CN110110212A - Combo discovering method, server, terminal installation and system - Google Patents
Combo discovering method, server, terminal installation and system Download PDFInfo
- Publication number
- CN110110212A CN110110212A CN201810063403.9A CN201810063403A CN110110212A CN 110110212 A CN110110212 A CN 110110212A CN 201810063403 A CN201810063403 A CN 201810063403A CN 110110212 A CN110110212 A CN 110110212A
- Authority
- CN
- China
- Prior art keywords
- score value
- community discovery
- cohesion
- objects
- server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The present invention discloses a kind of Combo discovering method and its relevant server, terminal installation, and system calculates equipment and storage medium.The Combo discovering method includes: the close relationship based on the object data constructed in advance, the cohesion score value between computing object;Community discovery algorithm is called as weight using the calculated cohesion score value of institute;Export the community discovery result obtained by community discovery algorithm.Community discovery algorithm based on cohesion as weight score value not only has specific termination condition, also avoids the case where computing repeatedly.Therefore, compared with the prior art, Combo discovering method of the invention significantly improves the efficiency and accuracy of community discovery.
Description
Technical field
The present invention relates to network structure detection technique fields, can more find precisely and more stably more particularly to one kind
The method of the community structure of hierarchy and its relevant server, terminal installation, system calculate equipment and storage medium.
Background technique
It with the booming of the social networks based on internet and is widely applied, more and more people utilize social network
Network carries out information interchange activity.For example, the moon in the maximum social network sites Facebook in the world of the third season in 2017 is active
User has broken through 2,000,000,000.In-depth study is carried out to the complicated community network of this large size, is not only had in terms of network security
There is actual guiding value, and in computer science, sociology, the fields such as biology all have important research significance.
Community structure is the most important structure feature in community network.In recent years, a large amount of researcher uses various
Theory and method carries out the detection of community structure.So far, the Combo discovering method for comparing core includes (i) figure segmentation side
Method, for example, K-L (Kernighan-Lin) algorithm, GN (Girvan-Newman) algorithm;(ii) modularity optimization method, for example,
Fast Newman algorithm, Louvain algorithm, Simulated Annealing algorithm;(iii) label transmission method, for example,
Label propagation Alogorithm algorithm, Hubs-based algorithm, COPRA algorithm;(iv) dynamic method, example
Such as, Finding and Extracting Communities algorithm, INFOMAP algorithm, Ronhovde-Nussinov algorithm.
In these Combo discovering methods, GN algorithm is a classical community discovery algorithm, it belongs to the hierarchical clustering algorithm of division,
Its basic thought be it is continuous delete have in network relative to institute's active node it is maximum while betweenness while, then, then weigh
It is new calculate in network it is remaining while the while betweenness relative to institute's active node, repeat this process, until in network, Suo Youbian
All it is deleted.Specifically, the step of GN algorithm, is as follows: (1) calculate each while while betweenness;(2) it is maximum to delete number of boundary
Side;(3) recalculate in network it is remaining while while order;(4) (3) and (4) step is repeated, until any top in network
Until point is as a community.GN algorithm causes its efficiency and accuracy low due to following defect: not knowing last meeting (a)
How many community;(b) it might have the case where much computing repeatedly shortest path when calculating side betweenness, the time is complicated
Du Taigao;(c) GN algorithm cannot judge algorithm final position.Therefore, it is desirable to which one kind can more precisely and more stably discovery layer
The method of the community structure of secondary property.
Summary of the invention
In order to overcome above-mentioned deficiency in the prior art, the present invention provides one kind being capable of more precisely and more stably discovery layer
The method of the community structure of secondary property and its relevant server, terminal installation, system calculate equipment and storage medium.
According to an aspect of the present invention, a kind of Combo discovering method is provided, the Combo discovering method includes: based on pre-
The close relationship of the object data first constructed, the cohesion score value between computing object;Made with the calculated cohesion score value of institute
Community discovery algorithm is called for weight;Export the community discovery result obtained by community discovery algorithm.
In Combo discovering method of the invention, the community discovery algorithm based on cohesion as weight score value not only has
Specific termination condition, also avoids the case where computing repeatedly.Therefore, compared with the prior art, Combo discovering method of the invention
Significantly improve the efficiency and accuracy of community discovery.
Preferably, the cohesion score value between computing object may include in given two objects and its all once passes
The cohesion score value between described two objects is calculated in the case where system and/or two degree of relationships by scores accumulated.
Preferably, the cohesion score value between computing object may include in given two objects and its all static passes
The cohesion score value between described two objects is calculated in the case where system and/or dynamic relationship by scores accumulated.
Preferably, the cohesion score value between computing object may include given two objects and its it is all two degree indirectly
The cohesion score value between described two objects is calculated by scores accumulated in the case where relationship and once static relation.
Preferably, the cohesion score value between computing object may include steps of: a) the given time range of basis,
All two degree of indirect relations between described two objects are calculated, and count the frequency of every class relationship by type;B) basis
The given time range, inquiry with described two objects it is associated institute it is in need calculating cohesion once static relation;
C) combining step a) and b) as a result, then according to each relationship occur number and type obtained from Integral Rule look-up table
Corresponding cohesion score value is taken, the parent between described two objects is then obtained by cumulative related cohesion score value
Density.
Preferably, the community discovery algorithm can be Fast Unfolding algorithm.
According to another aspect of the present invention, a kind of server for community discovery is provided, the server includes: parent
Density score value computing device, for the close relationship based on the object data constructed in advance, the cohesion between computing object is divided
Value;Community discovery algorithm calling device, for calling community discovery algorithm as weight using the calculated cohesion score value of institute;
Community discovery result output device, for exporting the community discovery result obtained by community discovery algorithm.
Preferably, the cohesion score value computing device may include scores accumulated unit, in given two objects
And its it is calculated between described two objects in the case where all once relationships and/or two degree of relationships by scores accumulated
Cohesion score value.
Preferably, the cohesion score value computing device may include scores accumulated unit, in given two objects
And its it is calculated between described two objects in the case where all static relations and/or dynamic relationship by scores accumulated
Cohesion score value.
Preferably, the cohesion score value computing device may include scores accumulated unit, in given two objects
And its all two degree of indirect relations and once in the case where static relation by scores accumulated be calculated described two objects it
Between cohesion score value.
Preferably, the scores accumulated unit may include: two degree of indirect relation determination units, for according to it is given when
Between range, calculate all two degree of indirect relations between described two objects, and count the frequency of every class relationship by type;
Once static relation determination unit, for according to the given time range, inquiry to be associated all with described two objects
Need to calculate the once static relation of cohesion;Cohesion score value obtaining unit, for single according to being determined by two degree of indirect relations
The number and type that member and each relationship that once static relation determination unit obtains occur are obtained from Integral Rule look-up table
Corresponding cohesion score value is taken, the parent between described two objects is then obtained by cumulative related cohesion score value
Density.
Preferably, the community discovery algorithm can be Fast Unfolding algorithm.
Preferably, the server can also include: storage device, for storing the parent with the object data constructed in advance
The relevant data of close relationship.
Preferably, the server can also include: configuration management device, include matching for source data for storage and management
It sets, the system configuration of transport protocol configuration, schema configuration and mapping configuration.
According to another aspect of the present invention, a kind of terminal installation for community discovery, the terminal installation packet are provided
Include: data storage cell is used for storage object data;Transmission unit, the number of objects for that will store in the data store
According to being sent to above-mentioned server;Receiving unit is exported for receiving from the community discovery result output device of the server
Community discovery result;Visualization, the community discovery result visualization for will be exported from the server.
According to another aspect of the present invention, a kind of system for community discovery is provided, the system comprises terminal dresses
It sets and server.The terminal installation includes: data storage cell, is used for storage object data;Transmission unit, being used for will be in number
Server is sent to according to the object data stored in storage unit;Receiving unit, for receiving the society exported from the server
Group's discovery result;Visualization, the community discovery result visualization for receiving receiving unit.The server packet
It includes: cohesion score value computing device, it is intimate between computing object for the close relationship based on the object data constructed in advance
Spend score value;Community discovery algorithm calling device, for calling community discovery as weight using the calculated cohesion score value of institute
Algorithm;Community discovery result output device, for exporting the community discovery result obtained by community discovery algorithm.
According to another aspect of the present invention, a kind of calculating equipment is provided, the calculating equipment includes: processor;And
Memory is stored thereon with executable code, when the executable code is executed by the processor, holds the processor
The above-mentioned patrol management method of row.
According to another aspect of the present invention, a kind of non-transitory machinable medium is provided, being stored thereon with can
Code is executed, when the executable code is executed by the processor of electronic equipment, the processor is made to execute above-mentioned patrol
Management method.
In the present invention, specific termination item is not only had as the community discovery algorithm of weight score value based on cohesion
Part also avoids the case where computing repeatedly.Therefore, compared with the prior art, Combo discovering method of the invention significantly improves society
The efficiency and accuracy of group's discovery.
Detailed description of the invention
Exemplary embodiment of the invention is described in more detail in conjunction with the accompanying drawings, it is of the invention above-mentioned and its
Its purpose, feature and advantage will be apparent, wherein in exemplary embodiment of the invention, identical reference label
Typically represent same parts.
Fig. 1 shows the thinking figure of Fast Unfolding algorithm in the prior art.
Fig. 2 shows the flow charts of Combo discovering method according to an embodiment of the present invention.
Fig. 3 shows the process of the calculating cohesion score value in Combo discovering method according to an embodiment of the present invention.
Fig. 4 is the structural block diagram for showing the server-side for realizing Combo discovering method according to an embodiment of the present invention.
Fig. 5 shows the structural block diagram of scores accumulated module included in server-side according to an embodiment of the present invention.
Fig. 6 is the system flow chart shown for realizing Combo discovering method according to an embodiment of the present invention.
Fig. 7 is the structural block diagram for showing the client of Combo discovering method according to an embodiment of the present invention.
Fig. 8 A to 8F shows for realizing the human-computer interaction interface of Combo discovering method according to an embodiment of the present invention.
Specific embodiment
The preferred embodiment that the present invention will be described in more detail below with reference to accompanying drawings.Although showing the present invention in attached drawing
Preferred embodiment, however, it is to be appreciated that may be realized in various forms the present invention without the embodiment party that should be illustrated here
Formula is limited.On the contrary, thesing embodiments are provided so that the present invention is more thorough and complete, and can will be of the invention
Range is completely communicated to those skilled in the art.
Before technical solution of the present invention is specifically described, the term mentioned in the present invention is fitted first
When explanation.
" community structure " mentioned herein: in a data network, object is equivalent to each point, passes through between object
Mutual relationship constitutes the structure of whole network, and within such networks, the connection between some objects is more close, has
Connection relationship between object is more sparse, and within such networks, more closely part can be seen as a society for connection
There is more close connection in area between internal node, and it is then opposite between the community Liang Ge connect more sparse, this is just known as
Community structure.
" static relation " mentioned herein: it will not change substantially with time change, such as father and son, mothers and sons, brother
Younger brother etc..
" dynamic relationship " mentioned herein: it as time change is continually changing, such as rides, surf the Internet, staying, even
Connect, buy etc..
" once relationship " mentioned herein: the single order of the partial points pair between two vertex in network is neighbouring to close
System, for the every opposite vertexes linked by side (u, v), the weight S on the sideU, vThe single order similarity between u and v is indicated, if in u
There is no side between v, their single order similarity is 0.
" two degree of relationships " mentioned herein: an opposite vertexes (u, v) are in a network between its proximity network structure
Second order proximity relations mathematically allows Pu=(SU, 1, SU, 2..., SU, | V |) indicate that the single order between u and every other vertex is adjacent
Recency, then the second order proximity relations between u and v is by PuAnd PvBetween similarity characterize.If none vertex is simultaneously
It is linked with u and v, then the second order similarity of u and v is 0.
" indirect relation " mentioned herein: it is a kind of based on static, dynamic relationship similitude abstraction relation, such as together
Trip, same to family, with social activity etc..
" cohesion " mentioned herein: cohesion is mainly used for portraying the intimate degree between two objects, except quiet
Outside state relationship, the common feature depended in daily behavior activity defines, and cohesion quantitative expression is this indirectly
The size of relationship.
" in betweenness " mentioned herein: the number of shortest path when any two node is by this in network.
" shortest path " mentioned herein: from certain vertex, the road that another summit is passed through is reached along the side of figure
In diameter, the smallest paths of weights sum on each side.
" GN algorithm " mentioned herein: being a classical community discovery algorithm, it belongs to the hierarchical clustering of division
Algorithm, basic thought be it is continuous delete have in network relative to institute's active node it is maximum while betweenness while, then,
Recalculate again in network it is remaining while the while betweenness relative to institute's active node, repeat this process, until in network, institute
There is side to be all deleted.
" modularity (Modularity) value " mentioned herein: be one be defined on [- 0.5,1) finger in section
Mark, algorithm is to consider the difference for connecting number of edges and expected value in each community for a certain community structure.It is high actually to connect side
In random expectation, illustrating node more has the tendency that concentrating in certain communities, i.e. the modular construction of network is more obvious.
" Fast Unfolding algorithm " mentioned herein: modularity becomes the important mark that measurement community divides superiority and inferiority
Standard, the network module angle value after division is bigger, and the effect for illustrating that community divides is better, and Fast Unfolding algorithm is to be based on
The algorithm that modularity divides community, Fast Unfolding algorithm are a kind of algorithms of iteration, and main target is constantly to divide
Community increases the modularity of the whole network after dividing constantly.Fast Unfolding algorithm mainly includes two stages,
As shown in Figure 1.First stage is known as modularity optimization (Modularity Optimization), mainly draws each node
It assigns in the community where the node being adjacent, so that the value of modularity constantly becomes larger;Second stage is known as community's polymerization
(Community Aggregation), the first step is mainly marked off to the community's polymerization come becomes a point, i.e., according to upper one
The community structure that step generates reconfigures network.Process more than repeating, until the structure in network no longer changes.Specifically
Algorithmic procedure it is as follows:
1) it initializes, each point is divided in different communities;
2) mould at this time is calculated by each community put where attempting to be divided into the point being adjacent to each node
Lumpiness judges to divide whether the difference DELTA Q of the modularity of front and back is positive number, if positive number, then receives this division, if not
Positive number then abandons this division;
3) process more than repeating, until cannot increase modularity again;
4) construction is new schemes, and what each point in new figure represented is that each community come is marked in step 3, continues to execute step
2 and step 3, until the structure until community no longer changes.
" community discovery algorithm " mentioned herein: corporations' knot of network-based structural information discovery nodes
The algorithm of structure, including based on figure segmentation algorithm, clustering algorithm, splitting algorithm, spectral method, dynamic algorithm, be based on statistical inference
Algorithm and algorithm based on modularity.The example of the community discovery algorithm used in the present invention includes Fast Unfolding
Algorithm, K-L algorithm compose Bisection Algorithms, GN algorithm, Newman fast algorithm, factions filtering CPM method (clique
Percolation method), based on the non-overlap community discovery algorithm LPA that label is propagated, the overlapping society propagated based on label
Area finds algorithm COPRA.
It is below with reference to accompanying drawings and right in conjunction with specific embodiments in order to be clearer and more clear technical solution of the present invention
The present invention is described in detail.
Fig. 2 shows the flow charts of Combo discovering method according to an embodiment of the present invention.As described in Figure 2, the community discovery side
Method includes: that cohesion score value calculates step S101, based on the close relationship of the object data constructed in advance, between computing object
Cohesion score value;Community discovery algorithm invocation step S102 calls corporations as weight using the calculated cohesion score value of institute
It was found that algorithm;Community discovery result exports step S103, exports the community discovery result obtained by community discovery algorithm.In parent
Before density score value calculates step S101, this method can also include receiving object data, the corresponding relationship of building object data
Then behavioral data constructs their close relationship.In addition, this method is also after community discovery result exports step S103
It may include that community discovery result is visualized and analyzed.
It is calculated in step S101 in cohesion score value, in given two object a and b and its all two degree of indirect relations and one
Spend static relation in the case where, between cohesion calculating logic be a simple summation process:
Wherein, S (ri) it is relationship r between object a and biIntimate score value, in actual mechanical process in a manner of look-up table
It realizes.Specifically, being defined according to cohesion, the dimensions such as static relation intensity, mutual-action behavior, common participation for checking between object are looked into
It looks for frequency, and calculates according to starting score value, beyond scoring paradigms such as partial scores, the score value upper limit, accumulative multiples between object
Cohesion score value.In this case, the cohesion score value between computing object includes: the time range that a) basis is given,
All two degree of indirect relations between described two objects are calculated, and count the frequency (step of every class relationship by type
S201);B) according to given time range, inquiry and described two objects it is associated calculating cohesion in need it is once quiet
State relationship (step S202);C) combining step a) and b) as a result, then according to each relationship occur number and type, from product
Divider then in look-up table, obtains corresponding cohesion score value, then by the cumulative related cohesion score value of institute to obtain
State the cohesion (step S203) between two objects.Although here in given two objects and its all two degree of indirect relations and
The cohesion score value between two objects once was calculated by scores accumulated in the case where static relation, but it is also possible to
In the case where given two objects and its all once relationship and/or two degree of relationships, alternatively, in given two objects and its
In the case where all static relations and/or dynamic relationship, the parent between described two objects is calculated by scores accumulated
Density score value.
In community discovery algorithm invocation step S102, a) by scheme (such as Fig. 1) in each node regard as one it is independent
Community, the number of community is identical as node number at this time;B) it to each node i, successively attempts node i to be assigned to each of which neighbour
The community where node is occupied, is calculated with the cohesion between node as weight (score value of step S101 calculating) and is distributed preceding and divide
Modularity after matching changes Delta Q, and records that maximum neighbor node of Delta Q, if max Delta Q > 0,
Node i is distributed the community where that maximum neighbor node of Delta Q, otherwise remained unchanged;C) it repeats b), until all
The affiliated community of node no longer changes;E) figure is compressed, by all Node compressions in the same community at a new section
Point, the weight on the side between community's interior nodes are converted into the weight of the ring of new node, and the side right between community is converted into new node again
Between side right weight;F) it repeats a) until the modularity of entire figure is no longer changed.That is, with the calculated cohesion of institute
Score value calls Fast Unfolding algorithm in the prior art as weight.But in community discovery algorithm invocation step
In S102, it can also be called using the calculated cohesion score value of institute as weight other based on optimization module angle value progress society
The algorithm of group's discovery, for example, K-L algorithm, composes Bisection Algorithms, GN algorithm, Newman fast algorithm, factions' filtering CPM method
(clique percolation method) is propagated based on the non-overlap community discovery algorithm LPA that label is propagated based on label
Overlapping community discovery algorithm COPRA.
In Combo discovering method of the invention, the community discovery algorithm based on cohesion as weight score value not only has
Specific termination condition, also avoids the case where computing repeatedly.Therefore, compared with the prior art, Combo discovering method of the invention
Significantly improve the efficiency and accuracy of community discovery.
Fig. 4 is the structural block diagram for showing the server-side 400 for realizing Combo discovering method according to an embodiment of the present invention.
As shown in figure 4, server-side 400 includes cohesion score value computing device 401, community discovery algorithm calling device 402 and corporations' hair
Now result output device 403.Cohesion score value computing device 401 is used for the close relationship based on the object data constructed in advance,
Cohesion score value between computing object.Community discovery algorithm calling device 402 is used to make with the calculated cohesion score value of institute
Community discovery algorithm is called for weight.Community discovery result output device 403 is obtained for exporting by community discovery algorithm
Community discovery result.In addition, cohesion score value computing device 401 may include scores accumulated module 401A.Scores accumulated mould
Block 401A is used in the case where given two objects and its all once relationship and/or two degree of relationships, right at given two
As and its all static relation and/or dynamic relationship in the case where, alternatively, given two objects and its it is all two degree indirectly
In the case where relationship and once static relation, the cohesion score value between described two objects is calculated by scores accumulated.
In one embodiment of the invention, as shown in figure 5, scores accumulated module 401A may include two degree of indirect relations
Determination unit 51, once static relation determination unit 52 and cohesion score value obtaining unit 53.Two degree of indirect relation determination units
51 for calculating all two degree of indirect relations between described two objects, and count by type according to given time range
The frequency of every class relationship.Once static relation determination unit 52 was used for according to the given time range, inquiry and institute
State two objects it is associated it is in need calculate cohesion once static relation.Cohesion score value obtaining unit 53 is used for basis
The number and type occurred by each relationship that two degree of indirect relation determination units and once static relation determination unit obtain, from
In Integral Rule look-up table, corresponding cohesion score value is obtained, is then obtained by the cumulative related cohesion score value of institute
Cohesion between described two objects.
In one embodiment of the invention, server-side 400 can also include the number of objects for storing with constructing in advance
According to the relevant data of close relationship storage device.The storage device can also store the relationship knot that knowledge based map defines
Structure comprising the triples such as vertex, side, attribute.
In one embodiment of the invention, server-side 400 can also include: configuration management device, for storing and managing
Reason includes the system configuration of the configuration of source data, transport protocol configuration, schema configuration and mapping configuration.
Fig. 6 is the system flow chart shown for realizing Combo discovering method according to an embodiment of the present invention.The system packet
Include client 300 and server-side 400.The structural block diagram of client 300 is shown in Figure 7.As shown in fig. 7, client 300 includes number
According to storage unit 301, transmission unit 302, receiving unit 303 and visualization 304.Data storage cell 301 is for storing
Object data.Transmission unit 302 is used to the object data stored in the data store being sent to server-side 400.It receives
Unit 303 is used to receive the community discovery result exported from the community discovery result output device 403 of server-side 400.Visualization
Unit 304 is used for the community discovery result visualization that will be exported from server-side 400.As shown in fig. 6, firstly, client 300 uploads
Then object data is sent to 400 (step of server-side by transmission unit 302 by object data (step S61) to be analyzed
S62).Next, server-side 400 constructs relationship behavior data (step S63) corresponding with the object data received.Then,
Cohesion score value computing device 401 in server-side 400 constructs the close relationship of data, and calculates cohesion score value (step
S64).Hereafter, the community discovery calling device 402 in server-side 400 is called using calculated cohesion score value as weight
Community discovery algorithm (step S65).Then, the output of community discovery result output device 403 in server-side 400 is sent out by corporations
The community discovery result (step S66) that existing algorithm obtains, and send it to the 303 (step of receiving unit in client 300
S67).Then, the visualization 304 in client 300 is visualized and is analyzed (step to received result
S68)。
In addition, also introducing the man-machine friendship for realizing Combo discovering method according to an embodiment of the present invention in the present embodiment
Mutual interface.Using human-computer interaction interface shown in Fig. 8 A to Fig. 8 F, user can intuitive, naturally, efficiently realize of the invention
Combo discovering method.It is briefly described below with reference to particular content of Fig. 8 A to Fig. 8 F to human-computer interaction interface.
Fig. 8 A to 8C shows the backstage configuration of cohesion building and is configured before all business diagnosis.Fig. 8 A institute
The interface display " basic information " shown and " decision rule " two." basic information " item includes " indirect relation " and " source relationship " two
A element." decision rule " item includes following element: (1) serial number;(2) Zuo Bianliang;(3) operator;(4) right variable;(5) standby
Note;(6) it operates;(7) logic.Interface display shown in Fig. 8 B " indirect relation overview " comprising following element: (1) serial number;
(2) indirect relation encodes;(3) indirect relation title;(4) remarks;(5) it operates." the VII configuration pipe of interface display shown in Fig. 8 C
Platform/relationship analysis configuration/intimate degree setting ", " setting details " include following element: (1) serial number;(2) relationship
Coding;(3) relation name;(4) integral mode;(5) starting score number;(6) intimate score value;(7) highest score;(8) it operates.
Fig. 8 D is the interface of the list page of Display Group analysis.The list includes following element: (1) type;(2) group's name
Claim;(3) personnel amount;(4) it operates.When clicking the button of " operation " on interface shown in Fig. 8 D, it just will appear Fig. 8 E
Shown in interface.The relationship analysis figure of interface display population analysis shown in Fig. 8 E.Fig. 8 F is the inquiry of the group shown in Fig. 8 E
Group later finds effect picture.
Combo discovering method according to the present invention and its relevant server above is described in detail by reference to attached drawing,
Terminal installation, system.
In addition, being also implemented as a kind of computer program or computer program product, the meter according to the method for the present invention
Calculation machine program or computer program product include the calculating for executing the above steps limited in the above method of the invention
Machine program code instruction.
In addition, the present invention can also be embodied as a kind of calculating equipment, which includes: processor;And memory,
It is stored thereon with executable code, when the executable code is executed by the processor, the processor is made to execute basis
Each step of method of the invention.
Alternatively, the present invention can also be embodied as a kind of (or the computer-readable storage of non-transitory machinable medium
Medium or machine readable storage medium), it is stored thereon with executable code (or computer program or computer instruction code),
When the executable code (or computer program or computer instruction code) by electronic equipment (or calculate equipment, server
Deng) processor execute when, so that the processor is executed each step according to the above method of the present invention.
Those skilled in the art will also understand is that, various illustrative logical blocks, mould in conjunction with described in disclosure herein
Block, circuit and algorithm steps may be implemented as the combination of electronic hardware, computer software or both.
The flow chart and block diagram in the drawings show the possibility of the system and method for multiple embodiments according to the present invention realities
Existing architecture, function and operation.In this regard, each box in flowchart or block diagram can represent module, a journey
A part of sequence section or code, a part of the module, section or code include one or more for realizing defined
The executable instruction of logic function.It should also be noted that in some implementations as replacements, the function of being marked in box can also
To be occurred with being different from the sequence marked in attached drawing.For example, two continuous boxes can actually be basically executed in parallel,
They can also be executed in the opposite order sometimes, and this depends on the function involved.It is also noted that block diagram and/or stream
The combination of each box in journey figure and the box in block diagram and or flow chart, can the functions or operations as defined in executing
Dedicated hardware based system realize, or can realize using a combination of dedicated hardware and computer instructions.
Various embodiments of the present invention are described above, above description is exemplary, and non-exclusive, and
It is not limited to disclosed each embodiment.Without departing from the scope and spirit of illustrated each embodiment, for this skill
Many modifications and changes are obvious for the those of ordinary skill in art field.The selection of term used herein, purport
In the principle, practical application or improvement to the technology in market for best explaining each embodiment, or make the art
Other those of ordinary skill can understand each embodiment disclosed herein.
Claims (18)
1. a kind of Combo discovering method, which is characterized in that the Combo discovering method includes:
Cohesion score value based on the close relationship of the object data constructed in advance, between computing object;
Community discovery algorithm is called as weight using the calculated cohesion score value of institute;
Export the community discovery result obtained by community discovery algorithm.
2. Combo discovering method according to claim 1, which is characterized in that the cohesion score value packet between computing object
It includes: in the case where given two objects and its all once relationship and/or two degree of relationships, being calculated by scores accumulated
Cohesion score value between described two objects.
3. Combo discovering method according to claim 1, which is characterized in that the cohesion score value packet between computing object
It includes: in the case where given two objects and its all static relation and/or dynamic relationship, being calculated by scores accumulated
Cohesion score value between described two objects.
4. Combo discovering method according to claim 1, which is characterized in that the cohesion score value packet between computing object
It includes: in the case where given two objects and its all two degree of indirect relations and once static relation, being calculated by scores accumulated
Obtain the cohesion score value between described two objects.
5. Combo discovering method according to claim 4, which is characterized in that the cohesion score value between computing object includes
Following steps:
A) according to given time range, all two degree of indirect relations between described two objects are calculated, and count by type
The frequency of every class relationship;
B) according to the given time range, inquiry and described two objects it is associated calculating cohesion in need once
Static relation;
C) combining step a) and b) as a result, then according to each relationship occur number and type, from Integral Rule look-up table
In, obtain corresponding cohesion score value, then obtained by the cumulative related cohesion score value of institute described two objects it
Between cohesion.
6. Combo discovering method according to any one of claims 1 to 5, which is characterized in that the community discovery is calculated
Method is Fast Unfolding algorithm.
7. a kind of server for community discovery, which is characterized in that the server includes:
Cohesion score value computing device, the parent for the close relationship based on the object data constructed in advance, between computing object
Density score value;
Community discovery algorithm calling device, for calling community discovery to calculate as weight using the calculated cohesion score value of institute
Method;
Community discovery result output device, for exporting the community discovery result obtained by community discovery algorithm.
8. server according to claim 7, which is characterized in that the cohesion score value computing device includes scores accumulated
Unit, based in the case where given two objects and its all once relationship and/or two degree of relationships through scores accumulated
Calculation obtains the cohesion score value between described two objects.
9. server according to claim 7, which is characterized in that the cohesion score value computing device includes scores accumulated
Unit, based in the case where given two objects and its all static relation and/or dynamic relationship through scores accumulated
Calculation obtains the cohesion score value between described two objects.
10. server according to claim 7, which is characterized in that the cohesion score value computing device includes aggregate value method of weighting
Sub-unit, for passing through aggregate value method of weighting in the case where given two objects and its all two degree of indirect relations and once static relation
Divide the cohesion score value being calculated between described two objects.
11. server according to claim 10, which is characterized in that the scores accumulated unit includes:
Two degree of indirect relation determination units, for calculating all two between described two objects according to given time range
Indirect relation is spent, and counts the frequency of every class relationship by type;
Once static relation determination unit, for inquiring associated with described two objects according to the given time range
It is in need calculate cohesion once static relation;
Cohesion score value obtaining unit, for basis, by two degree of indirect relation determination units and once, static relation determination unit is obtained
The number and type that each relationship arrived occurs obtain corresponding cohesion score value, then pass through from Integral Rule look-up table
Cumulative related cohesion score value obtains the cohesion between described two objects.
12. the server according to any one of claim 7 to 11, which is characterized in that the community discovery algorithm is
Fast Unfolding algorithm.
13. server according to claim 12, which is characterized in that the server further include:
Storage device, for storing data relevant to the close relationship of the object data constructed in advance.
14. server according to claim 12, which is characterized in that the server further include:
Configuration management device, for storage and management include source data configuration, transport protocol configuration, schema configuration and
The system configuration of mapping configuration.
15. a kind of terminal installation for community discovery, which is characterized in that the terminal installation includes:
Data storage cell is used for storage object data;
Transmission unit, for being sent to the object data stored in the data store according in claim 7 to 14
Described in any item servers;
Receiving unit, for receiving the community discovery result exported from the community discovery result output par, c of the server;
Visualization, the community discovery result visualization for will be exported from the server.
16. a kind of system for community discovery, the system comprises terminal installations and server, which is characterized in that
The terminal installation includes:
Data storage cell is used for storage object data;
Transmission unit, for the object data stored in the data store to be sent to server;
Receiving unit, for receiving the community discovery result exported from the server;
Visualization, the community discovery result visualization for receiving receiving unit, and
The server includes:
Cohesion score value computing device, the parent for the close relationship based on the object data constructed in advance, between computing object
Density score value;
Community discovery algorithm calling device, for calling community discovery to calculate as weight using the calculated cohesion score value of institute
Method;
Community discovery result output device, for exporting the community discovery result obtained by community discovery algorithm.
17. a kind of calculating equipment, comprising:
Processor;And
Memory is stored thereon with executable code, when the executable code is executed by the processor, makes the processing
Device executes the patrol management method as described in any one of claim 1-6.
18. a kind of non-transitory machinable medium, is stored thereon with executable code, when the executable code is electric
When the processor of sub- equipment executes, the processor is made to execute such as patrol manager described in any one of claims 1 to 6
Method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810063403.9A CN110110212A (en) | 2018-01-22 | 2018-01-22 | Combo discovering method, server, terminal installation and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810063403.9A CN110110212A (en) | 2018-01-22 | 2018-01-22 | Combo discovering method, server, terminal installation and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110110212A true CN110110212A (en) | 2019-08-09 |
Family
ID=67483471
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810063403.9A Pending CN110110212A (en) | 2018-01-22 | 2018-01-22 | Combo discovering method, server, terminal installation and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110110212A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102682039A (en) * | 2011-03-11 | 2012-09-19 | 北京千橡网景科技发展有限公司 | Method and device for displaying dynamic information of good friends |
CN105243593A (en) * | 2015-08-04 | 2016-01-13 | 电子科技大学 | Weighted network community clustering method based on hybrid measure |
CN107529656A (en) * | 2016-06-22 | 2018-01-02 | 腾讯科技(深圳)有限公司 | The division methods and server of a kind of myspace |
-
2018
- 2018-01-22 CN CN201810063403.9A patent/CN110110212A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102682039A (en) * | 2011-03-11 | 2012-09-19 | 北京千橡网景科技发展有限公司 | Method and device for displaying dynamic information of good friends |
CN105243593A (en) * | 2015-08-04 | 2016-01-13 | 电子科技大学 | Weighted network community clustering method based on hybrid measure |
CN107529656A (en) * | 2016-06-22 | 2018-01-02 | 腾讯科技(深圳)有限公司 | The division methods and server of a kind of myspace |
Non-Patent Citations (2)
Title |
---|
刘瑶: "基于节点亲密度和度的社会网络社团发现方法", 《计算机研究与发展》 * |
徐建民: "一种基于用户交互行为的微博社区发现方法", 《河北大学学报(自然科学版)》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Green et al. | A fast algorithm for streaming betweenness centrality | |
CN109033234B (en) | Streaming graph calculation method and system based on state update propagation | |
CN103179052B (en) | A kind of based on the central virtual resource allocation method and system of the degree of approach | |
Ljubić | Solving Steiner trees: Recent advances, challenges, and perspectives | |
US20140075002A1 (en) | Methods for clustering networks based on topology discovery and devices thereof | |
CN110555172B (en) | User relationship mining method and device, electronic equipment and storage medium | |
Singh et al. | IM‐SSO: Maximizing influence in social networks using social spider optimization | |
Dey et al. | Influence maximization in online social network using different centrality measures as seed node of information propagation | |
Chang et al. | Relative centrality and local community detection | |
CN112052404A (en) | Group discovery method, system, device and medium for multi-source heterogeneous relation network | |
CN111159577A (en) | Community division method and device, storage medium and electronic device | |
Liu et al. | Influence nodes identifying method via community-based backward generating network framework | |
Ghamgosar et al. | An extended distributed learning automata based algorithm for solving the community detection problem in social networks | |
Bagheri et al. | An efficient and fast influence maximization algorithm based on community detection | |
Zhu et al. | Roles of degree, H-index and coreness in link prediction of complex networks | |
Sharma et al. | Community mining in signed social networks-an automated approach | |
Li et al. | GAP: Genetic algorithm based large-scale graph partition in heterogeneous cluster | |
CN108011735B (en) | Community discovery method and device | |
CN110110212A (en) | Combo discovering method, server, terminal installation and system | |
Rashmi et al. | A review on overlapping community detection methodologies | |
Qiu et al. | A New Method for Identifying Influential Spreaders in Complex Networks | |
Liu et al. | A new method of identifying core designers and teams based on the importance and similarity of networks | |
CN103051476A (en) | Topology analysis-based network community discovery method | |
Niu et al. | On a two-stage progressive clustering algorithm with graph-augmented density peak clustering | |
Chakraborty | On the discovery of invariant groups in complex networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190809 |
|
RJ01 | Rejection of invention patent application after publication |