determined that the distance between each node of each user account is represented in community relation figure, can be known
The mutual concern tightness degree of each user account;It is lower at a distance of more remote user account, concern tightness degree;At a distance of nearer
User account, concern tightness degree it is higher.
Community can be drawn out using power guidance algorithm and closed by drawing script (JavaScript, abbreviation JS) storehouse of increasing income
System's figure, progress figure renders on webpage.In the community relation figure of display, it can be seen that the corpse account judged.Corpse
Account has following feature:Social influence power first is smaller, i.e. influence force value is smaller, and the node in community relation figure is all very
Small, if pressing diameter ranking, the diameter of these points is distributed in rear 20%;Concern relation between corpse account is very much,
But the account of other colonies is seldom paid close attention to, corpse account meeting high degree of internal polymerization, point is gathered into cluster, which can be away from it
His colony, the edge of cluster can be more than the diameter of the cluster with a distance from the edge of other colony's clusters.
Fig. 3 is the structure diagram for the corpse account detection device that the embodiment of the present invention three provides, as shown in figure 3, this reality
The corpse account detection device of example is applied, including:
Acquisition module 31, for obtaining user account set, and determine in user account set each user account mutually it
Between concern relation;
First determining module 32, for according to concern relation, determining colony and each user that each user account belonged to
Attraction degree between account, attraction degree are used to characterize the concern tightness degree between each user account;
Second determining module 33, if the attraction degree between each user account for belonging to the same group is more than the
One predetermined threshold value, and belong to the attraction between each user account of the same group and each user account for belonging to other colonies
The average of range degree is less than the second predetermined threshold value, it is determined that each user account for belonging to the same group is corpse account.
The corpse account detection device of the present embodiment can perform the corpse account detection method of the offer of the embodiment of the present invention one,
Its realization principle is similar, and details are not described herein again.
The present embodiment is by according to the mutual concern relation of each user account in user account set, obtaining each user
Attraction degree between colony that account is belonged to and each user account, belong to the same group each user account it
Between attraction degree be more than the first predetermined threshold value, and belong to the same group each user account and belong to other colonies
In the case that the average of attraction degree between each user account is less than the second predetermined threshold value, determine to belong to the same group
Each user account is corpse account.Realize and only gone to detect corpse account according to the concern relation of each user account, detect
The time of corpse account is shorter, improves detection efficiency, and saved resource.
Fig. 4 is the structure diagram for the corpse account detection device that the embodiment of the present invention four provides, in the base of embodiment three
On plinth, as shown in figure 4, the corpse account detection device of the present embodiment, further includes:
3rd determining module 41, for obtaining user account set in acquisition module 31, and determines in user account set
After the mutual concern relation of each user account, according to concern relation, the influence force value of each user account, influence power are determined
Value is used to characterize the degree that each user account is concerned;
Correspondingly, further include:
Judgment module 42, for determining that each user account for belonging to the same group is corpse account in the second determining module 33
Before number, whether the influence force value for each user account for judging to belong to the same group is less than the 3rd predetermined threshold value;If so, then hold
The step of definite each user account for belonging to the same group in the second determining module of row 33 is corpse account.
First determining module 32, is specifically used for:
Computing is carried out to concern relation using figure segmentation algorithm, obtains the colony that each user account is belonged to;
Computing is carried out to concern relation using power guidance algorithm, obtains the attraction degree between each user account.
3rd determining module 41, is specifically used for:
Computing is carried out to concern relation using PR algorithms, obtains the influence force value of each user account.
Further include:
4th determining module 43, for determining that each user account for belonging to the same group is deadlock in the second determining module 33
After corpse account, attraction degree between the colony that is belonged to according to definite each user account, each user account and really
Fixed corpse account, determines community relation figure and shows community relation figure, community relation figure characterize each user account mutually it
Between concern tightness degree, the group property and corpse account of each user account so that user check in community relation figure show
The content that the corpse account shown is issued.
The corpse account detection device of the present embodiment can perform the corpse account that the embodiment of the present invention two and embodiment provide
Number detection method, its realization principle is similar, and details are not described herein again.
The present embodiment by determine each user account influence force value, according to belong to the same group each user account it
Between attraction degree be more than the first predetermined threshold value, and belong to the same group each user account and belong to other colonies
The average of attraction degree between each user account is less than the second predetermined threshold value, and belongs to each user's account of the same group
Number influence force value be both less than the 3rd predetermined threshold value in the case of, it is corpse account to determine each user account for belonging to the same group
Number.Realizing and only gone to detect corpse account according to the concern relation of each user account, the time of detection corpse account is shorter,
Detection efficiency is improved, and has saved resource, while improves the accuracy rate of detection corpse account.
One of ordinary skill in the art will appreciate that:Realizing all or part of step of above-mentioned each method embodiment can lead to
The relevant hardware of programmed instruction is crossed to complete.Foregoing program can be stored in a computer read/write memory medium.The journey
Sequence upon execution, execution the step of including above-mentioned each method embodiment;And foregoing storage medium includes:ROM, RAM, magnetic disc or
Person's CD etc. is various can be with the medium of store program codes.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
The present invention is described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that:It still may be used
To modify to the technical solution described in foregoing embodiments, or equivalent substitution is carried out to which part technical characteristic;
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical solution spirit and
Scope.