CN105844577A - Relation network recognition method and device - Google Patents
Relation network recognition method and device Download PDFInfo
- Publication number
- CN105844577A CN105844577A CN201510015153.8A CN201510015153A CN105844577A CN 105844577 A CN105844577 A CN 105844577A CN 201510015153 A CN201510015153 A CN 201510015153A CN 105844577 A CN105844577 A CN 105844577A
- Authority
- CN
- China
- Prior art keywords
- node
- relationship
- information
- particular kind
- relational network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a relation network recognition method, and the method is used for providing a feasible scheme for precisely recognizing a relation network. The method comprises the steps: determining the ownership information of a relation network of a node set; and recognizing the relation network from the node set according to the ownership information of the relation network. The invention also discloses a relation network recognition device.
Description
Technical field
The application relates to Internet technical field, particularly relates to recognition methods and the device of a kind of relational network.
Background technology
Relational network, refers to the node set being made up of the node that there is at least one particular kind of relationship each other.
Node mentioned here, can be such as user, or the logic generated according to the relation between user is closed
It it is the node in figure.Such as, there is user's set that the user of " Peer Relationships " is constituted each other, just
It is a kind of relational network, and this relational network may be generally referred to as " Peer Relationships circle "；The most such as, each other it
Between there is user's set that the user of " classmate's relation " is constituted, be also a kind of relational network, can be described as " with
Learn relation loop ".It is similar to, it is also possible to have " relationship circle ", " friends circle ", etc..
The biggest some identical information may be liked (such as commodity owing to the user in same relation network has
Information), or hobby does some identical user behavior (such as buy same finance product), therefore,
The accurate identification of relational network, the most great for carrying out information recommendation meaning accurately for relational network.
But, prior art does not the most provide the feasible program for precisely identifying relational network.
Summary of the invention
The embodiment of the present application provides the recognition methods of a kind of relational network, is used for providing a kind of precisely identification relation
The feasible program of network.
The embodiment of the present application also provides for the identification device of a kind of relational network, is used for providing one precisely to identify pass
It it is the feasible program of network.
The embodiment of the present application employing following technical proposals:
The recognition methods of a kind of relational network, comprises determining that the relational network attaching information of node set；Root
According to described relational network attaching information, from described node set, identify relational network；
Wherein, the relational network attaching information of described node set includes: each node in described node set
The relational network attaching information of each pair of node in relational network attaching information, and/or described node set；
The relational network attaching information of individual node, including: represent and there is particular kind of relationship with described individual node
The information of other nodes particular kind of relationship degree of strength each other；
The relational network attaching information of single pair of node, including: represent the first set and the second set similarity degree
Information；First collection is combined into: with this, primary nodal point in node is existed particular kind of relationship in described node set
Node mark constitute set；Second collection be combined into: in described node set with this to second in node
There is the set that the mark of the node of particular kind of relationship is constituted in node.
The identification device of a kind of relational network, comprises determining that information unit, for determining the pass of node set
It is network home information；Identify relation unit, for according to described relational network attaching information, from described joint
Point set identifies relational network；
Wherein, the relational network attaching information of described node set includes: each node in described node set
The relational network attaching information of each pair of node in relational network attaching information, and/or described node set；
The relational network attaching information of individual node, including: represent and there is particular kind of relationship with described individual node
The information of other nodes particular kind of relationship degree of strength each other；
The relational network attaching information of single pair of node, including: represent the first set and the second set similarity degree
Information；First collection is combined into: with this, primary nodal point in node is existed particular kind of relationship in described node set
Node mark constitute set；Second collection be combined into: in described node set with this to second in node
There is the set that the mark of the node of particular kind of relationship is constituted in node.
At least one technical scheme above-mentioned that the embodiment of the present application uses can reach following beneficial effect:
Automatically can identify from node set according to the relational network attaching information of the node set determined
Relational network, thus provide the feasible program of a kind of accurate identification relational network.
Accompanying drawing explanation
Accompanying drawing described herein is used for providing further understanding of the present application, constitutes the part of the application,
The schematic description and description of the application is used for explaining the application, is not intended that the improper limit to the application
Fixed.In the accompanying drawings:
The flow process that implements of the recognition methods of a kind of relational network that Fig. 1-1 provides for the embodiment of the present application is shown
It is intended to；
The flow process that realizes identifying relational network from node set that Fig. 1-2 provides for the embodiment of the present application is illustrated
Figure；
The flow process that realizes identifying relational network from node set that Fig. 1-3 provides for the embodiment of the present application is illustrated
Figure；
The flow process that realizes identifying relational network from node set that Fig. 1-4 provides for the embodiment of the present application is illustrated
Figure；
Fig. 1-5 identifies occur that the flow process that realizes of concussion is shown during relational network for what the embodiment of the present application provided
It is intended to；
The concrete structure schematic diagram identifying device of a kind of relational network that Fig. 2 provides for the embodiment of the present application.
Detailed description of the invention
For making the purpose of the application, technical scheme and advantage clearer, specifically real below in conjunction with the application
Execute example and technical scheme is clearly and completely described by corresponding accompanying drawing.Obviously, described
Embodiment is only some embodiments of the present application rather than whole embodiments.Based on the enforcement in the application
Example, the every other enforcement that those of ordinary skill in the art are obtained under not making creative work premise
Example, broadly falls into the scope of the application protection.
Below in conjunction with accompanying drawing, describe the technical scheme that each embodiment of the application provides in detail.
Embodiment 1
Embodiment 1 provides the recognition methods of a kind of relational network, is used for providing one precisely to identify network of personal connections
The feasible program of network.The idiographic flow schematic diagram of the method is as it is shown in figure 1, comprise the steps:
Step 11, determines the relational network attaching information of node set.
Wherein, the relational network attaching information of node set includes: the relational network of each node in node set
The relational network attaching information of each pair of node in attaching information, and/or node set.
For the relational network attaching information of individual node, this information may include that expression and single joint
There is the information of other nodes particular kind of relationship degree of strength each other of particular kind of relationship in point.Such as, single joint
The relational network attaching information of point can be a numerical value, is used for representing relation power between other node
Degree.
If representing in this node set the relation between each node with logic relation picture, so that logical relation to connect
The line (limit) of different nodes represents and there is particular kind of relationship, then, there is particular kind of relationship with individual node
Other nodes particular kind of relationship degree of strength each other, can be presented as that in logic relation picture individual node coagulates
The size of the ability of poly-surroundings nodes.Line in logic relation picture, between individual node and neighbor node
(limit) is the most, represents that the ability of node cohesion surroundings nodes is the strongest, represents this individual node and neighbor node
Node corresponding respectively more may belong to a relational network together.
For the relational network attaching information of single pair of node, this information may include that expression the first set
Information with the second set similarity degree.Wherein, first collection be combined into: in node set with this in node
There is the set that the mark of the node of particular kind of relationship is constituted in primary nodal point；Second collection be combined into: in node set with
The set that the mark of this node that secondary nodal point in node exists particular kind of relationship is constituted.
Such as, it is assumed that node set has node A and node B there is particular kind of relationship each other this pair
Node；In this node set, with the presence of node B, node C, node D, node E respectively with node A
Particular kind of relationship；With the presence of node A, node C, node E and node F respectively with node B particular kind of relationship.
So, if node A be this to " primary nodal point " in node, node B is that this is to " second in node
Node ", then the first set can be that { B, C, D, E}, the second set can be " A, C, E, F ".
If the number of mark identical in gathering with second with the first set, as this, relational network of node is belonged to
Information, then this is " 2 " to the relational network attaching information of node.Owing to this number is the biggest, node is described
" social relations " similarity of A and node B is the biggest, thus illustrates that node A and node B more may belong to
In same relation network.
Step 12, according to relational network attaching information, identifies relational network from node set.
Use the method that embodiment 1 provides, can belong to according to the relational network of the node set determined
Information, automatically identifies relational network from node set, thus provides a kind of accurate identification relational network
Feasible program.
Some optional embodiments of step 11 and step 12 are discussed in detail individually below:
It is in one embodiment, more reliable in order to reach to determine the relational network attaching information of node set,
And then make the recognition result effect more accurately of relational network, can before performing step 11, according to
Particular kind of relationship information, chooses node configuration node set from start node set.
Wherein, start node set refers to the set that all nodes of belonging relation network to be identified are constituted；Joint
Point set, refers to according to particular kind of relationship information from all nodes of relational network to be identified, choose all or
Part is relatively reliable, it is more likely that the set that the node in relation belonging to network is constituted；Particular kind of relationship information is used
The degree of strength of particular kind of relationship between different nodes in expression, start node set.
Such as, as a example by the start node set that 13 employees of certain company are constituted, if particular kind of relationship information
Including one or more in this information: the similarity degree of shopping online ship-to, electronic account fund are past
The number of times come, and the MAC (Media Access Control, medium access control) of Office Area wireless aps
Address similarity degree, then, all use ship-to one if any in 13 employee hoc annos during net purchase；
10 employees had the record of treasury trade, 3 employee's no funds contact records between everyone；Permissible
Determine the particular kind of relationship having stronger degree between 10 employees, it is more likely that belong to a certain relational network,
And 3 employees may be not belonging to certain relational network, then can be formed node set by these 10 employees.
In one embodiment, according to particular kind of relationship information, from start node set, choose node constitute
Before node set, particular kind of relationship information can be determined in the following manner:
Perform respectively for every pair of node in start node set:
According to this similarity to node region respectively and equal to node with this in start node set
There is the number of the node of particular kind of relationship, determine this particular kind of relationship information to node.
It should be noted that this similarity to node respectively region mentioned here, can but do not limit
In including: the similarity of this hardware device respectively region corresponding to node；The terminal that node is corresponding is divided
The similarity of other region；The similarity etc. of certain system region respectively that node is corresponding.
Specifically,
The similarity of certain a pair node region respectively, can include one or more in following similarity:
The similarity of this ship-to that node is used respectively；The wireless network that node was respectively connected to by this accesses
The similarity of the physical address of point；This similarity to node used IP address respectively.
Such as, the similarity of ship-to that node A and node B uses respectively is determined by semantic analysis；
Determine the similarity of the wireless network access point that node A and node B was respectively connected to；And determine node A
Similarity with node B used IP address respectively.
The node of particular kind of relationship is all there is to node in start node set with this, including: start node set
In, be there is the first relation in a node in node with this, with this, another node in node is existed
The node of the second relation；
Wherein, first relation that exists can be, but not limited to include:
There is e-bank's treasury trade, interbehavior occurred in social networks, and/or each used
Electronic account bound same medium.
Second relation that exists can be, but not limited to include:
There is e-bank's treasury trade, interbehavior occurred in social networks, and/or each used
Electronic account bound same medium.
Such as, e-bank's treasury trade can be to transfer accounts, pay out the behaviors such as money；Mutual in social networks
Behavior can be concern behavior, forwards, comments on, praise；Same medium can be, but not limited to be same MAC
Address, same phone number, same mobile phone IMEI (International Mobile Equipment Identity
Number, international mobile device identification code).
Such as, node A (being called for short A) and node B (being called for short B) is a pair node, and node C (is called for short
C) there is the first relation and the second relation respectively with A and B, including: C respectively and between A and B all
Occurring the behavior of transferring accounts, C to use identical ship-to etc. respectively the most respectively with A and B, C is respectively
The message registration occurring in A and B is medium.It should be noted that the first relation and the second relation can be distinguished
Comprising different contents, such as, C had electronic account treasury trade respectively and paid close attention to mutually with A and B,
The node of particular kind of relationship can also be all there is to node in so C as this.
Above-mentioned first relation and the second relation, can be seen as one " direct correlation relation ".
In one embodiment, for every pair of node in start node set, start node set
In, with this, node all existed the node of described particular kind of relationship, it is also possible to include and this node is existed "
Connect incidence relation " node.Such as, in start node set, node is distinguished place with this by region
The Similarity value in region is above the node of similarity threshold, it is possible to be considered as with this that node is existed this is indirect
The node of incidence relation.
In the embodiment of the present application, the similarity of different node regions, including following similarity one or
Multiple:
The similarity of the ship-to that different nodes use respectively；
The similarity of the physical address of the wireless network access point that different nodes were respectively connected to；
The similarity of different nodes used IP address respectively.
In one embodiment, reach from just to reach, by one, particular kind of relationship information is done quantization
Beginning node set is chosen the effect of node configuration node set, can be realized by following manner:
Particular kind of relationship information according to every pair of node, and the relation degree of strength threshold value set, from start node
In set, choose each higher than the degree of strength represented by relation degree of strength threshold value of the degree of strength of particular kind of relationship
To node, configuration node set.
Specifically, the particular kind of relationship information of every pair of node can be calculated by following formula:
Score=W (work address)+W (work WIFI MAC)+W (work IP)+W (network
index)
Wherein,
Score is the relational network score value of a pair node, this score value i.e. this particular kind of relationship information to node；
W represents the method for transformation of index score value；Index may refer to similarity etc.；
Work address refers to that a period of time interior (such as one month) a pair node uses the phase of ship-to
Like degree.Such as, mono-" ship-to one " used the middle of the month of node A is made up of 10 characters, node
Mono-" ship-to two " used the middle of the month of B is made up of 8 characters, solves the two ship-to
Analysis, finds that they have 6 continuous print characters to be identical, is all " C, West, Xidan ", then can count
The Work address similarity calculating node A and node B is 6/ [(10+8)/2]=0.67；
Work WIFI MAC refers to the WIFI that a period of time interior (such as one month) a pair node uses
The similarity of MAC Address.Such as, node A is connected to router corresponding to MAC1 mono-middle of the month, joint
Point B is also connected to router corresponding to MAC1 mono-middle of the month, then the work WIFI of node A and node B
MAC similarity is 1；
Work IP refers to the similarity of the IP address of a pair node use of a period of time interior (such as one month).
Such as, if node A uses IP1 (192.168.140.2), node B to use IP2 mono-middle of the month mono-middle of the month
(192.168.140.20), then, it may be determined that the IP address that the two node uses is at the same network segment.If
Prespecified, the similarity of the IP address belonging to the same network segment together is 80%, then node A and node B
Work IP similarity is 0.8；
In network index refers to a period of time all there is particular kind of relationship with a pair node in (such as one month)
The number of node.Such as, it is " e-bank's treasury trade occurred " with the first relation and the second relation
As a example by, a middle of the month, have 10 nodes and node A that electronic account treasury trade occurred, and this 10
In individual node, 8 nodes and node B is had electronic account treasury trade to occur, then with node A and joint
It is 8 that some B exists the interstitial content of direct correlation relation.With the similarity of different node regions for " no
Similarity with the ship-to that node uses respectively " as a example by, if within a period of time (such as one month),
100 nodes and node A is had to be used in conjunction with in ship-to one, and these 100 nodes having 80 joints
Point shared router corresponding to mac1 with node B, then there is indirect association with node A and node B and close
The interstitial content of system is 80.Hence, it can be determined that go out network index=8+80=88.
After obtaining all indexs, use logistic regression algorithm, value corresponding for each index is converted into [0,1]
Score value, using each score value and value as Score.
Based on the relation degree of strength threshold value set, Score is more than each pair of i.e. structure of node of relation degree of strength threshold value
Become node set.
Such as, relation degree of strength threshold value can be set as 2.5, when by the particular kind of relationship information of a pair node
Calculating this to the node doubtful colleague score value of node is 2.8, then this be saved in node in node set.
For for step 11,
In one embodiment, include when the relational network attaching information of node set: each in node set
During the relational network attaching information of node, determine the relational network attaching information of node set, including:
According to the particular kind of relationship of each node, generate logic relation picture；Wherein, logic relation picture comprises each joint
Point；In logic relation picture, each bar limit represents and there is particular kind of relationship between the node at two ends, limit；
Wherein, each node refers to each node of the configuration node set chosen from start node set, specific
Relation refers to exist between each node stronger relation；Concrete stronger relation can be according to calculating every pair of joint above
The value of some Score determines.
Perform for each node:
The number of the adjacent node according to this node, and the number on limit between the adjacent node of this node, meter
Calculate the weighted value of this node, as the relational network attaching information of the individual node that this node represents.
Specifically,
For each node, calculate the weighted value of this node according to the following equation:
Wherein, E represents the quantity on the limit existed between the adjacent node of this node, and V represents this node
The quantity of adjacent node.Wherein, the adjacent node of this node, refer to exist the node on limit between this node.
For Web Community, node weights typically represents the ability of this node cohesion surrounding neighbours node,
Therefore can use node weights, weigh the cohesion of node.
In one embodiment, it is also possible to calculate the weighted value of each node according to the following equation:
In one embodiment, include when the relational network attaching information of node set: each in node set
During to the relational network attaching information of node, determine the relational network attaching information of node set, including:
According to the particular kind of relationship of each node, generate logic relation picture；Wherein, logic relation picture comprises each joint
Point；In logic relation picture, each bar limit represents and there is particular kind of relationship between the node at two ends, limit；
Perform for each edge:
Between other each nodes respectively and in addition to this is to node of a pair node according to these two ends, limit the most equal
There is limit, calculate the weighted value on this limit, the relational network as this single pair of node corresponding to node belongs to
Information.
Specifically,
Can calculate with following formula for each edge:
Wherein,
SijWeighted value for the limit between node i and node j；
K is node serial number；
N is the total number of logic relation picture interior joint；
A represents the two internodal situations that there is limit: when there is not limit between i, k, Aik=0, as i, k
Between when there is limit, Aik=1.Similarly, when there is not limit between j, k, Akj=0, between j, k
When there is limit, Akj=1.It should be noted that: (Aii-Aij)2=1, (Aij-Ajj)2=1.It is visible,
DijMinimum value be
Such as, can be by two formula in step 11, the relational network calculating certain node set is returned
Genus information, this information can be expressed as such as the original state of Fig. 1-2 interior joint set.
For for step 12,
Label propagation algorithm can be utilized, identify relational network.
Identification step is as follows:
The weight on node and limit, according to the result of calculation of step 11, is attached on corresponding node and limit, point
The each node of dispensing one initializes ID.
0th iteration, as shown in Figure 1-2.Each node is by own node ID, node weights, the power on limit
It is resent to neighbor node.Such as: node 2 is sent to the information format of node 1 and is: own node ID#
Node weights * limit weight: 2#0.49.
1st iteration, as Figure 1-3.Each node can inquire about all node ID received and
Corresponding weight, and the node ID of oneself is updated to the node ID that the weight maximum that receives is corresponding；
If node ID corresponding to weight limit value is unique, then take the node ID character string sorting of correspondence minimum.
Such as: initial node 2, the information receiving neighbor node is: node ID 3, the weighted value of ID1 are
0.49, the weighted value of node ID 4 is 0.3, then the node ID of start node 2 is updated to 1.
2nd iteration, as Figure 1-4.Similar with the 1st iteration, after iteration renewal, the left side is complete
Be updated to 1, the right be updated to 5 completely.
3rd iteration, as Figure 1-4.Owing to not having node ID to change, iteration terminates, logical
Cross 3 iteration, it is possible to perfectly mark off two relational networks.
It should be noted that for this algorithm, it may appear that such as the situation of Fig. 1-5:
After 3 iteration (t-1, t, t+1), two relational networks have divided, but after t+1 again
After t+n iteration, still can not stablize, node ID situation alternately, this feelings can be repeated
Condition is referred to as concussion occur.Solution for this kind of situation is: in iterative process, record each node
The node ID of nearest 10 times, if occur node ID alternately (in the 8th～10 iteration, a part of node
ID be followed successively by 2,3,2, the ID of another part node is followed successively by 3,2,3) situation, then it is assumed that
Vibration occurs, now stops iteration.
In one embodiment, in order to reach to understand the accuracy of the relational network identified, permissible
After identifying relational network, recognition result is estimated.
Following formula can be passed through, recognition result is estimated.
Wherein, m represents the relational network number marked off；
lpRepresent that being positioned at relational network in logic relation picture is designated the limit number of p；
L represents limit number total in relational network；
dpRepresent the degree sum that relational network is all nodes in p, dp=2 × lp+ a certain relational network identifies
For nodes all in p limit number to other relational network；
The codomain-1 < Q < 1 of Q, Q-value is the biggest, shows that relational network divides the most accurate.
It should be noted that the executive agent of each step of the provided method of embodiment 1 may each be same and sets
Standby, or, the method also by distinct device as executive agent.Such as, the holding of step 11 and step 12
Row main body can be equipment 1.The executive agent of the most such as step 11 can be equipment 1, holding of step 12
Row main body can be equipment 2；Etc..
Embodiment 2
Based on identical inventive concept, embodiment 2 provides the identification device of a kind of relational network, is used for carrying
Feasible program for a kind of accurate identification relational network.As in figure 2 it is shown, the identification device bag of this relational network
Include:
Determine information unit 21, be determined for the relational network attaching information of node set；
Identify relation unit 22, may be used for according to relational network attaching information, identify from node set and close
It it is network；
Wherein, the relational network attaching information of node set includes: the relational network of each node in node set
The relational network attaching information of each pair of node in attaching information, and/or node set；
The relational network attaching information of individual node, including: there is its of particular kind of relationship in expression and individual node
The information of his node particular kind of relationship degree of strength each other；
The relational network attaching information of single pair of node, including: represent the first set and the second set similarity degree
Information；First collection is combined into: with this, primary nodal point in node is existed particular kind of relationship in described node set
Node mark constitute set；Second collection be combined into: in described node set with this to second in node
There is the set that the mark of the node of particular kind of relationship is constituted in node.
In one embodiment, device also includes: choose aggregation units 23, may be used for execution and determines letter
Before interest statement unit 21,
According to particular kind of relationship information, from start node set, choose node configuration node set；
Wherein, particular kind of relationship information is used for representing: particular kind of relationship between the different nodes in start node set
Degree of strength.
In one embodiment, choose aggregation units 23, may be used for according to particular kind of relationship information, from just
Before beginning node set is chosen node configuration node set, divide for every pair of node in start node set
Not Zhi Hang:
According to this similarity to node region respectively and equal to node with this in start node set
There is the number of the node of particular kind of relationship, determine this particular kind of relationship information to node.
In one embodiment, choose aggregation units 23, may be used for:
Particular kind of relationship information according to every pair of node, and the relation degree of strength threshold value set, from start node
In set, choose each higher than the degree of strength represented by relation degree of strength threshold value of the degree of strength of particular kind of relationship
To node, configuration node set.
In one embodiment, this similarity to node region respectively, including in following similarity
One or more: the similarity of this ship-to that node is used respectively；Node was respectively connected to by this
The similarity of physical address of wireless network access point；This phase to node used IP address respectively
Like degree.
In one embodiment, all there is the joint of described particular kind of relationship to node with this in start node set
Point, including:
In start node set, be there is the first relation in a node in node with this, with this in node
Another node there is the node of the second relation；
Wherein, first relation that exists includes: e-bank's treasury trade occurred, occurs in social networks
Cross interbehavior, and/or the electronic account each used bound same medium.
Second relation that exists includes: occurred e-bank's treasury trade, occurred in social networks mutual
Behavior, and/or the electronic account each used bound same medium.
In one embodiment, include when the relational network attaching information of node set: each in node set
During the relational network attaching information of node, determine information unit 21, may be used for:
According to the particular kind of relationship of each node, generate logic relation picture；Wherein, logic relation picture includes each joint
Point；In logic relation picture, each bar limit represents and there is particular kind of relationship between the node at two ends, limit；For each
Node performs:
The number of the adjacent node according to this node, and the number on limit between the adjacent node of this node, meter
Calculate the weighted value of this node, as the relational network attaching information of the individual node that this node represents.
In one embodiment, include when the relational network attaching information of node set: each in node set
During to the relational network attaching information of node, determine information unit 21, may be used for:
According to the particular kind of relationship of each node, generate logic relation picture；Wherein, logic relation picture comprises each joint
Point；In logic relation picture, each bar limit represents and there is particular kind of relationship between the node at two ends, limit；For every
Limit performs:
Between other each nodes respectively and in addition to this is to node of a pair node according to these two ends, limit the most equal
There is limit, calculate the weighted value on this limit, the relational network as this single pair of node corresponding to node belongs to
Information.
In one embodiment, device also includes: assessment result unit 24, may be used for from node set
After middle identification relational network, recognition result is estimated.
Use this device that embodiment 2 provides, can belong to according to the relational network of the node set determined
Information, automatically identifies relational network from node set, thus provides a kind of accurate identification relational network
Feasible program.
Those skilled in the art are it should be appreciated that embodiments of the invention can be provided as method, system or meter
Calculation machine program product.Therefore, the present invention can use complete hardware embodiment, complete software implementation or knot
The form of the embodiment in terms of conjunction software and hardware.And, the present invention can use and wherein wrap one or more
Computer-usable storage medium containing computer usable program code (include but not limited to disk memory,
CD-ROM, optical memory etc.) form of the upper computer program implemented.
The present invention is with reference to method, equipment (system) and computer program product according to embodiments of the present invention
The flow chart of product and/or block diagram describe.It should be understood that can by computer program instructions flowchart and
/ or block diagram in each flow process and/or flow process in square frame and flow chart and/or block diagram and/
Or the combination of square frame.These computer program instructions can be provided to general purpose computer, special-purpose computer, embedding
The processor of formula datatron or other programmable data processing device is to produce a machine so that by calculating
The instruction that the processor of machine or other programmable data processing device performs produces for realizing at flow chart one
The device of the function specified in individual flow process or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions may be alternatively stored in and computer or the process of other programmable datas can be guided to set
In the standby computer-readable memory worked in a specific way so that be stored in this computer-readable memory
Instruction produce and include the manufacture of command device, this command device realizes in one flow process or multiple of flow chart
The function specified in flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, makes
Sequence of operations step must be performed to produce computer implemented place on computer or other programmable devices
Reason, thus the instruction performed on computer or other programmable devices provides for realizing flow chart one
The step of the function specified in flow process or multiple flow process and/or one square frame of block diagram or multiple square frame.
In a typical configuration, calculating equipment includes one or more processor (CPU), input/defeated
Outgoing interface, network interface and internal memory.
Internal memory potentially includes the volatile memory in computer-readable medium, random access memory
(RAM) and/or the form such as Nonvolatile memory, such as read only memory (ROM) or flash memory (flash RAM).
Internal memory is the example of computer-readable medium.
Computer-readable medium includes that removable media permanent and non-permanent, removable and non-can be by appointing
Where method or technology realize information storage.Information can be computer-readable instruction, data structure, program
Module or other data.The example of the storage medium of computer includes, but are not limited to phase transition internal memory
(PRAM), static RAM (SRAM), dynamic random access memory (DRAM), its
The random access memory (RAM) of his type, read only memory (ROM), electrically erasable are read-only
Memorizer (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read only memory
(CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tape, tape magnetic
Disk storage or other magnetic storage apparatus or any other non-transmission medium, can be used for storage can be calculated
The information that equipment accesses.According to defining herein, computer-readable medium does not include temporary computer-readable matchmaker
Body (transitory media), such as data signal and the carrier wave of modulation.
Also, it should be noted term " includes ", " comprising " or its any other variant are intended to non-
Comprising of exclusiveness, so that include that the process of a series of key element, method, commodity or equipment not only wrap
Include those key elements, but also include other key elements being not expressly set out, or also include for this process,
The key element that method, commodity or equipment are intrinsic.In the case of there is no more restriction, statement " include
One ... " key element that limits, it is not excluded that including the process of described key element, method, commodity or setting
Other identical element is there is also in Bei.
It will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer journey
Sequence product.Therefore, the application can use complete hardware embodiment, complete software implementation or combine software and
The form of the embodiment of hardware aspect.And, the application can use and wherein include calculating one or more
The computer-usable storage medium of machine usable program code (include but not limited to disk memory, CD-ROM,
Optical memory etc.) form of the upper computer program implemented.
The foregoing is only embodiments herein, be not limited to the application.For this area skill
For art personnel, the application can have various modifications and variations.All institutes within spirit herein and principle
Any modification, equivalent substitution and improvement etc. made, within the scope of should be included in claims hereof.
Claims (20)
1. the recognition methods of a relational network, it is characterised in that including:
Determine the relational network attaching information of node set；
According to described relational network attaching information, from described node set, identify relational network；
Wherein, the relational network attaching information of described node set includes: each node in described node set
The relational network attaching information of each pair of node in relational network attaching information, and/or described node set；
The relational network attaching information of individual node, including: represent and there is particular kind of relationship with described individual node
The information of other nodes particular kind of relationship degree of strength each other；
The relational network attaching information of single pair of node, including: represent the first set and the second set similarity degree
Information；First collection is combined into: with this, primary nodal point in node is existed particular kind of relationship in described node set
Node mark constitute set；Second collection be combined into: in described node set with this to second in node
There is the set that the mark of the node of particular kind of relationship is constituted in node.
2. the method for claim 1, it is characterised in that determine that the relational network of node set is returned
Before genus information, described method also includes:
According to particular kind of relationship information, from start node set, choose node constitute described node set；
Wherein, described particular kind of relationship information is used for representing: between the different nodes in described start node set
The degree of strength of particular kind of relationship.
3. method as claimed in claim 2, it is characterised in that according to particular kind of relationship information, from initially
Before choosing the node described node set of composition in node set, described method also includes:
Perform respectively for every pair of node in described start node set:
According to this similarity to node respectively region, and in described start node set with this to joint
All there is the number of the node of described particular kind of relationship in point, determines this particular kind of relationship information to node.
4. method as claimed in claim 3, it is characterised in that according to particular kind of relationship information, from initially
Node set is chosen node and constitutes described node set, including:
Particular kind of relationship information according to described every pair of node, and the relation degree of strength threshold value set, from initially
In node set, choose the degree of strength of particular kind of relationship higher than the power represented by described relation degree of strength threshold value
Each pair of node of degree, constitutes described node set.
5. method as claimed in claim 3, it is characterised in that this phase to node region respectively
Like degree, including one or more in following similarity:
The similarity of this ship-to that node is used respectively；
The similarity of the physical address of this wireless network access point that node was respectively connected to；
This similarity to node used IP address respectively.
6. method as claimed in claim 3, it is characterised in that right with this in described start node set
All there is the node of described particular kind of relationship in node, including:
In described start node set, be there is the first relation in a node in node with this, with this to joint
There is the node of the second relation in another node in point；
Wherein, first relation that exists includes:
There is e-bank's treasury trade, interbehavior occurred in social networks, and/or each used
Electronic account bound same medium；
Second relation that exists includes:
There is e-bank's treasury trade, interbehavior occurred in social networks, and/or each used
Electronic account bound same medium.
7. method as claimed in claim 6, it is characterised in that right with this in described start node set
All there is the node of described particular kind of relationship in node, also includes:
In described start node set, region is the highest to the Similarity value of node region respectively with this
Node in similarity threshold.
8. the method for claim 1, it is characterised in that when the relational network of node set belongs to
Information includes: in described node set during the relational network attaching information of each node, determine the pass of node set
It is network home information, including:
According to the particular kind of relationship of described each node, generate logic relation picture；Wherein, in described logic relation picture
Comprise described each node；In described logic relation picture, each bar limit represents and there is institute between the node at two ends, limit
State particular kind of relationship；
Perform for each node:
The number of the adjacent node according to this node, and the number on limit between the adjacent node of this node, meter
Calculate the weighted value of this node, as the relational network attaching information of the individual node that this node represents.
9. the method for claim 1, it is characterised in that when the relational network of described node set
Attaching information includes: in described node set during the relational network attaching information of each pair of node, determine set of node
The relational network attaching information closed, including:
According to the particular kind of relationship of described each node, generate logic relation picture；Wherein, in described logic relation picture
Respectively comprise described each node；In described logic relation picture, each bar limit represents existence between the node at two ends, limit
Described particular kind of relationship；
Perform for each edge:
Between other each nodes respectively and in addition to this is to node of a pair node according to these two ends, limit the most equal
There is limit, calculate the weighted value on this limit, the relational network as this single pair of node corresponding to node belongs to
Information.
10. method as claimed in claim 7 or 8, it is characterised in that identify from described node set
After relational network, described method also includes:
Recognition result is estimated.
The identification device of 11. 1 kinds of relational networks, it is characterised in that including:
Determine information unit, for determining the relational network attaching information of node set；
Identify relation unit, for according to described relational network attaching information, identify from described node set
Relational network；
Wherein, the relational network attaching information of described node set includes: each node in described node set
The relational network attaching information of each pair of node in relational network attaching information, and/or described node set；
The relational network attaching information of individual node, including: represent and there is particular kind of relationship with described individual node
The information of other nodes particular kind of relationship degree of strength each other；
The relational network attaching information of single pair of node, including: represent the first set and the second set similarity degree
Information；First collection is combined into: with this, primary nodal point in node is existed particular kind of relationship in described node set
Node mark constitute set；Second collection be combined into: in described node set with this to second in node
There is the set that the mark of the node of particular kind of relationship is constituted in node.
12. devices as claimed in claim 11, it is characterised in that described device also includes: choose collection
Close unit, before determining the relational network attaching information of node set,
According to particular kind of relationship information, from start node set, choose node constitute described node set；
Wherein, described particular kind of relationship information is used for representing: between the different nodes in described start node set
The degree of strength of particular kind of relationship.
13. devices as claimed in claim 12, it is characterised in that choose aggregation units, for basis
Particular kind of relationship information, before choosing the node described node set of composition from start node set,
Perform respectively for every pair of node in described start node set:
According to this similarity to node respectively region, and in described start node set with this to joint
All there is the number of the node of described particular kind of relationship in point, determines this particular kind of relationship information to node.
14. devices as claimed in claim 13, it is characterised in that choose aggregation units, be used for:
Particular kind of relationship information according to described every pair of node, and the relation degree of strength threshold value set, from initially
In node set, choose the degree of strength of particular kind of relationship higher than the power represented by described relation degree of strength threshold value
Each pair of node of degree, constitutes described node set.
15. devices as claimed in claim 13, it is characterised in that this is to node region respectively
Similarity, including one or more in following similarity:
The similarity of this ship-to that node is used respectively；
The similarity of the physical address of this wireless network access point that node was respectively connected to；
This similarity to node used IP address respectively.
16. devices as claimed in claim 13, it is characterised in that with this in described start node set
Node is all existed the node of described particular kind of relationship, including:
In described start node set, be there is the first relation in a node in node with this, with this to joint
There is the node of the second relation in another node in point；
Wherein, first relation that exists includes:
There is e-bank's treasury trade, interbehavior occurred in social networks, and/or each used
Electronic account bound same medium.
Second relation that exists includes:
There is e-bank's treasury trade, interbehavior occurred in social networks, and/or each used
Electronic account bound same medium.
17. devices as claimed in claim 16, it is characterised in that with this in described start node set
Node is all existed the node of described particular kind of relationship, also includes:
In described start node set, region is the highest to the Similarity value of node region respectively with this
Node in similarity threshold.
18. devices as claimed in claim 11, it is characterised in that when the relational network of node set is returned
Genus information includes: in described node set during the relational network attaching information of each node, determine information unit,
For:
According to the particular kind of relationship of described each node, generate logic relation picture；Wherein, in described logic relation picture
Comprise described each node；In described logic relation picture, each bar limit represents and there is institute between the node at two ends, limit
State particular kind of relationship；
Perform for each node:
The number of the adjacent node according to this node, and the number on limit between the adjacent node of this node, meter
Calculate the weighted value of this node, as the relational network attaching information of the individual node that this node represents.
19. devices as claimed in claim 11, it is characterised in that when the network of personal connections of described node set
Network attaching information includes: in described node set during the relational network attaching information of each pair of node, determine information
Unit, is used for:
According to the particular kind of relationship of described each node, generate logic relation picture；Wherein, in described logic relation picture
Comprise described each node；In described logic relation picture, each bar limit represents and there is institute between the node at two ends, limit
State particular kind of relationship；
Perform for each edge:
Between other each nodes respectively and in addition to this is to node of a pair node according to these two ends, limit the most equal
There is limit, calculate the weighted value on this limit, the relational network as this single pair of node corresponding to node belongs to
Information.
20. devices as described in claim 17 or 18, it is characterised in that described device also includes: comment
Estimate result unit, after identifying relational network from described node set,
Recognition result is estimated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510015153.8A CN105844577A (en) | 2015-01-12 | 2015-01-12 | Relation network recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510015153.8A CN105844577A (en) | 2015-01-12 | 2015-01-12 | Relation network recognition method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105844577A true CN105844577A (en) | 2016-08-10 |
Family
ID=57177884
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510015153.8A Pending CN105844577A (en) | 2015-01-12 | 2015-01-12 | Relation network recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105844577A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107862530A (en) * | 2016-09-19 | 2018-03-30 | 阿里巴巴集团控股有限公司 | Establish the method and device of user's interpersonal relationships information |
| CN108109069A (en) * | 2016-11-24 | 2018-06-01 | 方正国际软件(北京)有限公司 | A kind of community network establishment of connection method and device |
| CN108171612A (en) * | 2016-12-06 | 2018-06-15 | 北京国双科技有限公司 | A kind of correlating method and device |
| CN110956391A (en) * | 2019-11-28 | 2020-04-03 | 中信银行股份有限公司 | Information processing method based on finance and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102202012A (en) * | 2011-05-30 | 2011-09-28 | 中国人民解放军总参谋部第五十四研究所 | Group dividing method and system of communication network |
| US20130336216A1 (en) * | 2003-05-08 | 2013-12-19 | Mesh Dynamics, Inc. | Persistent mesh for isolated mobile and temporal networking |
-
2015
- 2015-01-12 CN CN201510015153.8A patent/CN105844577A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130336216A1 (en) * | 2003-05-08 | 2013-12-19 | Mesh Dynamics, Inc. | Persistent mesh for isolated mobile and temporal networking |
| CN102202012A (en) * | 2011-05-30 | 2011-09-28 | 中国人民解放军总参谋部第五十四研究所 | Group dividing method and system of communication network |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107862530A (en) * | 2016-09-19 | 2018-03-30 | 阿里巴巴集团控股有限公司 | Establish the method and device of user's interpersonal relationships information |
| CN108109069A (en) * | 2016-11-24 | 2018-06-01 | 方正国际软件(北京)有限公司 | A kind of community network establishment of connection method and device |
| CN108171612A (en) * | 2016-12-06 | 2018-06-15 | 北京国双科技有限公司 | A kind of correlating method and device |
| CN110956391A (en) * | 2019-11-28 | 2020-04-03 | 中信银行股份有限公司 | Information processing method based on finance and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alsheikh et al. | The accuracy-privacy trade-off of mobile crowdsensing | |
| CN105335409B (en) | A kind of determination method, equipment and the network server of target user | |
| CN105447028B (en) | The method and device of identification feature account | |
| WO2020034750A1 (en) | Classification model generation method and device, and data identification method and device | |
| CN112465627B (en) | Financial loan auditing method and system based on block chain and machine learning | |
| CN109063966A (en) | The recognition methods of adventure account and device | |
| CN109189937B (en) | Feature relationship recommendation method and device, computing device and storage medium | |
| CN103198161A (en) | Microblog ghostwriter identifying method and device | |
| CN107895038A (en) | A kind of link prediction relation recommends method and device | |
| CN105844577A (en) | Relation network recognition method and device | |
| CN110009365B (en) | User group detection method, device and equipment for abnormally transferring electronic assets | |
| CN109816323A (en) | Transaction auditing method, calculating equipment, storage medium for block chain network | |
| US11631090B2 (en) | Systems and methods for determining offer eligibility using a predicate logic tree against sets of input data | |
| CN104899762A (en) | Trust management method based on backward inference | |
| CN107294974A (en) | The method and apparatus for recognizing target clique | |
| CN107742217A (en) | Service control method and device based on businessman | |
| CN113609345B (en) | Target object association method and device, computing equipment and storage medium | |
| CN110276602A (en) | Block chain Classifying Consensus method, system and the electronic equipment of internet of things oriented | |
| US11218548B2 (en) | System and method for facilitating a data exchange amongst communication devices connected via one or more communication networks | |
| CN107622197A (en) | Device identification method and device, weighing computation method and device for equipment identification | |
| CN112132606B (en) | Dynamic price adjustment method and system based on graph attention algorithm | |
| CN106296252A (en) | User recommends method and device | |
| CN109146317A (en) | A kind of Benefit share method and relevant device | |
| US11720900B1 (en) | Systems and methods for adaptive learning to replicate peak performance of human decision making | |
| CN112560105B (en) | Joint modeling method and device for protecting multi-party data privacy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160810 |
|
| RJ01 | Rejection of invention patent application after publication |