CN109543077A - Community search method - Google Patents
Community search method Download PDFInfo
- Publication number
- CN109543077A CN109543077A CN201811205006.7A CN201811205006A CN109543077A CN 109543077 A CN109543077 A CN 109543077A CN 201811205006 A CN201811205006 A CN 201811205006A CN 109543077 A CN109543077 A CN 109543077A
- Authority
- CN
- China
- Prior art keywords
- node
- community
- result
- search
- candidate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of community search methods, comprising: according to user for the demand of community search, node is corresponded to node variable, writes out corresponding search condition；Multiple search terms are converted by search condition；Each search terms are carried out with the community search of individual event condition；The result of each individual event condition community search is merged, takes union to return community's result.Search condition is unified for Boolean expression form by the present invention, and the execution of personalized expression search need and complex condition community search is carried out convenient for user；It is not intended to appear in node in community due to considering user in search, result more meets the expectation of user；Due to allowing to include at least one in given node demand in community, a search condition be likely to be obtained multiple and different communities as a result, and all meet search condition so that selection more horn of plenty of the user to result；A variety of different implementations are provided, can be selected according to actual needs.
Description
Technical field
The present invention relates to search technique field more particularly to a kind of community search methods of complex condition.
Background technique
Computer science, biology are widely present in by the network structure that the connection relationship between great deal of nodes and node is formed
The fields such as, sociology.In the relevant research work of network, community (community) is given more sustained attention by people.Community
Refer generally to contact even closer subgraph between the node that connection is inside and outside compared to community between an internal node.In network
The community structure excavated facilitates people and carries out friend recommendation, criminal gang's identification and protein function prediction.Community is searched
Rope (local community discovery) refers to given one or more nodes, finds the community comprising them, sends out compared to global community
Existing, it is more concerned about local network structure, and returns to more personalized community's result.
Current community search method is mainly based upon the special topological structures such as k-clique, k-core, k-truss,
In addition there are the community search methods for partially having comprehensively considered topological structure and nodal community.
The community that method based on k-truss structure is found needs to meet following property: 1, the triangle where each edge
Number is all not less than k-2；2, any two sides can be transferred through a series of adjacent triangles arrival.A kind of typical method is note
The truss value on all of its neighbor side around each node is recorded, is then organized into the abutment points of each node according to the truss value on side
The index of tree structure, referred to as TCP-index, last according to given node and k value, constantly finding out from index can expand
The neighbor node of exhibition has just obtained the community of the k-truss structure comprising giving node until it can not extend.
Comprehensively consider the typical method of topological structure and nodal community, such as AGAR method, is exactly according to the category between node
Property similarity first to original image carry out side supplement, thus building one TA-graph, then on TA-graph according to k-truss tie
Structure carries out community search, finally obtains the community comprising given node.
The deficiency of current community search method is that the community comprising given node collection can only be found.We reality into
When row community search, be frequently encountered demands some in this way: 1, community not only will comprising certain set points, while do not allow include
Other set points；2, community at least will be comprising giving any one in several nodes.Existing community search method can not
Meet the demand.
Therefore, the prior art needs to improve.
Summary of the invention
One technical problem to be solved by the embodiment of the invention is that: a kind of community search method is provided, it is existing to solve
Technology there are the problem of, the community search method includes:
According to user for the demand of community search, node is corresponded into node variable, writes out corresponding search condition；
Multiple search terms are converted by search condition；
Each search terms are carried out with the community search of individual event condition；
The result of each individual event condition community search is merged, i.e., takes union to return community's result.
In another embodiment based on the above-mentioned community search method of the present invention, it is described according to user for community search
Demand, by node correspond to node variable, writing out corresponding search condition includes:
The node referred in user demand is corresponded into Boolean variable；
The node for not allowing to appear in community is non-modified with logic, and the node that community must include is without modification；
The node that must be appeared in simultaneously in community is connected with logical AND, the node that community must include and the packet not allowed
The node contained, is also connected with logical AND；
Community must need to indicate with several of logic or connection node comprising at least one in several nodes.
In another embodiment based on the above-mentioned community search method of the present invention, it is described convert search condition to it is multiple
Search terms include:
The node variable valued combinations for meeting search condition are enumerated, are extracted model to obtain with the master of search condition equivalence
Formula；
It is most simple and or formula that principal disjunctive normal form, which is passed through Quine-McCluskey algorithm abbreviation,；
By it is most simple with or each conjunct of formula be set as search terms；
The variable non-modified without logic that frequency of occurrence is most in different conjuncts is extracted, if containing the change
The conjunct of amount is more than 1, then these conjuncts is merged into new search terms, repeats this step until can be into without conjunct
Row merges.
It is described that list is carried out for each search terms in another embodiment based on the above-mentioned community search method of the present invention
The community search of condition includes:
For the search terms of conjunction expression form, the node for including is not allowed node that wherein community must include and to distinguish
It is organized into necessary node collection and forbids node collection as the input of individual event condition community search process；
For the search terms merged by multiple conjunction expressions, one or more common node variable root that will be extracted
According to whether capable of appearing in be organized into community and necessary node collection and forbid node collection as individual event condition community search process
Input, remainder are used for the differentiation to output result；
Individual event condition community search is carried out, using necessary node collection and node collection is forbidden to search for community's knot from network
Fruit, making gained community includes necessary node collection, while not including the node for forbidding node to concentrate.
In another embodiment based on the above-mentioned community search method of the present invention, the progress individual event condition community is searched
Rope using necessary node collection and forbids node collection search for community from network as a result, gained community is made to include necessity node collection,
Not including the node for forbidding node to concentrate simultaneously includes three kinds of implementations, is respectively as follows: the mode of community search after filtering, weighting
The mode of filtering, the mode filtered in search.
In another embodiment based on the above-mentioned community search method of the present invention, the mode of community search after the filtering
Include:
The remove ban node collection from network is obtained without the network for forbidding node；
To new network, necessary node collection is used to carry out community search as input.
In another embodiment based on the above-mentioned community search method of the present invention, the mode of the weighted filter includes:
It is assigned to numerical value weight for nodes all in network, enabling necessary node is 1, and forbidding node is -1, remaining node is
0；
Other than in addition to necessary node and forbidding node, iteration updates the weight of each node, is assigned a value of all neighbours
The mean value of node weights, it may be assumed that
Node weights threshold value λ is set, retains the node that node weights are more than or equal to λ, and by its leading in former network
Subgraph is extracted as new network out；
Necessary node collection is divided into community's result C by a01；
A02 obtains the corresponding induced subgraph of community's result according to given network, if export of community result C
Figure only has the node degree of a connected component and all nodes in induced subgraph and is both greater than equal to given threshold value k, then stops
Only and return to community's result；
Node in the same component is divided into same group according to the connected component of induced subgraph by a03；
The neighbor node of all nodes in community result C is divided into candidate node set Candidate, and excluded by a04
The node being present in community result C；
Community result C is set to empty set, and go to step a08 if Candidate is sky by a05；
A06 records its number a for connecting different connected components to each node in candidate node set Candidate,
Connect the node number b in community's result C, point degree d of the node in given network, later according to a, b and d to node collection
In node carry out multiple key descending sort；
A07, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set
It is removed in Candidate, goes to step a06, otherwise, community is added as a result, simultaneously adding the neighbor node of the node in node c
Enter to candidate node set Candidate, and node c is removed from candidate node set Candidate, goes to step a02；
The node of whole network figure is divided into community's result C by a08；
A09 stops and returns empty set, if society if the connected component number of the induced subgraph of community result C is greater than 1
The degree of smallest point is more than or equal to threshold value k in the induced subgraph of area result C, then stops and return to community result C；
A10, the node by degree in the induced subgraph of community result C lower than k are deleted from community result C, if deleted
Node be the member of necessary node collection, then stopping and returning empty set, otherwise go to step a09.
It is described to use necessary node collection as input in another embodiment based on the above-mentioned community search method of the present invention
Carrying out community search to new network includes:
Necessary node collection is divided into community's result C by b01；
B02 obtains the corresponding induced subgraph of community's result according to given network, if export of community result C
Figure only has the node degree of a connected component and all nodes in induced subgraph and is both greater than equal to given threshold value k, then stops
Only and return to community's result；
Node in the same component is divided into same group according to the connected component of induced subgraph by b03；
The neighbor node of all nodes in community result C is divided into candidate node set Candidate, and excluded by b04
The node being present in community result C；
Community result C is set to empty set, and go to step b08 if Candidate is sky by b05；
B06 records its number a for connecting different connected components to each node in candidate node set Candidate,
Connect the node number b in community's result C, point degree d of the node in given network, later according to the section concentrated to node
Point carries out multiple key descending sort；
B07, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set
It is removed in Candidate, goes to step b06, otherwise, community is added as a result, simultaneously adding the neighbor node of the node in node c
Enter to candidate node set Candidate, and node c is removed from candidate node set Candidate, goes to step b02；
The node of whole network figure is divided into community's result C by b08；
B09 stops and returns empty set, if society if the connected component number of the induced subgraph of community result C is greater than 1
The degree of smallest point is more than or equal to threshold value k in the induced subgraph of area result C, then stops and return to community result C；
B10, the node by degree in the induced subgraph of community result C lower than k are deleted from community result C, if deleted
Node be the member of necessary node collection, then stopping and returning empty set, otherwise go to step b09.
In another embodiment based on the above-mentioned community search method of the present invention, the mode packet filtered in search
It includes:
Necessary node collection is divided into community's result C by c01；
C02 obtains the corresponding induced subgraph of community result C according to given network.If the export of community result C
Subgraph only has a connected component and node degree of all nodes in induced subgraph is both greater than equal to given threshold value k, then
Stop and returns to community result C；
Node in the same component is divided into same group according to the connected component of induced subgraph by c03；
The neighbor node of all nodes in community result C is divided into candidate node set Candidate, and therefrom excluded by c04
The node that is already present in community result C and forbid node；
Community result C is set to empty set, and go to step c08 if Candidate is sky by c05；
C06 records its number for connecting different connected components for each node in candidate node set Candidate
A, connects the node number b in community's result C, node in given network with the non-number d for forbidding node to connect；According to a,
B and d carries out multiple key descending sort to the node that node is concentrated；
C07, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set
It is removed in Candidate, goes to step c06, otherwise, community's result C is added in node c, while by the neighbor node of the node
It is added to candidate node set Candidate, and node c is removed from candidate node set Candidate, goes to step c02；
The node of whole network figure is divided into community's result C and remove ban node by c08；
C09 stops and returns empty set, if society if the connected component number of the induced subgraph of community result C is greater than 1
Whether the degree of smallest point is more than or equal to threshold value k in the induced subgraph of area result C, check to contain in community's result C at this time and forbid
Node；
C10, the node by degree in the induced subgraph of community result C lower than k are deleted from community result C, if deleted
Node be necessary node collection member, then stop and return empty set, otherwise go to step c09.
Compared with prior art, the invention has the following advantages that
The invention proposes a kind of community search methods, and search condition is carried out table in the form of unified Boolean expression
Show, carries out personalized expression search need convenient for user, be also convenient for the execution of complex condition community search；Due to being searched in community
User is considered during rope and is not intended to appear in node in community, and the community's result searched for more meets the phase of user
It hopes, so that community's result is more personalized；Such demand in given node is included at least in community due to allowing to consider,
To a search condition be likely to be obtained multiple and different communities as a result, and all meet search condition, this makes user to result
Selection more horn of plenty；A variety of different implementations are provided, can be selected according to actual needs.
Below by drawings and examples, technical scheme of the present invention will be described in further detail.
Detailed description of the invention
The attached drawing for constituting part of specification describes the embodiment of the present invention, and together with description for explaining
The principle of the present invention.
The present invention can be more clearly understood according to following detailed description referring to attached drawing, in which:
Fig. 1 is the flow chart of one embodiment of community search method of the invention；
Fig. 2 is the flow chart of another embodiment of community search method of the invention；
Fig. 3 is the flow chart of another embodiment of community search method of the invention；
Fig. 4 is the flow chart of another embodiment of community search method of the invention；
Fig. 5 is the flow chart of another embodiment of community search method of the invention；
Fig. 6 is the flow chart of another embodiment of community search method of the invention；
Fig. 7 is the flow chart of another embodiment of community search method of the invention；
Fig. 8 is the flow chart of another embodiment of community search method of the invention.
Specific embodiment
Carry out the various exemplary embodiments of detailed description of the present invention now with reference to attached drawing.It should also be noted that unless in addition having
Body explanation, the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
The range of invention.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as part of specification.
Fig. 1 is the flow chart of one embodiment of community search method of the invention, as shown in Figure 1, the community search
Method includes:
10, according to user for the demand of community search, node is corresponded into node variable, writes out corresponding searching bar
Part, node variable are also known as Boolean variable, and search condition is indicated with Boolean expression；
20, multiple search terms are converted by search condition；
30, each search terms are carried out with the community search of individual event condition；
40, the result of each individual event condition community search is merged, i.e., takes union to return community's result.
Fig. 2 be community search method of the invention another embodiment flow chart, as shown in Fig. 2, it is described according to
Node is corresponded to node variable, writing out corresponding search condition includes: for the demand of community search by family
101, the node referred in user demand is corresponded into Boolean variable；
102, the node for not allowing to appear in community is non-modified with logic, and the node that community must include is without modification, institute
The expression symbol for stating logic NOT is
103, it is necessary to while appearing in the node in community and being connected with logical AND, the node and do not allow that community must include
The node for including, also connected with logical AND, the expression symbol of the logical AND is " ∧ ", such as: Boolean expressionIndicate that user wishes that community must not allow comprising node A and node B comprising node C.；
104, community must need to carry out table with several of logic or connection node comprising at least one in several nodes
Show, the logic or expression symbol be " ∨ ", such as: Boolean expression A ∨ B ∨ C indicate user wish that community includes at least
One in node A, node B and node C.
Fig. 3 is the flow chart of another embodiment of community search method of the invention, as shown in figure 3, described will search
Condition is converted into multiple search terms
201, the node variable valued combinations for meeting search condition are enumerated, are extracted to obtain with the master of search condition equivalence
Normal form；
202, it is most simple and or formula that principal disjunctive normal form, which is passed through Quine-McCluskey algorithm abbreviation,；
203, by it is most simple with or each conjunct of formula be set as search terms, such as: it is most simple with or formula (A ∧ B) ∨ (C ∧
It D is (A ∧ B) and (C ∧ D) respectively) just comprising two search terms, if it find that with the presence of the identical node of several search terms
Variable, then these search terms can be merged into a new search terms, such as: most simple and or formula
Two search terms (A ∧ B) andContaining same node point variables A, common node variables A can then be extracted, it will
The two conjuncts are merged intoFor reduce search terms number, to reduce subsequent progress individual event condition society
The number of area searching process achievees the purpose that save time overhead；
204, the variable non-modified without logic that frequency of occurrence is most in different conjuncts is extracted, if contained
The conjunct of the variable is more than 1, then these conjuncts is merged into new search terms, repeats this step until can without conjunct
To merge.
Fig. 4 is the flow chart of another embodiment of community search method of the invention, as shown in figure 4, described for every
One search terms carry out individual event condition community search include:
301, for the search terms of conjunction expression form, do not allow node that wherein community must include and the node for including
It is organized into necessary node collection respectively and forbids node collection as the input of individual event condition community search process, this is because it is only wrapped
The node occurred is not allowed containing the node that must occur in community and；
302, for the search terms merged by multiple conjunction expressions, one or more common node extracted is become
Amount is organized into necessary node collection and forbids node collection as individual event condition community search mistake according to whether can appear in community
The input of journey, remainder are used for the differentiation to output result, such as: in two search terms (A ∧ B) andIt is merged into
Search termsIn, input of the necessary node collection { A } as individual event condition community search process, i.e., for finding
Community comprising node A,As output result discriminate, i.e., for judge community's result whether contain node B or
Person does not contain node D；
303, individual event condition community search is carried out, using necessary node collection and node collection is forbidden to search for community from network
As a result, gained community is made to include necessary node collection, while the node for forbidding node to concentrate is not included.
The progress individual event condition community search using necessary node collection and forbids node collection to search for community from network
As a result, make gained community include necessary node collection, while not including the node for forbidding node to concentrate includes three kinds of implementations, point
Not are as follows: the mode of community search, the mode of weighted filter, the mode filtered in search after filtering.
The mode of community search includes: after the filtering
The remove ban node collection from network is obtained without the network for forbidding node；
To new network, necessary node collection is used to carry out community search as input.
Fig. 5 is the flow chart of another embodiment of community search method of the invention, as shown in figure 5, described weighted
The mode of filter includes:
401, it is assigned to numerical value weight for all nodes in network, enabling necessary node is 1, and forbidding node is -1, remaining section
Point is 0, it may be assumed that
402, other than in addition to necessary node and forbidding node, iteration updates the weight of each node, is assigned a value of owning
The mean value of neighbor node weight, it may be assumed that
403, node weights threshold value λ is set, retains the node that node weights are more than or equal to λ, and by it in former network
Induced subgraph extract as new network；
404, use necessary node collection to carry out community search to new network as input.
Fig. 6 is the flow chart of another embodiment of community search method of the invention, as shown in fig. 6, described to new
Network, using necessary node collection as progress community search is inputted includes:
501, necessary node collection is divided into community's result C；
502, according to given network, the corresponding induced subgraph of community's result is obtained, if export of community result C
Figure only has the node degree of a connected component and all nodes in induced subgraph and is both greater than equal to given threshold value k, then stops
Only and return to community's result；
503, according to the connected component of induced subgraph, the node in the same component is divided into same group；
504, the neighbor node of all nodes in community result C is divided into candidate node set Candidate, and exclude
The node being present in community result C；
505, if Candidate is sky, community result C is set to empty set, and go to step 508；
506, to each node in candidate node set Candidate, its number a for connecting different connected components is recorded,
Connect the node number b in community's result C, point degree d of the node in given network, later according to a, b and d to node collection
In node carry out multiple key descending sort；
507, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set
It is removed in Candidate, goes to step 506, otherwise, community is added as a result, simultaneously adding the neighbor node of the node in node c
Enter to candidate node set Candidate, and node c is removed from candidate node set Candidate, goes to step 502；
508, the node of whole network figure is divided into community's result C；
509, if the connected component number of the induced subgraph of community result C is greater than 1, stop and return empty set, if society
The degree of smallest point is more than or equal to threshold value k in the induced subgraph of area result C, then stops and return to community result C；
510, the node by degree in the induced subgraph of community result C lower than k is deleted from community result C, if deleted
Node be otherwise the member of necessary node collection goes to step 509 then stopping and returning empty set.
Fig. 7 is the flow chart of another embodiment of community search method of the invention, as shown in fig. 7, described with necessary
Node collection carries out community search to new network as input
601, necessary node collection is divided into community's result C；
602, according to given network, the corresponding induced subgraph of community's result is obtained, if export of community result C
Figure only has the node degree of a connected component and all nodes in induced subgraph and is both greater than equal to given threshold value k, then stops
Only and return to community's result；
603, according to the connected component of induced subgraph, the node in the same component is divided into same group；
604, the neighbor node of all nodes in community result C is divided into candidate node set Candidate, and exclude
The node being present in community result C；
605, if Candidate, community result C is set to empty set, and go to step 608；
606, to each node in candidate node set Candidate, its number a for connecting different connected components is recorded,
Connect the node number b in community's result C, point degree d of the node in given network, later according to a, b and d to node collection
In node carry out multiple key descending sort；
607, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set
It is removed in Candidate, goes to step 606, otherwise, community is added as a result, simultaneously adding the neighbor node of the node in node c
Enter to candidate node set Candidate, and node c is removed from candidate node set Candidate, goes to step 602；
608, the node of whole network figure is divided into community's result C；
609, if the connected component number of the induced subgraph of community result C is greater than 1, stop and return empty set, if society
The degree of smallest point is more than or equal to threshold value k in the induced subgraph of area result C, then stops and return to community result C；
610, the node by degree in the induced subgraph of community result C lower than k is deleted from community result C, if deleted
Node be otherwise the member of necessary node collection goes to step 609 then stopping and returning empty set.
Fig. 8 is the flow chart of another embodiment of community search method of the invention, as shown in figure 8, the side is searched for
Side filtering mode include:
701, necessary node collection is divided into community's result C；
702, according to given network, obtain the corresponding induced subgraph of community result C.If the export of community result C
Subgraph only has a connected component and node degree of all nodes in induced subgraph is both greater than equal to given threshold value k, then
Stop and returns to community result C；
703, according to the connected component of induced subgraph, the node in the same component is divided into same group；
704, the neighbor node of all nodes in community result C is divided into candidate node set Candidate, and therefrom exclude
The node that is already present in community result C and forbid node；
705, if Candidate is sky, community result C is set to empty set, and go to step 708；
706, for each node in candidate node set Candidate, record its number for connecting different connected components
A, connects the node number b in community's result C, node in given network with the non-number d for forbidding node to connect；According to a,
B and d carries out multiple key descending sort to the node that node is concentrated；
707, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set
It is removed in Candidate, goes to step 706, otherwise, community's result C is added in node c, while by the neighbor node of the node
It is added to candidate node set Candidate, and node c is removed from candidate node set Candidate；
708, the node of whole network figure is divided into community's result C and remove ban node；
709, if the connected component number of the induced subgraph of community result C is greater than 1, stop and return empty set, if society
The degree of smallest point is more than or equal to threshold value k in the induced subgraph of area result C, then stops and return to community result C；
710, degree in the induced subgraph of community result C is lower than to the knot removal of k, if the node deleted is necessary section
The member of point set then stops and returns empty set, otherwise jumps to step 709.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with its
The difference of its embodiment, the same or similar part cross-reference between each embodiment.For system embodiment
For, since it is substantially corresponding with embodiment of the method, so being described relatively simple, referring to the portion of embodiment of the method in place of correlation
It defends oneself bright.
Description of the invention is given for the purpose of illustration and description, and is not exhaustively or will be of the invention
It is limited to disclosed form.Many modifications and variations are obvious for the ordinary skill in the art.It selects and retouches
It states embodiment and is to more preferably illustrate the principle of the present invention and practical application, and those skilled in the art is enable to manage
The solution present invention is to design various embodiments suitable for specific applications with various modifications.
Claims (10)
1. a kind of community search method characterized by comprising
According to user for the demand of community search, node is corresponded into node variable, writes out corresponding search condition；
Multiple search terms are converted by search condition；
Each search terms are carried out with the community search of individual event condition；
The result of each individual event condition community search is merged, i.e., takes union to return community's result.
2. community search method according to claim 1, which is characterized in that it is described according to user for the need of community search
It asks, node is corresponded into node variable, writing out corresponding search condition includes:
The node referred in user demand is corresponded into Boolean variable；
The node for not allowing to appear in community is non-modified with logic, and the node that community must include is without modification；
The node that must be appeared in simultaneously in community is connected with logical AND, node that community must include and what is do not allowed include
Node is also connected with logical AND；
Community must need to indicate with several of logic or connection node comprising at least one in several nodes.
3. community search method according to claim 2, which is characterized in that described to convert multiple search for search condition
Include:
The node variable valued combinations for meeting search condition are enumerated, to obtain the principal disjunctive normal form with search condition equivalence；
It is most simple and or formula that principal disjunctive normal form, which is passed through Quine-McCluskey algorithm abbreviation,；
By it is most simple with or each conjunct of formula be set as search terms；
The variable non-modified without logic that frequency of occurrence is most in different conjuncts is extracted, if containing the variable
Conjunct is more than 1, then these conjuncts is merged into new search terms, repeats this step until can be closed without conjunct
And.
4. community search method according to claim 3, which is characterized in that described to carry out individual event item for each search terms
The community search of part includes:
For the search terms of conjunction expression form, the node for including is not allowed node that wherein community must include and to arrange respectively
At necessary node collection and forbid node collection as the input of individual event condition community search process；
For the search terms merged by multiple conjunction expressions, by one or more common node variable extracted according to being
No can appear in is organized into necessary node collection and forbids node collection as the input of individual event condition community search process in community,
Remainder is used for the differentiation to output result；
Individual event condition community search is carried out, using necessary node collection and node collection is forbidden to search for community from network as a result, making
Gained community includes necessary node collection, while not including the node for forbidding node to concentrate.
5. community search method according to claim 4, which is characterized in that the progress individual event condition community search, benefit
With necessary node collection and node collection is forbidden to search for community from network as a result, making gained community include necessary node collection, simultaneously
Include three kinds of implementations not comprising the node for forbidding node to concentrate, is respectively as follows: the mode of community search, weighted filter after filtering
Mode, the mode that filters in search.
6. community search method according to claim 5, which is characterized in that the mode packet of community search after the filtering
It includes:
The remove ban node collection from network is obtained without the network for forbidding node；
To new network, necessary node collection is used to carry out community search as input.
7. community search method according to claim 5, which is characterized in that the mode of the weighted filter includes:
It is assigned to numerical value weight for nodes all in network, enabling necessary node is 1, and forbidding node is -1, remaining node is 0；
Other than in addition to necessary node and forbidding node, iteration updates the weight of each node, is assigned a value of all neighbor nodes
The mean value of weight, it may be assumed that
Node weights threshold value λ is set, retains the node that node weights are more than or equal to λ, and its export in former network is sub
Figure is extracted as new network；
Necessary node collection is used to carry out community search to new network as input.
8. community search method according to claim 6, which is characterized in that it is described to new network, with necessary node
Collection carries out community search as input
Necessary node collection is divided into community's result C by a01；
A02 obtains the corresponding induced subgraph of community's result according to given network, if the induced subgraph of community result C is only
There is a connected component and node degree of all nodes in induced subgraph is both greater than equal to given threshold value k, then stops simultaneously
Return to community's result；
Node in the same component is divided into same group according to the connected component of induced subgraph by a03；
The neighbor node of all nodes in community result C is divided into candidate node set Candidate, and excludes to have existed by a04
Node in community result C；
Community result C is set to empty set, and go to step a08 if Candidate is sky by a05；
A06 records its number a for connecting different connected components to each node in candidate node set Candidate, connection
Node number b in community result C, point degree d of the node in given network, later concentrates node according to a, b and d
Node carries out multiple key descending sort；
A07, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set Candidate
Removal, goes to step a06, otherwise, community is added as a result, the neighbor node of the node is added to candidate section simultaneously in node c
Point set Candidate, and node c is removed from candidate node set Candidate, go to step a02；
The node of whole network figure is divided into community's result C by a08；
A09 stops and returns empty set if the connected component number of the induced subgraph of community result C is greater than 1, if community is tied
The degree of smallest point is more than or equal to threshold value k in the induced subgraph of fruit C, then stops and return to community result C；
A10, the node by degree in the induced subgraph of community result C lower than k are deleted from community result C, if the section deleted
Point is otherwise the member of necessary node collection goes to step a09 then stopping and returning empty set.
9. community search method according to claim 7, which is characterized in that described to use necessary node collection as input to new
Network carry out community search include:
Necessary node collection is divided into community's result C by b01；
B02 obtains the corresponding induced subgraph of community's result according to given network, if the induced subgraph of community result C is only
There is a connected component and node degree of all nodes in induced subgraph is both greater than equal to given threshold value k, then stops simultaneously
Return to community's result；
Node in the same component is divided into same group according to the connected component of induced subgraph by b03；
The neighbor node of all nodes in community result C is divided into candidate node set Candidate, and excludes to have existed by b04
Node in community result C；
Community result C is set to empty set, and go to step b08 if Candidate is sky by b05；
B06 records its number a for connecting different connected components to each node in candidate node set Candidate, connection
Node number b in community result C, point degree d of the node in given network, later according to the node that node is concentrated into
Row multiple key descending sort；
B07, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set Candidate
Removal, goes to step b06, otherwise, community is added as a result, the neighbor node of the node is added to candidate section simultaneously in node c
Point set Candidate, and node c is removed from candidate node set Candidate, go to step b02；
The node of whole network figure is divided into community's result C by b08；
B09 stops and returns empty set if the connected component number of the induced subgraph of community result C is greater than 1, if community is tied
The degree of smallest point is more than or equal to threshold value k in the induced subgraph of fruit C, then stops and return to community result C；
B10, the node by degree in the induced subgraph of community result C lower than k are deleted from community result C, if the section deleted
Point is otherwise the member of necessary node collection goes to step b09 then stopping and returning empty set.
10. community search method according to claim 5, which is characterized in that the mode that filters in search includes:
Necessary node collection is divided into community's result C by c01；
C02 obtains the corresponding induced subgraph of community result C according to given network.If the induced subgraph of community result C
Only the node degree of a connected component and all nodes in induced subgraph is both greater than equal to given threshold value k, then stops
And return to community result C；
Node in the same component is divided into same group according to the connected component of induced subgraph by c03；
The neighbor node of all nodes in community result C is divided into candidate node set Candidate, and therefrom excluded by c04
The node that is present in community result C and forbid node；
Community result C is set to empty set, and go to step c08 if Candidate is sky by c05；
C06 records its number a for connecting different connected components for each node in candidate node set Candidate, even
Meet the node number b in community result C, node in given network with the non-number d for forbidding node to connect；According to a, b and d
Multiple key descending sort is carried out to the node that node is concentrated；
C07, if the degree of the both candidate nodes c to rank the first is less than threshold value k, by node c from candidate node set Candidate
Removal, goes to step c06, otherwise, community's result C is added in node c, while the neighbor node of the node is added to candidate section
Point set Candidate, and node c is removed from candidate node set Candidate, go to step c02；
The node of whole network figure is divided into community's result C and remove ban node by c08；
C09 stops and returns empty set if the connected component number of the induced subgraph of community result C is greater than 1, if community is tied
Whether the degree of smallest point is more than or equal to threshold value k in the induced subgraph of fruit C, check to contain in community's result C at this time and forbid node；
C10, the node by degree in the induced subgraph of community result C lower than k are deleted from community result C, if the section deleted
Point is the member of necessary node collection, then stops and return empty set, otherwise go to step c09.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811205006.7A CN109543077B (en) | 2018-10-16 | 2018-10-16 | Community search method |
| PCT/CN2019/111419 WO2020078370A1 (en) | 2018-10-16 | 2019-10-16 | Community search method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811205006.7A CN109543077B (en) | 2018-10-16 | 2018-10-16 | Community search method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109543077A true CN109543077A (en) | 2019-03-29 |
| CN109543077B CN109543077B (en) | 2020-07-31 |
Family
ID=65843813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811205006.7A Active CN109543077B (en) | 2018-10-16 | 2018-10-16 | Community search method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109543077B (en) |
| WO (1) | WO2020078370A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020078370A1 (en) * | 2018-10-16 | 2020-04-23 | 清华大学 | Community search method |
| CN113254797A (en) * | 2021-04-19 | 2021-08-13 | 江汉大学 | Searching method, device and processing equipment for social network community |
| CN116485587A (en) * | 2023-04-21 | 2023-07-25 | 深圳润高智慧产业有限公司 | Community service acquisition method, community service providing method, electronic device and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103425662A (en) * | 2012-05-16 | 2013-12-04 | 腾讯科技（深圳）有限公司 | Information search method and device in network community |
| US8972557B2 (en) * | 2012-02-28 | 2015-03-03 | Samsung Electronics Co., Ltd. | Topic-based community index generation apparatus and method and topic-based community searching apparatus and method |
| CN105224555A (en) * | 2014-06-12 | 2016-01-06 | 北京搜狗科技发展有限公司 | A kind of methods, devices and systems of search |
| US20170032044A1 (en) * | 2006-11-14 | 2017-02-02 | Paul Vincent Hayes | System and Method for Personalized Search While Maintaining Searcher Privacy |
| CN106530039A (en) * | 2016-10-26 | 2017-03-22 | 深圳市亿家信息科技有限公司 | Data processing realization method and system of intelligent community |
| US9652875B2 (en) * | 2012-10-29 | 2017-05-16 | Yahoo! Inc. | Systems and methods for generating a dense graph |
| JP2017097449A (en) * | 2015-11-18 | 2017-06-01 | カシオ計算機株式会社 | Information processing system, electronic apparatus and program |
| US20170235848A1 (en) * | 2012-08-29 | 2017-08-17 | Dennis Van Dusen | System and method for fuzzy concept mapping, voting ontology crowd sourcing, and technology prediction |
| JP6332243B2 (en) * | 2015-11-18 | 2018-05-30 | カシオ計算機株式会社 | Information processing apparatus, electronic device, and program |
| CN108268603A (en) * | 2017-12-22 | 2018-07-10 | 中国电子科技集团公司第三十研究所 | A kind of community discovery method based on core member's identification |
| CN108319728A (en) * | 2018-03-15 | 2018-07-24 | 深圳大学 | A kind of frequent community search method and system based on k-star |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011032077A2 (en) * | 2009-09-11 | 2011-03-17 | University Of Maryland, College Park | System and method for data management in large data networks |
| US20160063110A1 (en) * | 2014-08-29 | 2016-03-03 | Matthew David Shoup | User interface for generating search queries |
| CN104636978B (en) * | 2015-02-12 | 2017-11-14 | 西安电子科技大学 | A kind of overlapping community detection method propagated based on multi-tag |
| CN109543077B (en) * | 2018-10-16 | 2020-07-31 | 清华大学 | Community search method |
-
2018
- 2018-10-16 CN CN201811205006.7A patent/CN109543077B/en active Active
-
2019
- 2019-10-16 WO PCT/CN2019/111419 patent/WO2020078370A1/en active Application Filing
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170032044A1 (en) * | 2006-11-14 | 2017-02-02 | Paul Vincent Hayes | System and Method for Personalized Search While Maintaining Searcher Privacy |
| US8972557B2 (en) * | 2012-02-28 | 2015-03-03 | Samsung Electronics Co., Ltd. | Topic-based community index generation apparatus and method and topic-based community searching apparatus and method |
| CN103425662A (en) * | 2012-05-16 | 2013-12-04 | 腾讯科技（深圳）有限公司 | Information search method and device in network community |
| US20170235848A1 (en) * | 2012-08-29 | 2017-08-17 | Dennis Van Dusen | System and method for fuzzy concept mapping, voting ontology crowd sourcing, and technology prediction |
| US9652875B2 (en) * | 2012-10-29 | 2017-05-16 | Yahoo! Inc. | Systems and methods for generating a dense graph |
| CN105224555A (en) * | 2014-06-12 | 2016-01-06 | 北京搜狗科技发展有限公司 | A kind of methods, devices and systems of search |
| JP2017097449A (en) * | 2015-11-18 | 2017-06-01 | カシオ計算機株式会社 | Information processing system, electronic apparatus and program |
| JP6332243B2 (en) * | 2015-11-18 | 2018-05-30 | カシオ計算機株式会社 | Information processing apparatus, electronic device, and program |
| CN106530039A (en) * | 2016-10-26 | 2017-03-22 | 深圳市亿家信息科技有限公司 | Data processing realization method and system of intelligent community |
| CN108268603A (en) * | 2017-12-22 | 2018-07-10 | 中国电子科技集团公司第三十研究所 | A kind of community discovery method based on core member's identification |
| CN108319728A (en) * | 2018-03-15 | 2018-07-24 | 深圳大学 | A kind of frequent community search method and system based on k-star |
Non-Patent Citations (3)
| Title |
|---|
| JUN CHEN,CHAOKUN WANG,JIANMIN WANG: "Recommendation for Repeat Consumption from", 《2017 IEEE 33RD INTERNATIONAL CONFERENCE ON DATA ENGINEERING (ICDE)》 * |
| LI JIN,LING FENG,GANGLI LIU,CHAOKUN WANG: "Personal Web Revisitation by Context and Content Keywords with Relevance Feedback", 《 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING》 * |
| 钱珺: "基于社区的动态网络节点介数中心度更新算法", 《清华大学 软件学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020078370A1 (en) * | 2018-10-16 | 2020-04-23 | 清华大学 | Community search method |
| CN113254797A (en) * | 2021-04-19 | 2021-08-13 | 江汉大学 | Searching method, device and processing equipment for social network community |
| CN113254797B (en) * | 2021-04-19 | 2022-09-20 | 江汉大学 | Searching method, device and processing equipment for social network community |
| CN116485587A (en) * | 2023-04-21 | 2023-07-25 | 深圳润高智慧产业有限公司 | Community service acquisition method, community service providing method, electronic device and storage medium |
| CN116485587B (en) * | 2023-04-21 | 2024-04-09 | 深圳润高智慧产业有限公司 | Community service acquisition method, community service providing method, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020078370A1 (en) | 2020-04-23 |
| CN109543077B (en) | 2020-07-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7814105B2 (en) | Method for domain identification of documents in a document database | |
| David | Association rule ontology matching approach | |
| EP1552437B1 (en) | Information extraction using an object based semantic network | |
| CN109543077A (en) | Community search method | |
| US20110016113A1 (en) | Method for re-ranking documents retrieved from a document database | |
| US20080168070A1 (en) | Method and apparatus for classifying multimedia artifacts using ontology selection and semantic classification | |
| EP1076299A2 (en) | A method to efficiently partition large hyperlinked databases by hyperlink structure | |
| Tao et al. | Nearest keyword search in xml documents | |
| CN105868366B (en) | Concept space navigation method based on concept association | |
| Simsek et al. | Decentralized search in networks using homophily and degree disparity | |
| KR20130098772A (en) | Topic-based community index generation apparatus, topic-based community searching apparatus, topic-based community index generation method and topic-based community searching method | |
| CN104199875A (en) | Search recommending method and device | |
| CN105740387B (en) | A kind of scientific and technical literature recommended method based on author's frequent mode | |
| Boden et al. | MiMAG: mining coherent subgraphs in multi-layer graphs with edge labels | |
| Bhardwaj et al. | A novel approach for content extraction from web pages | |
| Kumara et al. | Cluster-based web service recommendation | |
| Vu et al. | An efficient density-based clustering with side information and active learning: a case study for facial expression recognition task | |
| Asiler et al. | Bb-graph: A subgraph isomorphism algorithm for efficiently querying big graph databases | |
| Moreno et al. | SMOT+: Extending the SMOT algorithm for discovering stops in nested sites | |
| KR101693727B1 (en) | Apparatus and method for reorganizing social issues from research and development perspective using social network | |
| Cheng et al. | Context-based page unit recommendation for web-based sensemaking tasks | |
| CN106296537B (en) | A kind of group in information in public security organs industry finds method | |
| Li et al. | Graph summarization for source selection of querying over Linked Open Data | |
| CN113688164A (en) | Interest point query method and system based on knowledge graph correlation analysis | |
| Nguyen | Scalable classification method based on rough sets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |