CN107886112A - A kind of clustering objects method, apparatus and storage device - Google Patents
A kind of clustering objects method, apparatus and storage device Download PDFInfo
- Publication number
- CN107886112A CN107886112A CN201711014564.0A CN201711014564A CN107886112A CN 107886112 A CN107886112 A CN 107886112A CN 201711014564 A CN201711014564 A CN 201711014564A CN 107886112 A CN107886112 A CN 107886112A
- Authority
- CN
- China
- Prior art keywords
- cluster
- group
- degree
- user
- object group
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
The embodiment of the invention discloses clustering objects method, apparatus and storage device, applied to technical field of information processing.In the method for the present embodiment, clustering objects device can be worth to cluster degree according to the association between object group, then determine which object group clustered further according to cluster degree when carrying out clustering objects.The intimate degree between object group is so represented by the relating value between object group, compared with needing to represent the intimate degree between object by the distance of the characteristic vector between object in the prior art, greatly reduces amount of calculation；And eliminate with characteristic vector to represent an object, so as to lift the effect of clustering objects.
Description
Technical field
The present invention relates to technical field of information processing, more particularly to a kind of clustering objects method, apparatus and storage device.
Background technology
The application program of existing many community's classes, such as K songs or the application program such as live, between the user in same community
Interaction can be carried out by respective applications client, for example commented on, thumbed up, give a present or send flower etc., therefore, excavate with phase
In-depth analysis of the user for community with interest has very important significance.
Specifically, server can first determine each similar population, then user profile of the can in similar population
Information recommendation is carried out, such as, the information recommendation that the applications client of a user in similar population is paid close attention to is in similar population
Applications client of other users etc., and server can also guide other in the applications client concern similar population of user
User etc., this is interactive between applications client for strengthening community users, and improving the viscosity of product has very important meaning
Justice.
Wherein, it is determined that during similar population, server typically first obtains the characteristic vector of each user, then passes through cluster
Method, such as the mode such as K averages (k-means), same similar population is divided into by the user closer to the distance of characteristic vector
In.But by this method, it is necessary to calculate the characteristic information of the user when it is determined which similar population a user belong to
With the distance between characteristic vector of each user in each similar population, amount of calculation is bigger；In addition, with characteristic vector come table
It is also relatively difficult thing to show a user, and the quality of Feature Selection directly determines the effect of cluster.
The content of the invention
The embodiment of the present invention provides a kind of clustering objects method, apparatus and storage device, realizes according to the first object group
Relating value between at least one second object group respectively, determine that the first object group is clustered with which the second object group.
First aspect of the embodiment of the present invention provides a kind of clustering objects method, including：
It is determined that at least one second object group associated with the first object group, wherein, any object group is included at least
One object；
The first object group relating value between at least one second object group respectively is determined, obtains at least one pass
Connection value；
The cluster degree according to corresponding to the function calculating formula of at least one relating value and preset cluster degree calculates respectively,
Obtain at least one cluster degree；
If the first cluster degree meets preset first condition at least one cluster degree, by the first cluster degree
Corresponding second object group is with the first object group cluster into same target group.
Second aspect of the embodiment of the present invention provides a kind of clustering objects device, including：
Object group determining unit, for determining at least one second object group associated with the first object group, wherein, appoint
One object group includes at least one object；
Determining unit is associated, for determining the first object group associating between at least one second object group respectively
Value, obtains at least one relating value；
Computing unit, for being calculated respectively according to the function calculating formula of at least one relating value and preset cluster degree
Corresponding cluster degree, obtain at least one cluster degree；
Cluster cell, will if meeting preset first condition for the first cluster degree at least one cluster degree
Second object group corresponding to the first cluster degree is with the first object group cluster into same target group.
The third aspect of the embodiment of the present invention provides a kind of storage device, a plurality of instruction of storage device stored, the finger
Order is suitable to be loaded as processor and perform the clustering objects method as described in first aspect of the embodiment of the present invention.
Fourth aspect of the embodiment of the present invention provides a kind of server, including processor and storage device, the processor, uses
In realizing each instruction；
The storage device is used to store a plurality of instruction, described to instruct for being loaded by processor and being performed as of the invention real
Apply the clustering objects method described in a first aspect.
It can be seen that in the method for the present embodiment, clustering objects device, can be according between object group when carrying out clustering objects
Association be worth to cluster degree, then determine to be clustered which object group further according to cluster degree.So by object group it
Between relating value represent object group between intimate degree, with need in the prior art by the characteristic vector between object away from
Compared from come the intimate degree represented between object, greatly reduce amount of calculation；And eliminate with characteristic vector to represent one
Individual object, so as to lift the effect of clustering objects.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also
To obtain other accompanying drawings according to these accompanying drawings.
Fig. 1 is the schematic diagram for the scene that a kind of clustering objects method provided in an embodiment of the present invention is applied to；
Fig. 2 is a kind of flow chart for clustering objects method that one embodiment of the invention provides；
Fig. 3 a are the relation schematic diagrams between a kind of first object group and the second object group in one embodiment of the invention；
Fig. 3 b are the relation schematic diagrams between another first object group and the second object group in one embodiment of the invention；
Fig. 3 c are the relation schematic diagrams between another first object group and the second object group in one embodiment of the invention；
Fig. 3 d are the relation schematic diagrams between the first object group of another in one embodiment of the invention and the second object group；
Fig. 3 e are the relation schematic diagrams between a kind of each object group in one embodiment of the invention；
Fig. 3 f are the relation schematic diagrams between another each object group in one embodiment of the invention；
Fig. 4 is a kind of flow chart for clustering objects method that Application Example of the present invention provides；
Fig. 5 a are the relation schematic diagrams between each user in Application Example of the present invention；
Fig. 5 b are the relation schematic diagrams between a kind of user's group 1 and user's group 2 in Application Example of the present invention；
Fig. 5 c are the relation schematic diagrams between another user's group 1 and user's group 2 in Application Example of the present invention；
Fig. 6 is a kind of structural representation of clustering objects device provided in an embodiment of the present invention；
Fig. 7 is the structural representation of another clustering objects device provided in an embodiment of the present invention；
Fig. 8 is a kind of flow chart of server provided in an embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
Term " first ", " second ", " the 3rd " " in description and claims of this specification and above-mentioned accompanying drawing
The (if present)s such as four " are for distinguishing similar object, without for describing specific order or precedence.It should manage
The data that solution so uses can exchange in the appropriate case, so as to embodiments of the invention described herein for example can with except
Order beyond those for illustrating or describing herein is implemented.In addition, term " comprising " and " having " and theirs is any
Deformation, it is intended that including not exclusively is covered, for example, containing the process of series of steps or unit, method, system, production
Product or equipment are not necessarily limited to those steps clearly listed or unit, but may include not list clearly or for this
The intrinsic other steps of a little process, method, product or equipment or unit.
The embodiment of the present invention provides a kind of clustering objects method, the mainly method performed by clustering objects device.Can be with
Applied in scene as shown in Figure 1, the scene includes application server and multiple (illustrating in Fig. 1 exemplified by n) applications
Client.Wherein, applications client is the client of community's class application, such as K songs or the client such as live, and user can be with
Application server is logged in using user profile, and by applications client, such a applications client can correspond to a use
Family；Application server can identify user by user profile, and the user to identifying clusters.In this case,
Clustering objects device is specially application server, and application server is clustered user as object.
The clustering objects method of the embodiment of the present invention can also be applied to other scenes, such as, it is necessary to other virtual objects
Product, or the scene that song etc. is clustered, the scene of the clustering objects method of the embodiment of the present invention is not carried out herein
Limitation.
Specifically, in embodiments of the present invention, clustering objects device can carry out clustering objects by the following method：
It is determined that at least one second object group associated with the first object group, wherein, any object group is included at least
One object；Relating value of the first object group respectively between at least one second object group is determined, obtains at least one association
Value；The cluster degree according to corresponding to the function calculating formula of at least one relating value and preset cluster degree calculates respectively, is obtained at least
One cluster degree；It is if the first cluster degree meets preset first condition at least one cluster degree, the first cluster degree is corresponding
The second object group and the first object group cluster into same target group.
It can be seen that clustering objects device when carrying out clustering objects, can be worth to cluster degree according to the association between object group,
Then determine which object group clustered further according to cluster degree.The relating value expression pair between object group can so be passed through
As the intimate degree between group, with needing to represent between object by the distance of the characteristic vector between object in the prior art
Intimate degree compare, greatly reduce amount of calculation；And eliminate with characteristic vector to represent an object, so as to be lifted
The effect of clustering objects.
The embodiment of the present invention provides a kind of clustering objects method, the mainly method performed by clustering objects device, such as
Above-mentioned application server etc., flow chart as shown in Fig. 2 including：
Step 101, it is determined that at least one second object group associated with the first object group, wherein, in any object group
Including at least one object.
It is appreciated that clustering objects device can initiate the flow of the present embodiment according to certain cycle.
Wherein, at least one object can be included in each object group, object here can pass through application client
The virtual objects in the user for being registered to application server or game server are held, can also be that song server carries
Song of confession etc..It is right when starting cluster and each object group can be clustered to obtain by the method for the present embodiment
As clustering apparatus can so be grasped using an object as an object group by the repetition of step 101 to 104 in the present embodiment
Make, the object group after cluster meets some requirements, and may include multiple objects in so each object group.
Clustering objects device initiate the present embodiment flow after, can by the attribute information of each object, it is determined that with
The second associated object group of a certain object group (such as first object group).
(1) if object is user, the user of application server can be registered to by applications client according to user
Information, and information for being operated by applications client of user etc., it is determined that the other user's groups associated with a certain user's group.
For example, the user between a certain user's group and another user's group has paid close attention to identical by respective applications client
Information, or, the user of a certain user's group has paid close attention to the user in another user's group by applications client, or, a certain use
User between family group and another user's group has carried out interaction etc. by respective applications client, it is determined that a certain user's group
It is associated with another user's group.
For example, applications client is K song clients, user a sings the program request song collection of client 1 A by K in user's group 1
It is b1, b2 ... ..., bm that for a1, a2 ... ..., an, in user's group 2, user b sings the song collection B that requests tune of client 2 by K, is gathered
There is identical song in A and set B；Or user a has paid close attention to user b in user's group 2 by K song clients 1 in user's group 1；
Or user a sings the letter that client 1 forwarded user b in user's group 2 and pass through K and sing client 2 and issue by K in user's group 1
Breath, or, user a and user b in user's group 2 is good friend in user's group 1, or, user a in user's group 2 with using in user's group 1
Family b sings client by respective K and carries out interaction etc..Then user's group 1 associates with user's group 2.
(2), can be according to the content of song, the download user of song, the program request user of song if object is song
Etc. information, it is determined that another group of songs associated with a certain group of songs etc..
For example, the download user of a certain song, between the download user of another song, the quantity of same subscriber is more than one
Individual threshold value；Or the content of a certain song and another song is all the then phase between a certain song and another song on " love "
Association.
Step 102, relating value of the first object group respectively between at least one second object group is determined, obtains at least one
Individual relating value.
Here relating value can represent the intimate degree between the object of two object groups, if relating value is bigger, say
It is more intimate between the object of bright two object groups.Specifically, clustering objects device it is determined that the first object group with some second
During relating value between object group, can first determine the second object in the first object and the second object group in the first object group it
Between, based on the relationship score of at least one dimension, wherein, the first object and the second object are associated objects；Then first pair
As the relating value between group and the second object group has following several situations：
(1) if the first object and the second object are all one, clustering objects device can be by the first object and second pair
The mathematical calculation of relationship score based at least one dimension as between is as between the first object group and the second object group
Relating value.Such as using between the first object and the second object based on the relationship score sum of at least one dimension as first pair
As the relating value between group and the second object group, or, by between the first object and the second object based at least one dimension
The product of relationship score is as the relating value between the first object group and the second object group.
Wherein, if object is user, the first object group is the first user's group, and the second object group is second user group, then
Above-mentioned at least one dimension can include but is not limited to following information：First user and second user are respectively by respective
The information of applications client operation, and relation between the first user and second user etc..Then relationship score specifically has following two
Kind situation：
(11) if at least one dimension includes：The letter that first user and second user are operated by applications client respectively
Breath, the then relationship score based at least one dimension specifically include：The first information that first user is operated by applications client,
Identical information bar number in the second information operated with second user by applications client, divided by the first information and the second information
Information bar number sum, obtained quotient.
(12) if at least one dimension includes：Relation between first user and second user, then based at least one
The relationship score of dimension specifically includes：The number of species sum of relation between first user and second user, wherein, the first user
Relation between second user has a variety of, for example the first user and second user be good friend, or the first user and the second use
Family carries out interaction by respective applications client, or the first user has paid close attention to second user etc. by applications client.
For example, applications client is K song clients, if user a sings the program request song book of client 1 by K in user's group 1
Conjunction A is a1, a2 ... ..., an, and it is b1, b2 ... ..., bm that user b sings the song collection B that requests tune of client 2 by K in user's group 2,
It is p to have identical number of songs in set A and set B, then passes through the song of K song client program requests this dimension based on user
Relationship score s1 can be obtained by equation below 1：
If in user's group 1 user a by K sing client 1 paid close attention to user b in user's group 2, can using score value s21 as
1；If user a and user b in user's group 2 is good friend in user's group 1, can be using score value s22 as 1；If user in user's group 1
A and user b in user's group 2 sings client by respective K and carries out interaction, such as comment etc., then and score value s23 is 1, then is based on
Be related to this dimension between first user and second user, obtained relationship score be above-mentioned score value s21, s22 and s23 it
With.
Then the relating value between user's group 1 and user's group 2 can be above-mentioned relationship score s1 and relationship score s2 mathematics
Calculated value, such as s1 and s2 product, or s1 and s2 sums etc..
(2) if above-mentioned first object group and the second object group all include multiple objects, in so possible first object group
An object, it is all relevant between multiple objects in the second object group, i.e. the first object be one, the second object is more
Individual, then the relationship score based at least one dimension can include between clustering objects device determines the first object and the second object
One the first object relationship score based at least one dimension between multiple second objects respectively.
So clustering objects device can also calculate first object and be based at least one between multiple second objects respectively
The mathematical calculation of the relationship score of individual dimension, obtains multiple mathematical calculations, then using this multiple mathematical calculation sum as
The relating value of first object group and the second object group.
Such as shown in Fig. 3 a, an object is represented with a round dot, between line two objects of expression between points
It is associated, and the length of line can represent the intimate degree between two objects between points.The object 1 of first object group
All associated respectively with the object 2 and object 3 of the second object group, then clustering objects device can calculate object 1 and is based on extremely with object 2
The mathematical calculation s12 of the relationship score of a few dimension, and object 1 and relationship score of the object 3 based at least one dimension
Mathematical calculation s13, and using mathematical calculation s12 and s13 sum as the pass between the first object group and the second object group
Connection value.
(3) if above-mentioned first object group and the second object group all include multiple objects, in so possible first object group
Multiple objects, relevant between multiple objects in the second object group respectively, i.e. to be multiple, the second object is the first object
Multiple, then the relationship score based at least one dimension can wrap between clustering objects device determines the first object and the second object
Include each first object association based at least one dimension between the second associated object respectively in multiple first objects
Score value.
So clustering objects device can also calculate in multiple first objects each first object respectively with associated second
The mathematical calculation of relationship score based at least one dimension between object, multiple mathematical calculations are obtained, then this is multiple
Relating value of the mathematical calculation sum as the first object group and the second object group.
Such as shown in Fig. 3 b, an object is represented with a round dot, between line two objects of expression between points
It is associated, and the length of line can represent the cohesion between two objects between points.The object 1 of first object group with
The object 2 of second object group is associated, and the object 3 of the first object group is associated with the object 4 of the second object group, then clustering objects
Device can calculate the mathematical calculation s12 of object 1 and relationship score of the object 2 based at least one dimension, and object 3 and object
The mathematical calculation s34 of 4 relationship scores based at least one dimension, and using mathematical calculation s12 and s34 sum as first
Relating value between object group and the second object group.
(4) if above-mentioned first object group and the second object group all include multiple objects, in so possible first object group
Multiple objects, relevant between an object in the second object group respectively, i.e. to be multiple, the second object is the first object
One, then the relationship score based at least one dimension can wrap between clustering objects device determines the first object and the second object
Include in multiple first objects each first object relationship score based at least one dimension between second object respectively.
Clustering objects device can also calculate in multiple first objects each first object respectively between second object
The mathematical calculation of relationship score based at least one dimension, obtains multiple mathematical calculations, then by this multiple mathematical computations
It is worth relating value of the sum as the first object group and the second object group.
Such as shown in Fig. 3 c, an object is represented with a round dot, between line two objects of expression between points
It is associated, and the length of line can represent the cohesion between two objects between points.The object 1 of first object group with
The object 2 of second object group is associated, and the object 3 of the first object group is associated with the object 2 of the second object group, then clustering objects
Device can calculate the mathematical calculation s12 of object 1 and relationship score of the object 2 based at least one dimension, and object 3 and object
The mathematical calculation s32 of 2 relationship scores based at least one dimension, and using mathematical calculation s12 and s32 sum as first
Relating value between object group and the second object group.
Step 103, calculated respectively correspondingly according to the function calculating formula of above-mentioned at least one relating value and preset cluster degree
Cluster degree, obtain at least one cluster degree.
Here, according to corresponding to some cluster degree for being worth to of association refers to the relating value the first object group with second pair
After group cluster to same target group, the extent of polymerization of the object group after cluster.Wherein, the function of preset cluster degree calculates
Formula can be the calculation formula that the relating value determined according to above-mentioned steps 102 calculates a cluster degree, can be specifically to appoint
The calculation formula of meaning form.In a specific embodiment, the function calculating formula of the cluster degree can include equation below 2：
Wherein, ∑inFor the relating value sum between object in the first object group, in a certain second object group between object
Relating value sum, the relating value between the first object group and a certain second object group be added after value；∑totFor first
The relating value sum between object in object group and a certain second object group in object, with other object groups；It is all right that m includes
Relating value sum as between.
It should be noted that if a certain object group includes an object, ∑ is being calculatedinWhen, it is right in the object group
Relating value sum as between is zero.
Such as shown in Fig. 3 d, an object is represented with a round dot, between line two objects of expression between points
It is associated, and the length of line can represent the cohesion between two objects between points.First object group includes object
1,2 and 3, a certain second object group includes object 4,5 and 6, then ∑inRelating value s12 between object 1 and object 2, object 2
Relating value s23 between object 3, the relating value between the first object group and the second object group is (i.e. between object 3 and object 4
Relating value s34), the relating value s56 sums between relating value s45 and object 5 and object 6 between object 4 and object 5.
∑totRelating value s1t between object 1 and other objects, and the relating value s6p sums between object 6 and other objects.
Step 104, if the first cluster degree at least one cluster degree meets preset first condition, by the first cluster
Second object group corresponding to degree is with the first object group cluster into same target group.
It is appreciated that clustering objects device needs first to judge which cluster degree meets preset at least one cluster degree
First condition, if some cluster degree meets that the cluster degree is the first cluster degree, by second pair corresponding to the first cluster degree
As group and the first object group cluster are into same target group；If any cluster degree is all unsatisfactory for, will not be clustered.
Here preset first condition be determine the first object group and the second object group whether the condition that can be clustered, specifically may be used
With including the following two kinds situation：
(1) if at least one cluster degree includes a cluster degree, illustrate to associate with the first object group only one the
Two object groups, then the cluster degree is the first cluster degree, and preset first condition is：Integrally degree of cluster is big corresponding to first cluster degree
In current overall cluster degree, current overall cluster degree is that the second object group corresponding to the first cluster degree is clustering with the first object group
The cluster degree sum of all object groups before, integrally degree of cluster is second pair corresponding to the first cluster degree corresponding to the first cluster degree
As organizing the cluster degree sum with the first object group all object groups after cluster.Overall cluster degree specifically can be by following public
Formula 3 represents, wherein, c represents all object groups：
Such as shown in Fig. 3 e, an object is represented with a round dot, between line two objects of expression between points
It is associated, and the length of line can represent the cohesion between two objects between points.All object groups include first
To including object 1,2 and 3, the second object group as group, the second object group, the object group 7 that a certain object 7 forms, the first object group
Including object 4,5 and 6.Associated with the first object group only a second object group, then according to the first object group and this second
A cluster degree can be calculated in the relating value of object group.Then it is determined that whether the first object group and the second object group can gather
, it is necessary to calculate current overall cluster degree during class, and integrally degree of cluster corresponding to a cluster degree being calculated.
Wherein, when the current overall cluster of calculating is spent, all object groups are the first object group, the second object group, and object
Group 7；When integrally cluster is spent corresponding to one cluster degree of calculating, all object groups are the first object group and the second object group cluster
Object group (in figure shown in dotted line frame) afterwards, and object group 7.
(2) if at least one cluster degree includes multiple cluster degree, a certain cluster degree is first poly- in multiple cluster degree
Class degree, preset first condition are：Integrally degree of cluster is more than current overall cluster degree, and the first cluster corresponding to first cluster degree
Integrally degree of cluster integrally clusters more than corresponding to other cluster degree in multiple cluster degree in addition to the first cluster degree corresponding to degree
Degree.
Such as shown in Fig. 3 f, an object is represented with a round dot, between line two objects of expression between points
It is associated, and the length of line can represent the cohesion between two objects between points.All object groups include first
To including object 1 as group, the second object group, the object group 7 and object group 8 that a certain object 7 and 8 separately constitutes, the first object group,
2 and 3, the second object group includes object 4,5 and 6.What is associated with the first object group has two object groups, i.e. the second object group and right
As group 7, then cluster degree 1 can be calculated according to the relating value of the first object group and the second object group, according to object group 7 and
Cluster degree 2 is calculated in the relating value of first object group.Then when it is determined that the first object group can cluster with which object group, need
Calculate current overall cluster degree, overall overall degree of cluster corresponding to degree of cluster and cluster degree 2 corresponding to cluster degree 1.
Wherein, when the current overall cluster of calculating is spent, all object groups are the first object group, the second object group, and object
Group 7 and 8；When integrally cluster is spent corresponding to calculating cluster degree 1, all object groups are the first object group and the second object group cluster
Object group afterwards, and object group 7 and 8；When integrally cluster is spent corresponding to calculating cluster degree 2, all object groups are the first object
Object group after group and the cluster of object group 7, and second group and object group 8.
It should be noted that above-mentioned steps 101 to 104 are the sides performed by a certain object group (i.e. the first object group)
Method, after above-mentioned steps 101 to 104 have been carried out for all object groups, clustering objects device can further judge to gather
Whether the object group after class meets preset second condition, if it is satisfied, then terminating flow；If be unsatisfactory for, for cluster
Object group afterwards, which returns, performs step 101, that is, performs above-mentioned the second object of determination group, determine relating value, calculates cluster degree and gathers
The step of class.
Wherein preset second condition is determined whether to need to be directed to the object group after all clusters, again in circulation execution
The condition of step 101 to 104 flow is stated, can specifically be included：Corresponding to all object groups after the cluster that previous cycle obtains
Cluster degree sum (the overall cluster degree obtained after previous cycle), all object groups after obtained cluster are circulated with the last time
Corresponding cluster degree sum is identical (the overall cluster degree obtained after last time circulation) or approximate, i.e., overall cluster degree will not
When changing, it may not be necessary to circulated again.
Further, if object is user, object group is user's group, then when clustering objects device passes through above-mentioned flow
, can also will be other in addition to the first user in the user's group after cluster after carrying out the user's group after clustering objects are clustered
The information of user's operation, and/or user profile are sent to the applications client of the first user.First user can also be operated
Information, and/or user profile are sent to the applications client of other users.Wherein, the information of each user's operation can be this
The information that user is operated by corresponding applications client, such as information of concern or forwarding etc..
It can be seen that in the method for the present embodiment, clustering objects device, can be according between object group when carrying out clustering objects
Association be worth to cluster degree, then determine to be clustered which object group further according to cluster degree.So by object group it
Between relating value represent object group between intimate degree, with need in the prior art by the characteristic vector between object away from
Compared from come the intimate degree represented between object, greatly reduce amount of calculation；And eliminate with characteristic vector to represent one
Individual object, so as to lift the effect of clustering objects.
Illustrate the clustering objects method of the present invention with a specific embodiment below, the method for the present embodiment can answer
For in the scene shown in above-mentioned Fig. 1, specifically, in the present embodiment, applications client can be that K sings client, application service
Device is that K sings server, and object is user, and object group is user's group, and clustering objects device is that K sings server.With reference to shown in figure 4,
The method of the present embodiment may include steps of：
Step 201, K sings server and is directed to the user for singing server to K by K song client registers, initiates the present embodiment
Flow, K songs server can be first using each user as a user's group, for a user's group (such as user's group 1), really
The fixed user's group 2 associated with the user's group 1, specifically, in the present embodiment by user's group 2 be it is multiple exemplified by illustrate.
Such as shown in Fig. 5 a, user 1 to 6 is respectively as an independent user's group, for a certain user 6, with user 6
It is associated for user 1,2 and 5；It is associated with user 3 for user 2 and 4 for user 3.In figure one is represented with a round dot
Individual user, it is associated between representing user with the line between round dot.
Step 202, K sings server and determines relating value of the user's group 1 respectively between multiple user's groups 2, obtains multiple passes
Connection value.
Specifically, it is determined that during relating value between user's group 1 and a user's group 2, can first determine in user's group 1
One user (than user 6 as illustrated in fig. 5 a) and second user in user's group 1 (than user 1 as illustrated in fig. 5 a), are based on
The relationship score of multiple dimensions, then using the product of the relationship score of this multiple dimension as between user's group 1 and user's group 2
Relating value.
For example the relationship score s1 of this dimension of the song of client program request is sung by K based on user, it is user 6 and user
1 sings the quantity for the same song that client is requested tune by K, sings all songs of client program request by K with user 6 and user 1
Quantity ratio.And based on this dimension that is related between the first user and second user, obtained relationship score s2 is use
The number of species sum of relation between family 6 and user 1.Then using relationship score s1 and s2 product as between two user's groups
Relating value.
Step 203, K sings server and calculated respectively correspondingly according to the function calculating formula of multiple relating values and preset cluster degree
Cluster degree, obtain multiple cluster degree.Wherein, the according to corresponding to some cluster degree for being worth to of association refers to the relating value
One user's group and second user group cluster are to after same user's group, the extent of polymerization of the user's group after cluster.
Specifically, the function calculating formula of preset cluster degree can include above-mentioned formula 2 here, herein without superfluous
State.Such as shown in Fig. 5 a, calculated according to the relating value between user 1 and user 6 the two user's groups and the function of cluster degree
Formula calculate corresponding to cluster when spending, then ∑inFor the relating value between user 1 and user 6, ∑totFor user 6 respectively with user 2 and
Relating value sum between 5.
Step 204, if the first cluster degree meets preset first condition in multiple cluster degree, K sings server by first
User's group 2 corresponding to cluster degree is clustered in same user's group with user's group 1.
Wherein, integrally degree of cluster needs to be more than current overall cluster degree, and the first cluster degree pair corresponding to the first cluster degree
The overall cluster degree answered is more than overall degree of cluster corresponding to other cluster degree in multiple cluster degree in addition to the first cluster degree.
Such as shown in Fig. 5 a, when corresponding cluster is calculated according to user 6 and user 1,2 and 5 respectively in K songs server
Degree, obtains 3 cluster degree, i.e. cluster degree 1,2 and 5；Then overall degree of cluster corresponding to determining this 3 cluster degree respectively again, such as
Integrally degree of cluster 1 is maximum in these overall cluster degree of fruit, should if the overall cluster degree 1 is more than current overall cluster degree
User's group (such as user 1) corresponding to overall cluster degree 1 corresponding to cluster degree (such as cluster degree 1) arrives same with the cluster of user 6
In one user's group.
It should be noted that by the flow of above-mentioned steps 201 to 204, the side carried out just for some user's group
Method flow, K song servers can be directed to all user's groups, and circulation performs above-mentioned steps 201 to 204, the user after being clustered
Group.
Such as shown in Fig. 5 b, K sings server and clusters user 1 to 6 into user's group 1, the cluster of user 7 to 9 is arrived into user
In group 2, the user's group after cluster also includes other user's groups, does not draw in figure 5b.If also need to perform the stream of cluster
Journey, then, can be directly by the pass between user 4 and user 7 during relating value between calculating and user's group 2 for user's group 1
Connection value is as the relating value between user's group 1 and user's group 2.
Then when being spent according to relating value calculating cluster, if the function calculating formula of cluster degree is above-mentioned formula 2,
In this case, ∑inIncluding the relating value between user in the relating value between user 4 and user 7, in addition to user's group 1 it
With, and the relating value sum in user's group 2 between user.Relating value sum wherein in user's group 1 between user is specially：With
Relating value between family 1 and user 6, the relating value between user 6 and user 2, the relating value between user 6 and user 5, user
Relating value between 2 and user 3, the relating value sum between the relating value between user 3 and user 4, and user 4 and user 5.
Relating value sum in user's group 2 between user is specially：Relating value between user 7 and user 8, between user 7 and user 9
Relating value, and the relating value sum between user 9 and user 8.
And ∑totThen include the relating value between user's group 1 and other associated user's group (not shown)s.
In another case, if more than one user is associated with the user in user's group 2 in user's group 1, such as scheme
Shown in 5c, the user 3 and 4 in user's group 1 is associated with the user 7 in user's group 2, then is calculating user's group 1 and user's group 2
Between relating value when, can using user 7 respectively the relating value sum between user 3 and 4 as user's group 1 and user's group 2
Between relating value.
The embodiment of the present invention also provides a kind of clustering objects device, than application server described above, its structural representation
As shown in fig. 6, it can specifically include：
Object group determining unit 10, for determining at least one second object group associated with the first object group, wherein,
Any object group includes at least one object；
Associate determining unit 11, for determine the first object group respectively with the object group determining unit 10 determine
Relating value between at least one second object group, obtains at least one relating value.
The association determining unit 11, specifically for it is determined that between the first object group and some second object group
During relating value, the first object in the first object group is determined, between the second object in a certain second object group, is based on
The relationship score of at least one dimension；First object and the second object are associated objects；If first object
All it is one with the second object, then by the relationship score based at least one dimension between first object and the second object
Mathematical calculation is as the relating value between the first object group and the second object group.
Further, the association determining unit 11, if it is one to be specifically additionally operable to first object, the second object is
It is multiple, then calculate first object respectively between multiple second objects the relationship score based at least one dimension mathematics
Calculated value, multiple mathematical calculations are obtained, using the multiple mathematical calculation sum as the first object group and second pair
As the relating value of group；
If first object is multiple, the second object is multiple, then calculates respectively every in the multiple first object
The mathematical calculation of the relationship score based at least one dimension, is obtained more between individual first object and the second associated object
Individual mathematical calculation, the relating value using the multiple mathematical calculation sum as the first object group and the second object group.
Wherein, if above-mentioned object is user, the first object is the first user, and the second object is second user, then described
At least one dimension includes：The information that first user and second user are operated by applications client respectively, the then base
Specifically included in the relationship score of at least one dimension：The first information that first user is operated by applications client, with
Identical information bar number in the second information that second user is operated by applications client, divided by the first information and the second letter
The information bar number sum of breath, obtained quotient；If at least one dimension includes：Pass between first user and second user
System, the then relationship score based at least one dimension specifically include：The number of species of relation between first user and second user
Sum.
Computing unit 12, at least one relating value obtained according to the association determining unit 11 and preset cluster
The function calculating formula of degree calculates corresponding cluster degree respectively, obtains at least one cluster degree.Here, obtained according to some relating value
To cluster degree refer to the first object group corresponding to the relating value and the second object group cluster to after same target group, after cluster
The extent of polymerization of the object group
The function calculating formula of preset cluster degree can include：
Wherein, ∑inIt is a certain for the relating value sum between object in the first object group
Relating value phase between relating value sum in second object group between object, with the first object group and a certain second object group
Value after adding；The ∑totFor object in the first object group and a certain second object group, with the objects of other object groups it
Between relating value sum；The m includes the relating value sum between all objects.
Cluster cell 13, if the first cluster degree at least one cluster degree obtained for the computing unit 12 is expired
The preset first condition of foot, by the second object group corresponding to the first cluster degree and the first object group cluster to same target group
In.
The cluster cell 13, if including a cluster degree specifically at least one cluster degree, described first is poly-
Integrally degree of cluster is more than current overall cluster degree corresponding to class degree, and the current overall cluster degree is corresponding for the first cluster degree
The second object group and the first object group all object groups before cluster cluster degree sum, corresponding to the first cluster degree
Overall cluster degree is the second object group corresponding to the first cluster degree and the first object group all object groups after cluster
Cluster degree sum；
If at least one cluster degree includes multiple cluster degree, integrally degree of cluster is big corresponding to the first cluster degree
In the current overall cluster degree, and corresponding to the first cluster degree integrally degree of cluster more than removing institute in the multiple cluster degree
State corresponding to other cluster degree outside the first cluster degree integrally degree of cluster.
It can be seen that in the device of the present embodiment, computing unit 12 can be worth to cluster degree according to the association between object group,
Then cluster cell 13 determines which object group clustered further according to cluster degree.So pass through the relating value between object group
The intimate degree between object group is represented, with needing to represent pair by the distance of the characteristic vector between object in the prior art
Intimate degree as between is compared, and greatly reduces amount of calculation；And eliminate with characteristic vector to represent an object, so as to
The effect of clustering objects can be lifted.
With reference to shown in figure 7, in a specific embodiment, except that can include in the clustering objects device of the present embodiment
Outside structure as shown in Figure 6, circulation determining unit 14 and information transmitting unit 15 can also be included, wherein：
Determining unit 14 is circulated, for determining whether the object group after the cluster meets preset second condition, if
It is unsatisfactory for, notifies the object group determining unit 10 to perform the second object group of the determination for the object group after the cluster
Step.
Information transmitting unit 15, if being user for above-mentioned object, the first object is the first user, according to cluster cell
User's group after 13 obtained clusters, the letter that other users in the user's group after the cluster in addition to the first user are operated
Breath, and/or user profile are sent to the applications client of first user.
The embodiment of the present invention also provides a kind of server, its structural representation as shown in figure 8, the server can because configuration or
Performance is different and produces bigger difference, can include one or more central processing units (central
Processing units, CPU) 20 (for example, one or more processors) and memory 21, one or more are deposited
Store up the storage medium 22 (such as one or more mass memory units) of application program 221 or data 222.Wherein, store
Device 21 and storage medium 22 can be of short duration storage or persistently storage.Be stored in storage medium 22 program can include one or
More than one module (diagram does not mark), each module can include operating the series of instructions in server.Further
Ground, central processing unit 20 could be arranged to communicate with storage medium 22, perform on the server a series of in storage medium 22
Command operating.
Specifically, the application program 221 stored in storage medium 22 includes the application program of clustering objects, and the program
The object group determining unit 10 in above-mentioned clustering objects device can be included, associate determining unit 11, computing unit 12, cluster list
Member 13, determining unit 14 and information transmitting unit 15 are circulated, herein without repeating.Further, central processing unit 20 can
To be arranged to communicate with storage medium 22, it is corresponding that the application program with cluster stored in storage medium 22 is performed on the server
Sequence of operations.
Server can also include one or more power supplys 23, one or more wired or wireless network interfaces
24, one or more input/output interfaces 25, and/or, one or more operating systems 223, such as Windows
ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc..
Described in above method embodiment can be based on the service shown in the Fig. 8 as the step performed by application server
The structure of device.
The embodiment of the present invention also provides a kind of storage device, a plurality of instruction of storage device stored, and the instruction is suitable to
Loaded as processor and perform the clustering objects method as performed by above-mentioned application server.
The embodiment of the present invention also provides a kind of server, including processor and storage device, the processor, for realizing
Each instruction；
The storage device is used to store a plurality of instruction, and described instruct is applied as described above for being loaded by processor and being performed
Clustering objects method performed by server.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
To instruct the hardware of correlation to complete by program, the program can be stored in a computer-readable recording medium, storage
Medium can include：Read-only storage (ROM), random access memory ram), disk or CD etc..
The clustering objects method, apparatus and storage device provided above the embodiment of the present invention is described in detail,
Specific case used herein is set forth to the principle and embodiment of the present invention, and the explanation of above example is simply used
Understand the method and its core concept of the present invention in help；Meanwhile for those of ordinary skill in the art, according to the present invention's
Thought, there will be changes in specific embodiments and applications, in summary, this specification content should not be construed as
Limitation of the present invention.
Claims (14)
- A kind of 1. clustering objects method, it is characterised in that including：It is determined that at least one second object group associated with the first object group, wherein, any object group includes at least one Object；The first object group relating value between at least one second object group respectively is determined, obtains at least one association Value；The cluster degree according to corresponding to the function calculating formula of at least one relating value and preset cluster degree calculates respectively, is obtained At least one cluster degree；It is if the first cluster degree meets preset first condition at least one cluster degree, the first cluster degree is corresponding The second object group and the first object group cluster into same target group.
- 2. the method as described in claim 1, it is characterised in that determine the first object group and some second object group it Between relating value, specifically include：The first object in the first object group is determined, between the second object in a certain second object group, based at least The relationship score of one dimension；First object and the second object are associated objects；If first object and the second object are all one, will be based at least between first object and the second object The mathematical calculation of the relationship score of one dimension is as the relating value between the first object group and the second object group.
- 3. method as claimed in claim 2, it is characterised in that determine the first object group and some second object group it Between relating value, in addition to：If first object is one, the second object to be multiple, then calculate first object respectively with multiple second pairs The mathematical calculation of relationship score based at least one dimension as between, obtains multiple mathematical calculations, by the multiple number Learn relating value of the calculated value sum as the first object group and the second object group；If first object is multiple, the second object is multiple, then calculate respectively each the in the multiple first object The mathematical calculation of the relationship score based at least one dimension, obtains more numbers between one object and the second associated object Calculated value is learned, the relating value using the multiple mathematical calculation sum as the first object group and the second object group.
- 4. method as claimed in claim 2, it is characterised in thatIf the object is user, first object is the first user, and the second object is second user, described at least one Dimension includes：The information that first user and second user are operated by applications client respectively, then it is described to be based at least one The relationship score of individual dimension specifically includes：The first information that first user is operated by applications client, with second user Identical information bar number in the second information operated by applications client, divided by the information of the first information and the second information Bar number sum, obtained quotient.
- 5. the method as described in claim 1, it is characterised in that the function calculating formula of the preset cluster degree includes：Wherein, ∑inFor the relating value sum between object in the first object group, a certain second pair As the relating value sum between object in group, after relating value is added between the first object group and a certain second object group Value；The ∑totFor the pass in the first object group and a certain second object group between object, and the object of other object groups Connection value sum；The m includes the relating value sum between all objects.
- 6. the method as described in any one of claim 1 to 5, it is characterised in that the first cluster at least one cluster degree Degree meets preset first condition, specifically includes：If at least one cluster degree includes a cluster degree, integrally degree of cluster is more than and worked as corresponding to the first cluster degree Preceding overall cluster degree, the current overall cluster degree is that the second object group corresponding to the first cluster degree exists with the first object group The cluster degree sum of all object groups before cluster, integrally degree of cluster is the first cluster degree corresponding to the first cluster degree The cluster degree sum of corresponding second object group and the first object group all object groups after cluster；If at least one cluster degree includes multiple cluster degree, integrally degree of cluster is more than institute corresponding to the first cluster degree State current overall cluster degree, and corresponding to the first cluster degree integrally degree of cluster more than removing described the in the multiple cluster degree Overall degree of cluster corresponding to other cluster degree outside one cluster degree.
- 7. the method as described in any one of Claims 1-4, it is characterised in that methods described also includes：If the object group after the cluster is unsatisfactory for preset second condition, for described in the object group execution after the cluster The second object group is determined, determines relating value, the step of calculating cluster degree and cluster.
- 8. the method as described in any one of Claims 1-4, it is characterised in that if the object is user, object group is use Family group, the first object are the first user, and methods described also includes：The information that other users in user's group after the cluster in addition to the first user are operated, and/or user profile hair Give the applications client of first user.
- A kind of 9. clustering objects device, it is characterised in that including：Object group determining unit, for determining at least one second object group associated with the first object group, wherein, it is any right As group includes at least one object；Determining unit is associated, for determining the first object group relating value between at least one second object group respectively, Obtain at least one relating value；Computing unit, for being calculated respectively correspondingly according to the function calculating formula of at least one relating value and preset cluster degree Cluster degree, obtain at least one cluster degree；Cluster cell, if meeting preset first condition for the first cluster degree at least one cluster degree, by described in Second object group corresponding to first cluster degree is with the first object group cluster into same target group.
- 10. device as claimed in claim 9, it is characterised in thatThe association determining unit, specifically for it is determined that associating between the first object group and some second object group During value, the first object in the first object group is determined, between the second object in a certain second object group, based at least The relationship score of one dimension；First object and the second object are associated objects；If first object and Two objects are all one, then by the mathematics of the relationship score based at least one dimension between first object and the second object Calculated value is as the relating value between the first object group and the second object group.
- 11. the device as described in claim 9 or 10, it is characterised in thatThe cluster cell, if including a cluster degree, the first cluster degree specifically at least one cluster degree Corresponding overall cluster degree is more than current overall cluster degree, and the current entirety cluster degree is the corresponding to the first cluster degree The cluster degree sum of two object groups and the first object group all object groups before cluster, entirety corresponding to the first cluster degree Cluster degree is the cluster of the second object group corresponding to the first cluster degree and the first object group all object groups after cluster Spend sum；If at least one cluster degree includes multiple cluster degree, integrally degree of cluster is big corresponding to the first cluster degree In the current overall cluster degree, and corresponding to the first cluster degree integrally degree of cluster more than removing institute in the multiple cluster degree State corresponding to other cluster degree outside the first cluster degree integrally degree of cluster.
- 12. the device as described in claim 9 or 10, it is characterised in that also include：Information transmitting unit, if being user for the object, the first object is the first user, and object group is user's group, will The information of other users operation in user's group after the cluster in addition to the first user, and/or user profile are sent to institute State the applications client of the first user.
- 13. a kind of storage device, it is characterised in that a plurality of instruction of storage device stored, the instruction are suitable to by processor Load and perform the clustering objects method as described in any one of claim 1 to 8.
- A kind of 14. server, it is characterised in that including processor and storage device, the processor, for realizing each finger Order；The storage device is used to store a plurality of instruction, described to instruct for being loaded by processor and performing such as claim 1 to 8 Clustering objects method described in any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711014564.0A CN107886112B (en) | 2017-10-26 | 2017-10-26 | Object clustering method and device and storage equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711014564.0A CN107886112B (en) | 2017-10-26 | 2017-10-26 | Object clustering method and device and storage equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107886112A true CN107886112A (en) | 2018-04-06 |
| CN107886112B CN107886112B (en) | 2020-09-08 |
Family
ID=61782447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711014564.0A Active CN107886112B (en) | 2017-10-26 | 2017-10-26 | Object clustering method and device and storage equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107886112B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111324518A (en) * | 2020-02-03 | 2020-06-23 | 中国银联股份有限公司 | Application association method and device |
| CN113221016A (en) * | 2021-07-08 | 2021-08-06 | 北京达佳互联信息技术有限公司 | Resource recommendation method and device, computer equipment and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103971256A (en) * | 2013-01-25 | 2014-08-06 | 阿里巴巴集团控股有限公司 | Information push method and device |
| US20150039619A1 (en) * | 2012-03-19 | 2015-02-05 | Microsoft Corporation | Grouping documents and data objects via multi-center canopy clustering |
| CN104598449A (en) * | 2013-10-30 | 2015-05-06 | Sap欧洲公司 | Preference-based clustering |
| CN106095843A (en) * | 2016-06-02 | 2016-11-09 | 腾讯科技（深圳）有限公司 | Social account method for digging and device |
-
2017
- 2017-10-26 CN CN201711014564.0A patent/CN107886112B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150039619A1 (en) * | 2012-03-19 | 2015-02-05 | Microsoft Corporation | Grouping documents and data objects via multi-center canopy clustering |
| CN103971256A (en) * | 2013-01-25 | 2014-08-06 | 阿里巴巴集团控股有限公司 | Information push method and device |
| CN104598449A (en) * | 2013-10-30 | 2015-05-06 | Sap欧洲公司 | Preference-based clustering |
| CN106095843A (en) * | 2016-06-02 | 2016-11-09 | 腾讯科技（深圳）有限公司 | Social account method for digging and device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111324518A (en) * | 2020-02-03 | 2020-06-23 | 中国银联股份有限公司 | Application association method and device |
| CN111324518B (en) * | 2020-02-03 | 2024-05-03 | 中国银联股份有限公司 | Application association method and device |
| CN113221016A (en) * | 2021-07-08 | 2021-08-06 | 北京达佳互联信息技术有限公司 | Resource recommendation method and device, computer equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107886112B (en) | 2020-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108229555B (en) | Sample weights distribution method, model training method, electronic equipment and storage medium | |
| CN104778173B (en) | Target user determination method, device and equipment | |
| CN110266745B (en) | Information flow recommendation method, device, equipment and storage medium based on deep network | |
| CN110213164B (en) | Method and device for identifying network key propagator based on topology information fusion | |
| CN109299420A (en) | Social media account processing method, device, equipment and readable storage medium storing program for executing | |
| KR101356948B1 (en) | User interest inference method and system in sns using topics on social activities with neighbors | |
| CN107229966A (en) | A kind of model data update method, apparatus and system | |
| CN108985954A (en) | A kind of method and relevant device of incidence relation that establishing each mark | |
| CN110599312A (en) | Credit-based interactive credit evaluation method and device | |
| CN106789338B (en) | Method for discovering key people in dynamic large-scale social network | |
| CN110008999A (en) | Determination method, apparatus, storage medium and the electronic device of target account number | |
| CN111639969B (en) | Dynamic incentive calculation method, system, equipment and medium for crowdsourcing system | |
| CN107886112A (en) | A kind of clustering objects method, apparatus and storage device | |
| US20070198252A1 (en) | Optimum design management apparatus, optimum design calculation system, optimum design management method, and optimum design management program | |
| US11468521B2 (en) | Social media account filtering method and apparatus | |
| CN105096161B (en) | It is a kind of enter row information displaying method and apparatus | |
| CN104537011B (en) | Sing arrangement method and device | |
| US20210291046A1 (en) | Providing Video Game Content to an Online Connected Game | |
| CN110505276B (en) | Object matching method, device and system, electronic equipment and storage medium | |
| CN110222297B (en) | Identification method of tag user and related equipment | |
| CN106960397A (en) | Recommend the method and device of college entrance examination universities and colleges | |
| CN106844504B (en) | A kind of method and apparatus for sending song and singly identifying | |
| CN110035126A (en) | A kind of document handling method, calculates equipment and storage medium at device | |
| CN111429949B (en) | Pitch line generation method, device, equipment and storage medium | |
| CN106301880B (en) | Method and equipment for determining network relationship stability and recommending internet service |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |