CN105654132A - Community detection method and device - Google Patents
Community detection method and device Download PDFInfo
- Publication number
- CN105654132A CN105654132A CN201511023521.XA CN201511023521A CN105654132A CN 105654132 A CN105654132 A CN 105654132A CN 201511023521 A CN201511023521 A CN 201511023521A CN 105654132 A CN105654132 A CN 105654132A
- Authority
- CN
- China
- Prior art keywords
- sampling
- network structure
- community
- burst
- denser
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
Abstract
The invention relates to a community detection method and device. The method includes the following steps that: an inputted intensive network structure is obtained; sampling fusion processing is performed on the intensive network structure, so that the sampling subgraph of the intensive network structure is obtained; an METIS algorithm is utilized to detect the sampling subgraph, so that the initial community detection data of the sampling subgraph can be obtained; and clustering processing is performed on the initial community detection data, so that the network community division result of the intensive network structure can be obtained.
Description
Technical field
The present invention relates to information science field, particularly relate to a kind of community detection method and device.
Background technology
Current complex network often all has very typical modular organization, and this shows to exist between node in the same module special similarity. This similarity can and intercommunal limit density contrast internal from community from showing.
In the prior art, the method developing many quickness and high efficiency detects for community. But, most prior art, when carrying out community's detection, is largely dependent upon the number of links on limit. Such as, Fast-Newman algorithm proposes the strategy of a cohesion level, and this strategy colony summit successively merges, and then forms bigger community so that modularity increases. Again such as, label propagation algorithm (LPA), each node of this algorithm initialization has the label of uniqueness, then allows label propagate whole network again.
But, the time complexity of Fast-Newman algorithm and LPA algorithm is O (mlogn2) and O (m+n) respectively, which has limited above two algorithm and finds community structure from bigger limit denser network, has certain restricted.
Summary of the invention
The invention provides a kind of community detection method and device, cannot find there is community structure restrictive problem from bigger limit denser network for solving Fast-Newman algorithm and LPA algorithm in existing algorithm.
For achieving the above object, in first aspect, the invention provides a kind of community detection method, described method includes:
Obtain the denser network structure of input;
Described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure;
Utilize METIS algorithm that described sampling subgraph is detected, it is thus achieved that the initial community detection data of described sampling subgraph;
Data are detected in described initial community and carries out clustering processing, it is thus achieved that Web Community's division result of described denser network structure.
Preferably, described described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure, specifically include:
Described denser network structure is carried out horizontal resection burst process, obtains multiple network structure burst and store the order information of described network structure burst;
Described network structure burst is carried out stochastic sampling process, it is thus achieved that the sampling results of described network structure burst;
According to described order information, described sampling results is carried out fusion treatment, obtain the sampling subgraph of described denser network structure.
Preferably, described described network structure burst is carried out stochastic sampling process, also includes before:
Determine the sampling proportion value n that described network structure burst is carried out stochastic sampling processv;
Described sampling proportion value nvMeet nv=min{dv,��(lndv+ ln2) };
Wherein, described dvFor the degree of each node in described network structure burst; Described �� is nonnegative constant.
Preferably, described described sampling results is carried out fusion treatment, obtains the sampling subgraph of described denser network structure, specifically include:
Determine and described sampling results belongs to a stochastic sampling and the sub-sample network with identical key assignments;
Belonging to described with a stochastic sampling and have the sub-sample network mapping of identical key assignments in same node, described node includes multiple described sub-sample network;
The multiple described sub-sample network included by described node is as the sampling subgraph of described denser network structure.
In second aspect, the invention provides a kind of community detecting device, described device includes:
Acquiring unit, for obtaining the denser network structure of input;
Sampling integrated unit, for described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure;
Detection unit, is used for utilizing METIS algorithm that described sampling subgraph is detected, it is thus achieved that the initial community detection data of described sampling subgraph;
Cluster cell, carries out clustering processing for described initial community is detected data, it is thus achieved that Web Community's division result of described denser network structure.
Preferably, described sampling integrated unit specifically includes:
Cutting subelement, for described denser network structure is carried out horizontal resection burst process, obtains multiple network structure burst and stores the order information of described network structure burst;
Sub-unit, for carrying out stochastic sampling process by described network structure burst, it is thus achieved that the sampling results of described network structure burst;
Fusant unit, for according to described order information, carrying out fusion treatment to described sampling results, obtain the sampling subgraph of described denser network structure.
Preferably, described sampling integrated unit also includes:
Determine subelement, for determining the sampling proportion value n that described network structure burst is carried out stochastic sampling processv;
Described sampling proportion value nvMeet nv=min{dv,��(lndv+ ln2) };
Wherein, described dvFor the degree of each node in described network structure burst; Described �� is nonnegative constant.
Preferably, described fusant unit also particularly useful for,
Determine and described sampling results belongs to a stochastic sampling and the sub-sample network with identical key assignments;
Belonging to described with a stochastic sampling and have the sub-sample network mapping of identical key assignments in same node, described node includes multiple described sub-sample network;
The multiple described sub-sample network included by described node is as the sampling subgraph of described denser network structure.
Therefore, by applying community detection method and the device of the embodiment of the present invention, denser network result is sampled fusion treatment by terminal, obtains sampling subgraph; Utilize METIS algorithm that sampling subgraph is detected, it is thus achieved that the initial community detection data of sampling subgraph; Initial community is detected data and carries out clustering processing by terminal, it is thus achieved that Web Community's division result of denser network structure. Owing in the present invention, denser network result is carried out multiple sampling fusion treatment by terminal, and then sampling subgraph is detected, after clustering processing, obtain Web Community's division result, and then solve Fast-Newman algorithm and LPA algorithm in existing algorithm and cannot find there is community structure restrictive problem from bigger limit denser network. Achieve and simply and efficiently find community structure, expanded being widely applied property.
Accompanying drawing explanation
The community detection method flow chart that Fig. 1 provides for the embodiment of the present invention;
The SIMPLE model structure that Fig. 2 provides for the embodiment of the present invention;
The denser network structure that Fig. 3 provides for the embodiment of the present invention;
The result schematic diagram that Fig. 4-A to Fig. 4-F detects for community for each algorithm that the embodiment of the present invention provides;
The community detection method structure drawing of device that Fig. 5 provides for the embodiment of the present invention.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is explicitly described, it is clear that, described embodiment is a part of embodiment of the present invention, rather than whole embodiments. Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under not making creative work premise, broadly fall into the scope of protection of the invention.
For ease of the understanding to the embodiment of the present invention, being further explained explanation below in conjunction with accompanying drawing with specific embodiment, embodiment is not intended that the restriction to the embodiment of the present invention.
Describing, for Fig. 1, the community detection method that the embodiment of the present invention one provides in detail below, the community detection method flow chart that Fig. 1 provides for the embodiment of the present invention, subject of implementation can be the terminal unit such as PC or desktop computer in embodiments of the present invention. As it is shown in figure 1, this embodiment specifically includes following steps:
The denser network structure that step 110, acquisition input.
Specifically, the embodiment of the present invention illustrates for terminal unit. The community detection method that the embodiment of the present invention provides can realize under SIMPLE model structure, as in figure 2 it is shown, be SIMPLE model structure.
Terminal obtains denser network structure, as it is shown on figure 3, have 10 nodes in this denser network structure, forms limit between each node.
Step 120, described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure.
Specifically, denser network structure is first sampled processing by terminal, then the network structure after sampling is carried out fusion treatment, obtains the sampling subgraph of denser network structure.
In embodiments of the present invention, sample process may proceed to less once.
Further, denser network structure is sampled fusion treatment by terminal, obtain the sampling subgraph of denser network structure, specifically include: denser network structure is carried out horizontal resection burst process by terminal, obtain multiple network structure burst and store the order information of network structure burst; Network structure burst is carried out stochastic sampling process by terminal, it is thus achieved that the sampling results of network structure burst; According to order information, sampling results is carried out fusion treatment by terminal, obtains the sampling subgraph of denser network structure.
In embodiments of the present invention, denser network structure level is specifically divided into a series of burst for terminal by described horizontal resection burst, and each burst keeps the integrity of web transactions. Such as, the fractionation order in Linux can be used for denser network structure is split. Each burst, according to the configuration of Hadoop distributed file system (HDFS), is sized to 64M. All bursts all will be stored on the HDFS of many station terminals, and storage position is transparent to user. Each computing node can share burst by access HDFS.
Further, network structure burst is carried out stochastic sampling process by terminal, also includes before: terminal determines the sampling proportion value n that network structure burst carries out stochastic sampling processv;
Described sampling proportion value nvMeet following formula one:
nv=min{dv,��(lndv+ ln2) } formula one
Wherein, described dvFor the degree of each node in described network structure burst; Described �� is nonnegative constant. The degree of described each node can be regarded as the limit number that each node connects.
Above-mentioned formula one can be determined by following theorem. Specifically, sampling figure to store the node pair of high similarity as much as possible, can safeguard the community structure of primitive network structure better with this. Therefore, in the embodiment of the present invention, first should determine that sampling proportion value nv. Due to more little nvRepresent the limit keeping more few in network structure, follow-up detecting step efficiency also can be improved a lot, but also can be much larger with this error brought.
Based on this, first the empty model link reference model as stochastic sampling is proposed. Given similarity threshold �� �� [0,1]; The neighbor choice function f of high similarityvFollowing formula two can be defined as:
Wherein, sv,u(sv,u�� [0,1]) between v and u, I (.) is the function of an index. fvValue be the ratio of neighbours of high similarity of v.
If s=nvIt is the sample size of node v �� V, { (v, u1),...,(v,us) it is the limit sampled. IfFor each j �� [1, s]. So, f can be obtained by the sampling limit of vvEstimated valueTherefore, as s >=3 (lndv+ln2)/fv��2Time, the error of estimation about Pr (| Xv-fv|�ݦ�fv)��1/dv. That is to say, as s >=3 (lndv+ln2)/fv��2Time, the error that sampling brings will significantly increase.
Further, sampling results is carried out fusion treatment by terminal, obtains the sampling subgraph of denser network structure, specifically includes: terminal is determined and belonged in sampling results with a stochastic sampling and have the sub-sample network of identical key assignments; Terminal will belong to a stochastic sampling and has the sub-sample network mapping of identical key assignments in same node, and described node includes multiple sub-sample network; Multiple sub-sample networks that node is included by terminal are as the sampling subgraph of denser network structure.
It should be noted that in embodiments of the present invention, each burst repeatable stochastic sampling r time, produce a sampling burst after sampling every time, therefore, it is right that terminal obtains a series of " key/value ". Wherein key [i], 1��i��r. Corresponding to i & lt stochastical sampling, value [i] records the sub-sample network of this sampling. Therefore, i-th " key/value " represents passing through following formula three:
Therefore, the sampling burst with identical key assignments will be mapped to that same node, and multiple sub-sample networks that this node includes are as the sampling subgraph of denser network structure.
Example as shown in Figure 3, this network structure is carried out horizontal resection burst process by terminal, it is assumed that network structure is cut into 3, and terminal obtains 3 bursts, and (node 1, node 2, node 3 belong to the first burst; Node 4, node 5, node 6 belong to the second burst; Node 7, node 8, node 9, node 10 belong to the 3rd burst). Meanwhile, terminal also stores the order information of burst; Three bursts are sampled processing by terminal respectively, obtain multiple sampling burst, and terminal is further according to order information, it is determined that belong to a stochastic sampling and the sub-sample network with identical key assignments in sampling results; Terminal will belong to a stochastic sampling and has the sub-sample network mapping of identical key assignments in same node, and described node includes multiple sub-sample network; Multiple sub-sample networks that node is included by terminal are as the sampling subgraph of denser network structure.
Step 130, utilize METIS algorithm that described sampling subgraph is detected, it is thus achieved that the initial community detection data of described sampling subgraph.
Specifically, utilizing METIS algorithm, sampling subgraph is detected by terminal, it is thus achieved that the initial community detection data of sampling subgraph.
Described initial community detection data are the label information of each sub-sample network.
Step 140, to described initial community detect data carry out clustering processing, it is thus achieved that Web Community's division result of described denser network structure.
Specifically, by KCC algorithm, terminal carries out clustering processing to finishing apprenticeship community's detection data, it is thus achieved that Web Community's division result of denser network structure.
It should be noted that in the step 120, each sub-sample network that sampling subgraph includes is individually for 1 community, that is to say that sampling subgraph has been divided into K community. The label (being initial community detection data) of each community is integrated into n* (rK) matrix ��={ ��1,��2,...��r}��
This step is obtaining on the basis of initial community detection data, and the concordance clustering algorithm (KCC) based on K-means obtains Web Community's division result. First KCC algorithm selects k node as initial barycenter. Based on COS distance, each node is assigned to its nearest centre of form by KCC algorithm, and then forms k bunch (i.e. community), then recalculates the barycenter of each bunch. Repeatedly repeat aforementioned iterative process, until the label of barycenter or community no longer changes.
In embodiments of the present invention, described METIS algorithm, KCC algorithm are existing algorithm, then this is not repeating its operation principle.
Therefore, by applying the community detection method that the embodiment of the present invention provides, denser network result is sampled fusion treatment by terminal, obtains sampling subgraph; Utilize METIS algorithm that sampling subgraph is detected, it is thus achieved that the initial community detection data of sampling subgraph; Initial community is detected data and carries out clustering processing by terminal, it is thus achieved that Web Community's division result of denser network structure. Owing in the present invention, denser network result is carried out multiple sampling fusion treatment by terminal, and then sampling subgraph is detected, after clustering processing, obtain Web Community's division result, and then solve Fast-Newman algorithm and LPA algorithm in existing algorithm and cannot find there is community structure restrictive problem from bigger limit denser network. Achieve and simply and efficiently find community structure, expanded being widely applied property.
In order to verify the effectiveness of preceding method, in one group of experimental data, choose 6 true social networkies and test as data set, the effectiveness being index measure algorithm with modularity (Q). Community's Detection results that final experimental data indicates preceding method is notable. The attribute of data set and modularity formula are respectively as shown in table 1 below and following formula four:
The attribute of table 1 data set
The result schematic diagram that each algorithm as shown in Fig. 4-A to Fig. 4-F detects for community, wherein, the SIMPLE of "-*-" line represents the community detection method that the embodiment of the present invention provides;Line represents PaK-means algorithm; "-��-" line represents METIS algorithm; "--" line represents Fast-Newman algorithm. From this 6 width figure, can be seen that the community detection method that the embodiment of the present invention provides has higher modularity.
Wherein, Fig. 4-A is Oklahoma network modularity under four kinds of algorithms; Fig. 4-B is UNC network modularity under four kinds of algorithms;Fig. 4-C is Twitter network modularity under four kinds of algorithms; Fig. 4-D is Gowalla network modularity under four kinds of algorithms; Fig. 4-E is Skitter network modularity under three kinds of algorithms; Fig. 4-F is LiveJournal network modularity under two kinds of algorithms.
The method that above-described embodiment describes all can realize overlapping community mining method, correspondingly, the embodiment of the present invention additionally provides a kind of community detecting device, in order to realize the community detection method provided in previous embodiment, as it is shown in figure 5, described device includes: acquiring unit 510, sampling integrated unit 520, detection unit 530 and cluster cell 540.
The acquiring unit 510 that described device includes, for obtaining the denser network structure of input;
Sampling integrated unit 520, for described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure;
Detection unit 530, is used for utilizing METIS algorithm that described sampling subgraph is detected, it is thus achieved that the initial community detection data of described sampling subgraph;
Cluster cell 540, carries out clustering processing for described initial community is detected data, it is thus achieved that Web Community's division result of described denser network structure.
Further, described sampling integrated unit 520 specifically includes:
Cutting subelement 521, for described denser network structure is carried out horizontal resection burst process, obtains multiple network structure burst and stores the order information of described network structure burst;
Sub-unit 522, for carrying out stochastic sampling process by described network structure burst, it is thus achieved that the sampling results of described network structure burst;
Fusant unit 523, for according to described order information, carrying out fusion treatment to described sampling results, obtain the sampling subgraph of described denser network structure.
Further, described sampling integrated unit 520 also includes:
Determine subelement 524, for determining the sampling proportion value n that described network structure burst is carried out stochastic sampling processv;
Described sampling proportion value nvMeet nv=min{dv,��(lndv+ ln2) };
Wherein, described dvFor the degree of each node in described network structure burst; Described �� is nonnegative constant.
Further, described fusant unit 523 also particularly useful for,
Determine and described sampling results belongs to a stochastic sampling and the sub-sample network with identical key assignments;
Belonging to described with a stochastic sampling and have the sub-sample network mapping of identical key assignments in same node, described node includes multiple described sub-sample network;
The multiple described sub-sample network included by described node is as the sampling subgraph of described denser network structure.
Therefore, by applying community's detecting device that the embodiment of the present invention provides, denser network result is sampled fusion treatment by described device, obtains sampling subgraph; Utilize METIS algorithm that sampling subgraph is detected, it is thus achieved that the initial community detection data of sampling subgraph; Initial community is detected data and carries out clustering processing by described device, it is thus achieved that Web Community's division result of denser network structure. Owing to denser network result is carried out multiple sampling fusion treatment by heretofore described device, and then sampling subgraph is detected, after clustering processing, obtain Web Community's division result, and then solve Fast-Newman algorithm and LPA algorithm in existing algorithm and cannot find there is community structure restrictive problem from bigger limit denser network.Achieve and simply and efficiently find community structure, expanded being widely applied property.
Professional should further appreciate that, the unit of each example described in conjunction with the embodiments described herein and algorithm steps, can with electronic hardware, computer software or the two be implemented in combination in, in order to clearly demonstrate the interchangeability of hardware and software, generally describe composition and the step of each example in the above description according to function. These functions perform with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme. Professional and technical personnel specifically can should be used for using different methods to realize described function to each, but this realization is it is not considered that beyond the scope of this invention.
The method described in conjunction with the embodiments described herein or the step of algorithm can use the software module that hardware, processor perform, or the combination of the two is implemented. Software module can be placed in any other form of storage medium known in random access memory (RAM), internal memory, read only memory (ROM), electrically programmable ROM, electrically erasable ROM, depositor, hard disk, moveable magnetic disc, CD-ROM or technical field.
Above-described detailed description of the invention; the purpose of the present invention, technical scheme and beneficial effect have been further described; it is it should be understood that; the foregoing is only the specific embodiment of the present invention; the protection domain being not intended to limit the present invention; all within the spirit and principles in the present invention, any amendment of making, equivalent replacement, improvement etc., should be included within protection scope of the present invention.
Claims (8)
1. a community detection method, it is characterised in that described method includes:
Obtain the denser network structure of input;
Described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure;
Utilize METIS algorithm that described sampling subgraph is detected, it is thus achieved that the initial community detection data of described sampling subgraph;
Data are detected in described initial community and carries out clustering processing, it is thus achieved that Web Community's division result of described denser network structure.
2. community detection method according to claim 1, it is characterised in that described described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure, specifically includes:
Described denser network structure is carried out horizontal resection burst process, obtains multiple network structure burst and store the order information of described network structure burst;
Described network structure burst is carried out stochastic sampling process, it is thus achieved that the sampling results of described network structure burst;
According to described order information, described sampling results is carried out fusion treatment, obtain the sampling subgraph of described denser network structure.
3. community detection method according to claim 2, it is characterised in that described described network structure burst is carried out stochastic sampling process, also includes before:
Determine the sampling proportion value n that described network structure burst is carried out stochastic sampling processv;
Described sampling proportion value nvMeet nv=min{dv,��(lndv+ ln2) };
Wherein, described dvFor the degree of each node in described network structure burst; Described �� is nonnegative constant.
4. community detection method according to claim 2, it is characterised in that described described sampling results is carried out fusion treatment, obtains the sampling subgraph of described denser network structure, specifically includes:
Determine and described sampling results belongs to a stochastic sampling and the sub-sample network with identical key assignments;
Belonging to described with a stochastic sampling and have the sub-sample network mapping of identical key assignments in same node, described node includes multiple described sub-sample network;
The multiple described sub-sample network included by described node is as the sampling subgraph of described denser network structure.
5. community's detecting device, it is characterised in that described device includes:
Acquiring unit, for obtaining the denser network structure of input;
Sampling integrated unit, for described denser network structure is sampled fusion treatment, obtains the sampling subgraph of described denser network structure;
Detection unit, is used for utilizing METIS algorithm that described sampling subgraph is detected, it is thus achieved that the initial community detection data of described sampling subgraph;
Cluster cell, carries out clustering processing for described initial community is detected data, it is thus achieved that Web Community's division result of described denser network structure.
6. community according to claim 5 detecting device, it is characterised in that described sampling integrated unit specifically includes:
Cutting subelement, for described denser network structure is carried out horizontal resection burst process, obtains multiple network structure burst and stores the order information of described network structure burst;
Sub-unit, for carrying out stochastic sampling process by described network structure burst, it is thus achieved that the sampling results of described network structure burst;
Fusant unit, for according to described order information, carrying out fusion treatment to described sampling results, obtain the sampling subgraph of described denser network structure.
7. community according to claim 6 detecting device, it is characterised in that described sampling integrated unit also includes:
Determine subelement, for determining the sampling proportion value n that described network structure burst is carried out stochastic sampling processv;
Described sampling proportion value nvMeet nv=min{dv,��(lndv+ ln2) };
Wherein, described dvFor the degree of each node in described network structure burst; Described �� is nonnegative constant.
8. community according to claim 6 detecting device, it is characterised in that described fusant unit also particularly useful for,
Determine and described sampling results belongs to a stochastic sampling and the sub-sample network with identical key assignments;
Belonging to described with a stochastic sampling and have the sub-sample network mapping of identical key assignments in same node, described node includes multiple described sub-sample network;
The multiple described sub-sample network included by described node is as the sampling subgraph of described denser network structure.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201511023521.XA CN105654132A (en) | 2015-12-30 | 2015-12-30 | Community detection method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201511023521.XA CN105654132A (en) | 2015-12-30 | 2015-12-30 | Community detection method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105654132A true CN105654132A (en) | 2016-06-08 |
Family
ID=56490882
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201511023521.XA Pending CN105654132A (en) | 2015-12-30 | 2015-12-30 | Community detection method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105654132A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111275597A (en) * | 2020-01-16 | 2020-06-12 | 华南理工大学 | Community life circle space identification method, system, computer equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103778192A (en) * | 2014-01-06 | 2014-05-07 | 华中师范大学 | Complex network local community discovery method |
CN104346481A (en) * | 2014-11-28 | 2015-02-11 | 西安电子科技大学 | Community detection method based on dynamic synchronous model |
CN104765733A (en) * | 2014-01-02 | 2015-07-08 | 华为技术有限公司 | Method and device for analyzing social network event |
-
2015
- 2015-12-30 CN CN201511023521.XA patent/CN105654132A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104765733A (en) * | 2014-01-02 | 2015-07-08 | 华为技术有限公司 | Method and device for analyzing social network event |
CN103778192A (en) * | 2014-01-06 | 2014-05-07 | 华中师范大学 | Complex network local community discovery method |
CN104346481A (en) * | 2014-11-28 | 2015-02-11 | 西安电子科技大学 | Community detection method based on dynamic synchronous model |
Non-Patent Citations (1)
Title |
---|
ZHIANG WU 等: ""SIMPLE:a simplifying-ensembling framework for parallel community detection from large networks"", 《CLUSTER COMPUT》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111275597A (en) * | 2020-01-16 | 2020-06-12 | 华南理工大学 | Community life circle space identification method, system, computer equipment and storage medium |
CN111275597B (en) * | 2020-01-16 | 2023-02-14 | 华南理工大学 | Community life circle space identification method, system, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Du et al. | The echo chamber effect in Twitter: does community polarization increase? | |
Kolda et al. | Counting triangles in massive graphs with MapReduce | |
Viggianiello et al. | Experimental generalized quantum suppression law in Sylvester interferometers | |
CN104408667B (en) | A kind of method and system of electric energy quality synthesis evaluation | |
CN106156145A (en) | The management method of a kind of address date and device | |
CN109034562A (en) | A kind of social networks node importance appraisal procedure and system | |
Chen et al. | Distinct counting with a self-learning bitmap | |
CN104598632A (en) | Hot event detection method and device | |
Chen et al. | Synthesis of n‐port resistive networks containing 2n terminals | |
CN102281075B (en) | Hierarchical encoding, operation and indexing method of hexagonal grid with aperture of 4 | |
CN105678626A (en) | Overlapped community excavation method and apparatus | |
CN107070897A (en) | Network log storage method based on many attribute Hash duplicate removals in intruding detection system | |
CN109684517A (en) | A kind of historical data storage method, reading/writing method, storage device and equipment | |
CN105654132A (en) | Community detection method and device | |
Horvát et al. | Connectedness matters: construction and exact random sampling of connected networks | |
CN107590160B (en) | Method and device for monitoring internal structure of radix tree to realize test | |
CN104317675B (en) | The disaster tolerance treating method and apparatus of application | |
Cable et al. | Exact and efficient simulation of concordant computation | |
CN105975634B (en) | The storage method of multidimensional ordered data in distributed data-storage system | |
Lai et al. | Correlation enhanced modularity-based belief propagation method for community detection in networks | |
CN114611713A (en) | Method and system for constructing tree model based on longitudinal federated learning | |
Yang et al. | Large-scale metagenomic sequence clustering on map-reduce clusters | |
Haraguchi et al. | B LOCK S UM is NP-Complete | |
CN103957012A (en) | Method and device for compressing DFA matrix | |
Zhang et al. | Detangling the multilayer structure from an aggregated network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160608 |
|
WD01 | Invention patent application deemed withdrawn after publication |