WO2021227130A1 - Heterogeneous network community detection method, device, computer apparatus, and storage medium - Google Patents

Abstract

The present application discloses a heterogeneous network community detection method, a device, a computer apparatus, and a storage medium. The method comprises: pre-configuring and representing a type constraint s of a heterogeneous network G by means of a 3-tuple <I₁, I₂, k>; acquiring, according to a user community requirement, a type constraint set S meeting the community requirement and a node type set L_S; determining whether each node v in the heterogeneous network G is a type that belongs to the node type set L_S or meets the type constraint set S, obtaining and deleting an invalid node, forming a corresponding message from the invalid node and each neighbor node thereof, and adding the message to a message queue Q; and proceeding to determine whether a new invalid node is generated in the message queue Q, and deleting the new invalid node to obtain a community meeting a query condition. The present application meets a user demand of performing node constraint-based customized community queries in heterogeneous networks.

Description

Heterogeneous network community detection method, device, computer equipment and storage medium

This application is based on a Chinese patent application with an application number of 202010401730.8 and an application date of May 13, 2020, and claims its priority. The entire content of this application is hereby incorporated into this application as a whole.

Technical field

This application relates to the field of heterogeneous networks, and in particular to a method, device, computer equipment, and storage medium for detecting a heterogeneous network community based on node constraints.

Background technique

In graph data mining, community detection is one of the more important steps. The community detection algorithm can be used to search for a specific community structure. At present, the community structure based on k-core is more popular.

However, in a heterogeneous network, due to the existence of different types of nodes, users often need to customize the number of multiple different types of neighbor nodes when performing community detection; and the current community structure based on k-core It can only satisfy community detection in a homogeneous network, and cannot customize the number of neighbor nodes of multiple different types.

The community structure of k-core is defined as follows: Given a positive integer k≥0, the k-core of network G is the largest subgraph H _k that satisfies the following conditions: For any _{node v ∈ H k} , the node degree of node v

Figure 1 is an example k-core community, where the subgraph {A,B,C,D} is a 3-core community, because in this subgraph, each node has at least three neighbor nodes. The graph {A,B,C,D,E,F,G,H,I} is a 1-nucleus community, which consists of two connected subgraphs: {A,B,C,D,E,F, G} and {H,I}, each node in this subgraph has at least one neighbor node. The larger the parameter k, the smaller the detected community, and at the same time, the closer the nodes in the community are.

The k-core community is a community structure defined in a homogeneous network, and cannot be used for community detection in a heterogeneous network, because heterogeneous networks often have different types of nodes. In a heterogeneous graph, the structure of the k-core cannot be used directly to detect the community. At the same time, in a heterogeneous network, due to the existence of different types of nodes, users often need to make different settings for different types of neighbor nodes. The definition of k-core community has only one parameter k, so in this case, and Can not meet the needs of practical applications.

Figure 2 is an example academic network, which contains three types of nodes: A: author, P: paper, and V: conference. Each node in the graph is marked with the type of the node. In addition, side A-P indicates that the author participated in writing the paper, and side P-V indicates that the paper was published in the conference. Suppose a user conducts data mining related analysis for the heterogeneous network, and hopes to find the relevant authors and academic community of papers, satisfying that each author in the community has posted at least 2 papers, and each paper has at least 3 papers Collaborator. The dotted part in Figure 3 is an academic community in the heterogeneous network that meets the user's query needs. However, the current community structure based on k-cores cannot dig out such an academic community: when k=1, the k-core community is the entire heterogeneous network; when k=2, its corresponding k-core The community is represented by the dashed part of Fig. 4, when k=3, the corresponding k-core community is an empty set. Therefore, the current community structure based on k-core cannot satisfy the community search and detection that includes multiple types of nodes in a heterogeneous network.

Therefore, it is difficult to solve the community detection for multiple types of nodes in a heterogeneous network in the prior art, and there is no corresponding method to detect such a community structure.

Application content

The purpose of this application is to provide a heterogeneous network community detection method, device, computer equipment, and storage medium based on node constraints, aiming to solve the problem of community detection for heterogeneous network multi-type nodes in the prior art. A detection algorithm for community structure.

In the first aspect, an embodiment of the present application provides a method for detecting a heterogeneous network community based on node constraints, which includes: presetting a triplet <l ₁ ,l ₂ ,k> to indicate the type constraint of the heterogeneous network G s; where the triplet <l ₁ ,l ₂ ,k> is used to indicate that each _{node of type l 1} has at least k neighboring nodes _{of type l 2 , l 1} ,l ₂ ∈L _G , k≥1, L _G is the type of network node the heterogeneous set G;

Obtain the user's community needs according to the user's query conditions, and obtain the type constraint set S that meets the community needs, where S={s ₁ ,s ₂ ,...,s _t }, and the s ₁ , s ₂ ,..., _st represent different type constraints;

_{Obtain a node type set L S} that meets the needs of the community according to the type constraint set S;

Judging whether the type of each node v in the heterogeneous network G belongs to the node type set L _S , and whether each node v satisfies each type constraint in the type constraint set S;

If the type of node v in the heterogeneous network G does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, then the node v is determined as Invalid node;

All non-valid nodes in the heterogeneous network G are added to the node set H to be deleted and deleted, and the non-valid nodes in the node set H to be deleted and each neighbor node form a corresponding message, and add To the message queue Q;

Iteratively judge whether a new invalid node is generated in the message queue Q according to each type constraint of the type constraint set S, and if so, delete the new invalid node, so as to obtain a community that satisfies the query condition.

In the second aspect, an embodiment of the present application provides a heterogeneous network community detection device based on node constraints, which includes:

The preset unit is used to preset the type constraint s of the heterogeneous network G expressed _{by the triplet <l 1} ,l _{2 ,k>; wherein, the triplet <l 1} ,l ₂ ,k> is used to represent Each node of type l ₁ has at least k neighbor nodes _{of type l 2 , l 1} , l ₂ ∈ L _G , k≥1, where L _G is a set of types of nodes of the heterogeneous network G;

The first obtaining unit is configured to obtain the user's community needs according to the user's query conditions, and obtain the type constraint set S that meets the community needs, where S={s ₁ ,s ₂ ,...,s _t }, the s ₁ , s ₂ ,..., _st represent different type constraints;

The second obtaining unit is configured to obtain, according to the type constraint set S, a node type set L _S that meets the needs of the community;

The first judgment unit is used to judge whether the type of each node v in the heterogeneous network G belongs to the node type set L _S , and whether each node v satisfies each type in the type constraint set S constraint;

The second judgment unit is configured to: if the type of the node v in the heterogeneous network G does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, then Determining the node v as an invalid node;

The deleting unit is used to add all the non-valid nodes in the heterogeneous network G to the node set H to be deleted and delete them, and compose the non-valid nodes in the node set H to be deleted with each neighbor node The corresponding message is added to the message queue Q;

The iterative judgment unit is used to iteratively judge whether a new invalid node has been generated in the message queue Q according to each type constraint of the type constraint set S, and if so, delete the new invalid node so as to satisfy the query condition Community.

In a third aspect, an embodiment of the present application provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and running on the processor, and the processor executes the computer The program implements the heterogeneous network community detection method based on node constraints described in the first aspect above.

In a fourth aspect, an embodiment of the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the above-mentioned first On the one hand, the heterogeneous network community detection method based on node constraints.

This application discloses a method, device, computer equipment and storage medium for detecting a heterogeneous network community based on node constraints, wherein the method includes: presetting a triplet <l ₁ ,l ₂ ,k> to indicate a heterogeneous network The type constraint s of G; where the triplet <l ₁ ,l ₂ ,k> is used to indicate that each _{node of type l 1} has at least k neighboring nodes _{of type l 2 , l 1} ,l ₂ ∈L _G , k≥1, the L _G is the type set of nodes of the heterogeneous network G; obtain the user's community needs according to the user's query conditions, and obtain the type constraint set S that meets the community needs , Where S={s ₁ ,s ₂ ,...,s _t }, the s ₁ ,s ₂ ,...,s _t represent different type constraints; according to the type constraint set S, obtain the The node type set L _S required by the community; determine whether the type of each node v in the heterogeneous network G belongs to the node type set L _S , and whether each node v satisfies the type constraint set S If the type of node v in the heterogeneous network G does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, then The node v is determined to be an invalid node; all the invalid nodes in the heterogeneous network G are added to the node set H to be deleted and deleted, and the invalid nodes in the node set H to be deleted are Each neighbor node composes a corresponding message and adds it to the message queue Q; iteratively judges whether a new invalid node has been generated in the message queue Q according to each type constraint of the type constraint set S, and if so, the new Invalid nodes are deleted, so as to obtain a community that meets the query conditions. According to the user’s community needs, this method obtains the type constraint set S and the node type set L _S that meet the needs of the community, and then according to the type constraint set S and the node type set L _S , all ineffective nodes in the heterogeneous network G are obtained Find out and delete, and finally get the community that meets the query conditions. The embodiments of the present application fulfill the needs of users for personalized community queries based on node constraints in a heterogeneous network.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present application more clearly, the following will briefly introduce the drawings used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. Ordinary technicians can obtain other drawings based on these drawings without creative work.

Figure 1 is a schematic diagram of the k-core community provided by the prior art;

Figure 2 is a schematic diagram of an academic network provided by the prior art;

Figure 3 is a schematic diagram of the actual community that meets user needs in Figure 2;

Figure 4 is a schematic diagram of the 2-core community found based on the prior art;

FIG. 5 is a schematic flowchart of a method for detecting a heterogeneous network community based on node constraints according to an embodiment of the application;

6 is a schematic diagram of a sub-process of a method for detecting a heterogeneous network community based on node constraints according to an embodiment of the application;

FIG. 7 is a schematic diagram of another sub-process of the method for detecting a heterogeneous network community based on node constraints according to an embodiment of this application;

FIG. 8 is a schematic diagram of another sub-process of the method for detecting a heterogeneous network community based on node constraints according to an embodiment of the application;

FIG. 9 is a schematic diagram of a network for classifying and numbering nodes on an academic network according to an embodiment of the application;

10 is a schematic block diagram of a heterogeneous network community detection device based on node constraints provided by an embodiment of the application;

FIG. 11 is a schematic block diagram of a computer device provided by an embodiment of the application.

Detailed ways

The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

It should be understood that when used in this specification and appended claims, the terms "including" and "including" indicate the existence of the described features, wholes, steps, operations, elements and/or components, but do not exclude one or The existence or addition of multiple other features, wholes, steps, operations, elements, components, and/or collections thereof.

It should also be understood that the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in the specification of this application and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

It should be further understood that the term "and/or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations .

Please refer to FIG. 5, which is a schematic flowchart of a method for detecting a heterogeneous network community based on node constraints according to an embodiment of the application;

As shown in Fig. 5, the method includes steps S501 to S507.

S501. Pre-set to express the type constraint s of the heterogeneous network G _{through the triplet <l 1} ,l _{2 ,k>; wherein, the triplet <l 1} ,l ₂ ,k> is used to indicate that each type is The node of l ₁ has at least k neighbor nodes _{of type l 2 , l 1} , l ₂ ∈ L _G , k≥1, and the L _G is the type set of nodes of the heterogeneous network G.

A heterogeneous network (or FIG isomers) can be represented by a four-tuple _{G (V G, E G,} L G, φ>, wherein, V _G representative of a set of nodes; the representative set of edges E _G; L _G represents a set of types, the type used to represent all nodes in the network; [Phi] is a function mapping each node to a network type of L _G:

For a subset of nodes H∈V _G , G[H] represents a subgraph of H. The set of nodes in this subgraph is H, and the set of edges is (H×H)∩E _G. For any node v, define its neighbor node as a node with an edge between it and v, which is represented by N _G (v)={u:(v,u)∈E _G }, and the node degree of node v is d _v = |N _G (v)={u:(v,u)∈E _G }|. In addition, N _G (v,l) represents the node of type l among the neighbor nodes of node v, that is to say, N _G (v,l)={u:(v,u)∈E _G ∧φ (u)=l}.

In a heterogeneous network, there are many different types of nodes, and each type of nodes may be connected or independent. Each node in the heterogeneous network G has at least one neighbor node. In order to more conveniently constrain or limit the relationship between nodes, this step sets the expression of the type constraint s: <l ₁ ,l ₂ ,k>; for example: l ₁ is A, l ₂ It is P, k=2, that is, the type constraint s=<A,P,2>, which means that a node with a node type of A has at least two neighboring nodes with a node type of P.

S502. Obtain the user's community needs according to the user's query conditions, and obtain a type constraint set S that meets the community needs, where S={s ₁ ,s ₂ ,...,s _t }, and the s ₁ ,s ₂ ,...,s _t represent different types of constraints.

In this embodiment, a community requirement that meets the user's query condition is found in a heterogeneous network according to the user's query condition, and the type constraint set S of the community requirement is obtained according to the community requirement.

For example, the academic network shown in Figure 2 contains three types of nodes, namely: A author, P paper, and V meeting. Edge AP means that the author participated in writing the paper, and edge PV means that the paper was published in the meeting. , Suppose the user wants to inquire about the academic community of the author and the paper. The academic community satisfies that each author has posted at least 2 papers, and each paper has at least 3 collaborators. Then the type constraint set S that meets the needs of the community can be obtained, where S={s ₁ ,s ₂ }, where s ₁ =<A,P,2>, s ₂ =<P,A,3>.

_{S503: Obtain a node type set L S} that meets the needs of the community according to the type constraint set S.

In this embodiment, the type constraint set S is defined as the constraints satisfied by the community searched by the user, that is, all the node types in the type constraint set S meet the needs of the community; correspondingly, the type constraint set S is also defined _{A set of node types L S} related to the search community is created, so the set of node types L _S also meets the needs of the community.

Specifically, L _S ={l|<l,l',k>∈S∨<l',l,k>∈S}, for a node v, if

Then the node v must not be in the community that satisfies the constraints.

In an embodiment, as shown in FIG. 6, step S503 includes:

S601. Obtain each type constraint in the type constraint set S;

S602. Obtain the node type in each type of constraint.

S603. Add all the acquired node types to the node type set L _S.

In this embodiment, taking the academic community as an example above, each type constraint in its type constraint set S is obtained, including: s ₁ =<A,P,2>, s ₂ =<P,A,3 〉; Get _{the node types in s 1} : A, P, get _{the node types in s 2} : P, A; add the node type A and the node type P to the node type set L _S , that is, L _S = {A ,P}.

S504: Determine whether the type of each node v in the heterogeneous network G belongs to the node type set L _S , and whether each node v satisfies each type constraint in the type constraint set S.

Taking the above-mentioned academic network as an example, the types of nodes of the academic network include: author A, paper P, and meeting V. The academic network node type set L _S = {A, P}, the type constraint set S = {s ₁ ,s ₂ }, where s ₁ =<A,P,2>, s ₂ =<P,A,3>; compare the type of each node v of the academic network with the set of node types L _S , you can It is determined whether the node type v belongs to the node type set L _S. By comparing each node v in the academic network with each type constraint in the type constraint set S, it can be determined whether the node v satisfies each type constraint in the type constraint set S.

S505: If the type of the node v in the heterogeneous network G does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, then the node v Determined as an invalid node.

In this embodiment, in the heterogeneous network G, as long as the type of a node v does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, the node v Determined as an invalid node.

S506. Add all non-valid nodes in the heterogeneous network G to the node set H to be deleted and delete them, and compose a corresponding message between the non-valid nodes in the node set H to be deleted and each neighbor node. , Added to the message queue Q.

In this embodiment, the non-valid nodes in the heterogeneous network G are obtained according to the node type set L _S and the type constraint set S, and these non-valid nodes are added to the node set H to be deleted and deleted. Since a non-valid node is deleted, it will affect the effectiveness of its neighbor nodes, so the non-valid node in the node set H to be deleted and each neighbor node form a corresponding message and add it to the message queue Q. In the subsequent steps It is necessary to perform iterative judgment on the message queue Q.

S507: Iteratively judge whether a new invalid node is generated in the message queue Q according to each type constraint of the type constraint set S, and if so, delete the new invalid node, so as to obtain a community that satisfies the query condition.

In this embodiment, after deleting an invalid node, the neighbor nodes of the invalid node (nodes that are not listed as invalid nodes for the time being) may become a new invalid node, and the new invalid node's own neighbors The node may also be a potential new ineffective node. Therefore, in this embodiment, iteratively determine and delete until all the ineffective nodes are deleted. The remaining nodes are valid nodes, and the subgraph is composed of the remaining nodes. That is, the community that satisfies S.

In one embodiment, as shown in FIG. 7, step S507 includes:

S701: Traverse the messages in the message queue Q;

S702. Acquire neighbor nodes in the message.

S703: Determine whether the neighbor node has been deleted from the heterogeneous network, if so, consume the corresponding message, if not, proceed to the next step;

S704. Determine whether the neighbor node satisfies each type constraint in the type constraint set S, if so, reserve the neighbor node in the heterogeneous network G, if otherwise, use the neighbor node as a new non-valid node, And delete the new non-valid node from the heterogeneous network G;

S705: Consume the corresponding message, and process the next message until all messages in the message queue Q are consumed.

This embodiment provides an algorithm for iteratively deleting nodes to query the community. The algorithm is based on a message-passing framework. When an invalid node v is deleted, for each neighbor node of the invalid node v u (nodes that have not yet been deleted) generates a message (u, v), which indicates that u's neighbor v has been deleted. This message is used to record potentially invalid nodes, because the deletion of v will affect the validity of u; at the same time; In the process of iteratively deleting nodes in the algorithm, the generated messages are saved by the message queue Q, and the messages in the message queue Q are also consumed as a candidate set for detecting invalid nodes; therefore, the message queue Q is in the iterative process of the algorithm It is constantly changing; when the message queue Q is an empty set at that time, it means that there are no more potential invalid nodes, and the algorithm ends.

In an embodiment, as shown in FIG. 8, step S507 further includes:

S801: When the neighbor node is deleted from the heterogeneous network G, obtain each neighbor node of the neighbor node itself, and compose a new message between the neighbor node and each neighbor node of the neighbor node;

S802: Determine whether the new message is the same as the consumed or unconsumed message in the message queue, if otherwise, add the new message to the message queue, and if so, proceed to the next step.

In this embodiment, for a deleted neighbor node, each of its own neighbor nodes is also a potential ineffective node, the neighbor node and each of its own neighbor nodes form a new message, and then the new message is obtained by the above method. The message is consumed, and the potential ineffective node may have a new neighbor node to continue to form another new message, which is also consumed by the above method; this process is the process of algorithm iteration, and finally all messages are consumed After that, you can get the communities that meet the query conditions.

Taking the above academic network as an example, as shown in Figure 9, for easier description, the academic network is classified and node numbered; in this academic network, its type constraint set, S={s ₁ ,s ₂ }, where s ₁ =<A,P,2>, s ₂ =<P,A,3> type constraint set L _S ={A,P}, it can be obtained that the node type in the searched community only contains the author and Two types of papers.

Specifically, each node in the academic network is traversed to find the set of nodes to be deleted H={3,11,14,2,6}, where the node type of {3,11,14} is a meeting, so it is determined as Invalid nodes; in addition, {2,6} partners are less than 3 and do not meet the constraint S, so they are judged as invalid nodes; then the nodes in H are deleted; finally, the invalid nodes in H are Each neighbor node composes a corresponding message and adds it to the message queue Q. The state of the message queue Q is obtained as:

Q={(4,3),(9,3),(2,3),(6,11),(12,6),(13,14),(1,2),(7,2) ,(7,6),(1,6)}.

According to the message queue Q={(4,3),(9,3),(2,3),(6,11),(12,6),(13,14),(1,2),(7 ,2),(7,6),(1,6)}, detect potential invalid nodes; for message (4,3), because node 4 has 3 authors (collaborators), node 4 is a valid node Therefore, the message did not cause node 4 to be deleted, and at the same time, the message was consumed.

The message queue Q at this time={(9,3),(2,3),(6,11),(12,6),(13,14),(1,2),(7,2), (7,6),(1,6)}, for the message (9,3), because node 9 has 3 partners, node 9 is a valid node, therefore, the message did not cause node 9 to be deleted, and at the same time , The message is consumed.

The message queue Q at this time={(2,3),(6,11),(12,6),(13,14),(1,2),(7,2),(7,6), (1,6)}, for the message (2,3), because node 2 is deleted, the message generates new potential invalid node 1 and potential invalid node 7, and at the same time, the message is consumed. For potentially non-valid nodes, a new message is generated in the manner described above, and whether to add to the message queue is determined, and if it is added to the message queue, it is necessary to continue to judge the new message.

The message queue Q at this time={(6,11),(12,6),(13,14),(1,2),(7,2),(7,6),(1,6)} For the message (6, 11), because node 6 has been deleted, the message does not generate a new message about the potentially invalid node, and at the same time, the message is consumed.

The message queue Q at this time={(12,6),(13,14),(1,2),(7,2),(7,6),(1,6)}, for the message (12, 6) Because node 12 has two papers, node 12 is a valid node. Therefore, the message does not cause node 12 to be deleted, and at the same time, the message is consumed.

At this time, the message queue Q={(13,14),(1,2),(7,2),(7,6),(1,6)}, for the message (13,14), node 13 has There are 3 partners, so node 13 is a valid node. Therefore, the message does not cause node 13 to be deleted, and at the same time, the message is consumed.

At this time, the message queue Q={(1,2),(7,2),(7,6),(1,6)}, for the message (1,2) because nodes 2 and 6 are deleted, the node 1 has no neighbor nodes, therefore, potentially invalid node 1 becomes an invalid node and is deleted. At the same time, because node 1 has no neighbor nodes, deleting node 1 does not generate a new potential invalid node.

At this time, the message queue Q={(7,2),(7,6),(1,6)}, for the message (7,2), because nodes 2 and 6 are deleted, node 7 has no neighbor nodes, Therefore, the potentially invalid node 7 becomes an invalid node and is deleted. At the same time, because the node 7 has no neighbor nodes, the deletion of the node 7 does not generate a new potential invalid node.

At this time, the message queue Q={(7,6),(1,6)}, for the message (7,6), because node 7 has been deleted, there is no need to verify whether node 7 is valid, and the message is consumed .

At this time, the message queue Q={(1,6)}, because node 1 has been deleted, there is no need to verify whether node 1 is valid, and the message is consumed.

After processing according to the above algorithm, the message queue at this time

The remaining nodes in the academic network are valid nodes, and the subgraph formed by these valid nodes is the community that meets the query conditions.

In an embodiment, the step S107 further includes:

Obtain all connected subgraphs that meet the query conditions, and treat each connected subgraph as a community that meets the query conditions.

In a heterogeneous network G, there may be multiple connected subgraphs satisfying the type constraint set S, and each connected subgraph is connected by ineffective nodes. Each connected subgraph of the heterogeneous network G can be used as Communities that meet the query conditions.

In an embodiment, the acquiring all connected subgraphs that satisfy the query condition, and taking each connected subgraph as a community that satisfies the query condition, further includes:

The nodes reserved in the heterogeneous network G are divided to obtain multiple connected subgraphs, and each connected subgraph is regarded as a community that satisfies the query condition.

In this embodiment, ineffective nodes between multiple connected subgraphs in the heterogeneous network G can be deleted to obtain multiple separate connected subgraphs, or a single connected subgraph can be divided to obtain multiple connected subgraphs. Subgraphs, and each divided connected subgraph can also be used as a community that satisfies the query conditions.

In an embodiment, the method further includes:

It is preset that the node v in the heterogeneous network G satisfies the following conditions, then the node v satisfies the type constraint s: φ(v)≠l ₁ , or

Among them, N _G (v, l ₂ ) represents a node of type l ₂ among the neighbor nodes of v.

In this embodiment, the specific meaning of the definition is that if the type of v is l ₁ and it has at least k _{neighbor nodes of type l 2} , then v satisfies the condition s; in addition, if the type of v is not l ₁ , then the type constraint s is not defined in v, which means that s does not restrict the number of nodes of type _{l 2 that v must satisfy.} Therefore, in this case, it is also considered that v satisfies the condition s (because the constraint does not work).

The embodiment of the present application also provides a heterogeneous network community detection device based on node constraint, which is used to perform any of the aforementioned node constraint-based heterogeneous network community detection methods. Examples. Specifically, please refer to FIG. 10, which is a schematic block diagram of a heterogeneous network community detection device based on node constraints provided by an embodiment of the present application.

As shown in FIG. 10, a heterogeneous network community detection device 1000 based on node constraints includes: a preset unit 1001, a first acquisition unit 1002, a second acquisition unit 1003, a first judgment unit 1004, a second judgment unit 1005, The deletion unit 1006 and the iteration judgment unit 1007.

The preset unit 1001 is configured to preset the type constraint s of the heterogeneous network G expressed _{by the triplet <l 1} ,l _{2 ,k>; wherein, the triplet <l 1} ,l ₂ ,k> is used for l represent each type of node ₁ is at least k types of neighbors l _{2, l 1, l 2 ∈L G} , k≥1, L _G is the set of nodes of the network in the heterogeneous type the G ；

The first acquiring unit 1002 is configured to acquire the user's community needs according to the user's query conditions, and acquire a type constraint set S that meets the community needs, where S={s ₁ ,s ₂ ,...,s _t }, the s ₁ , s ₂ ,..., _st represent different type constraints;

The second obtaining unit 1003 is configured to obtain, according to the type constraint set S, a node type set L _S that meets the needs of the community;

The first determining unit 1004 is configured to determine whether the type of each node v in the heterogeneous network G belongs to the node type set L _S , and whether each node v satisfies each of the type constraint set S Type constraint

The second judgment unit 1005 is configured to: if the type of the node v in the heterogeneous network G does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, Then determine the node v as an invalid node;

The deleting unit 1006 is configured to add all the non-valid nodes in the heterogeneous network G to the node set H to be deleted and delete them, and combine the non-valid nodes in the node set H to be deleted and each neighbor node Compose the corresponding message and add it to the message queue Q;

The iterative judgment unit 1007 is configured to iteratively judge whether a new invalid node has been generated in the message queue Q according to each type constraint of the type constraint set S, and if so, delete the new invalid node to obtain a satisfactory query Conditional community.

According to the user’s community needs, the device obtains the type constraint set S and the node type set L _S that meet the needs of the community, and then finds all the non-valid nodes in the heterogeneous network G according to the type constraint set S and the node type set L _S Finally, the community that meets the query conditions is obtained, so as to meet the needs of users for personalized community query based on node constraints in a heterogeneous network.

Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working process of the device and unit described above can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

The foregoing device for detecting a heterogeneous network community based on node constraints can be implemented in the form of a computer program, which can be run on a computer device as shown in FIG. 11.

Please refer to FIG. 11, which is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device 1100 is a server, and the server may be an independent server or a server cluster composed of multiple servers.

Referring to FIG. 11, the computer device 1100 includes a processor 1102, a memory, and a network interface 1105 connected through a system bus 1101, where the memory may include a non-volatile storage medium 1103 and an internal memory 1104.

The non-volatile storage medium 1103 can store an operating system 11031 and a computer program 11032. When the computer program 11032 is executed, the processor 1102 can execute a method for detecting a heterogeneous network community based on node constraints.

The processor 1102 is used to provide computing and control capabilities, and support the operation of the entire computer device 1100.

The internal memory 1104 provides an environment for the operation of the computer program 11032 in the non-volatile storage medium 1103. When the computer program 11032 is executed by the processor 1102, the processor 1102 can execute the method for detecting heterogeneous network communities based on node constraints. .

The network interface 1105 is used for network communication, such as providing data information transmission. Those skilled in the art can understand that the structure shown in FIG. 11 is only a block diagram of part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 1100 to which the solution of the present application is applied. The specific computer device The 1100 may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

Those skilled in the art can understand that the embodiment of the computer device shown in FIG. 11 does not constitute a limitation on the specific configuration of the computer device. In other embodiments, the computer device may include more or less components than those shown in the figure. Or some parts are combined, or different parts are arranged. For example, in some embodiments, the computer device may only include a memory and a processor. In such an embodiment, the structures and functions of the memory and the processor are consistent with the embodiment shown in FIG. 11, and will not be repeated here.

It should be understood that, in this embodiment of the application, the processor 1102 may be a central processing unit (Central Processing Unit, CPU), and the processor 1102 may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSPs), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor.

In another embodiment of the present application, a computer-readable storage medium is provided. The computer-readable storage medium may be a non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, where the computer program is executed by a processor to implement the node constraint-based heterogeneous network community detection method of the embodiment of the present application.

The storage medium is a physical, non-transitory storage medium, such as a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a magnetic disk, or an optical disk, etc., which can store program codes. medium.

Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working process of the above-described equipment, device, and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

The above are only specific implementations of this application, but the protection scope of this application is not limited to this. Any person skilled in the art can easily think of various equivalents within the technical scope disclosed in this application. Modifications or replacements, these modifications or replacements shall be covered within the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims (10)

1. A heterogeneous network community detection method based on node constraints, which is characterized in that it includes:

   It is preset that the type constraint s of the heterogeneous network G is represented by the triplet <l ₁ ,l _{2 ,k>; wherein, the triplet <l 1} ,l ₂ ,k> is used to indicate that each type is l ₁ The node of has at least k neighbor nodes _{of type l 2 , l 1} , l ₂ ∈ L _G , k≥1, where L _G is the type set of nodes of the heterogeneous network G;

   Obtain the user’s community needs according to the user’s query conditions, and obtain the type constraint set S that meets the community needs, where S={s ₁ , s ₂ ,..., s _t }, the s ₁ , s ₂ ,...,s _t represents different types of constraints;

   _{Obtain a node type set L S} that meets the needs of the community according to the type constraint set S;

   Judging whether the type of each node v in the heterogeneous network G belongs to the node type set L _S , and whether each node v satisfies each type constraint in the type constraint set S;

   If the type of node v in the heterogeneous network G does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, then the node v is determined as Invalid node;

   All non-valid nodes in the heterogeneous network G are added to the node set H to be deleted and deleted, and the non-valid nodes in the node set H to be deleted and each neighbor node form a corresponding message, add To the message queue Q;

   Iteratively judge whether a new invalid node is generated in the message queue Q according to each type constraint of the type constraint set S, and if so, delete the new invalid node, so as to obtain a community that meets the query condition.
2. The method for detecting a heterogeneous network community based on node constraints according to claim 1, wherein the iteratively judging whether a new message queue Q is generated according to each type constraint of the type constraint set S Invalid node, if it is, delete the new invalid node to obtain the community that meets the query conditions, including:

   Traverse the messages in the message queue Q;

   Acquiring neighbor nodes in the message;

   Determine whether the neighbor node has been deleted from the heterogeneous network, if so, consume the corresponding message, if not, go to the next step;

   Determine whether the neighbor node satisfies each type constraint in the type constraint set S, if so, keep the neighbor node in the heterogeneous network G, if otherwise, treat the neighbor node as a new non-valid node and follow Deleting the new non-valid node from the heterogeneous network G;

   The corresponding message is consumed, and the next message is processed until all messages in the message queue Q are consumed.
3. The method for detecting a heterogeneous network community based on node constraints according to claim 2, wherein the iteratively judging whether a new message queue Q is generated according to each type constraint of the type constraint set S Invalid node, if it is, delete the new invalid node to obtain a community that meets the query conditions. It also includes:

   When deleting the neighbor node from the heterogeneous network G, obtain each neighbor node of the neighbor node itself, and compose a new message between the neighbor node and each neighbor node of the neighbor node;

   It is judged whether the new message is the same as the consumed or unconsumed message in the message queue, if not, the new message is added to the message queue, and if so, the next step is performed.
4. The method for detecting a heterogeneous network community based on node constraints according to claim 1, wherein the obtaining a node type set L _S that meets the needs of the community according to the type constraint set S comprises:

   Obtain each type constraint in the type constraint set S;

   Get the node type in each type of constraint;

   Add all the acquired node types to the node type set L _S.
5. The heterogeneous network community detection method based on node constraints according to claim 1, wherein said iteratively judging whether a new message queue Q is generated according to each type constraint of the type constraint set S If it is a non-valid node, delete the new non-valid node to obtain a community that satisfies the query conditions. It also includes:

   Obtain all connected subgraphs that meet the query conditions, and treat each connected subgraph as a community that meets the query conditions.
6. The method for detecting a heterogeneous network community based on node constraints according to claim 5, wherein the acquiring all connected subgraphs that meet the query conditions, and using each connected subgraph as a community that meets the query conditions, includes :

   The nodes reserved in the heterogeneous network G are divided to obtain multiple connected subgraphs, and each connected subgraph is regarded as a community that satisfies the query condition.
7. The method for detecting a heterogeneous network community based on node constraints according to claim 1, further comprising:

   It is preset that the node v in the heterogeneous network G satisfies the following conditions, and the node v satisfies the type constraint s:

   φ(v)≠l ₁ , or

   

   Among them, N _G (v, l ₂ ) represents a node of type l ₂ among neighboring nodes of node v.
8. A heterogeneous network community detection device based on node constraints, which is characterized in that it comprises:

   The preset unit is used to preset the type constraint s of the heterogeneous network G expressed _{by the triplet <l 1} ,l _{2 ,k>; wherein, the triplet <l 1} ,l ₂ ,k> is used to represent Each node of type l ₁ has at least k neighbor nodes _{of type l 2 , l 1} , l ₂ ∈ L _G , k≥1, where L _G is a set of types of nodes of the heterogeneous network G;

   The first obtaining unit is used to obtain the user's community needs according to the user's query conditions, and obtain the type constraint set S that meets the community needs, where S={s ₁ ,s ₂ ,...,s _t }, The s ₁ , s ₂ ,..., _st represent different types of constraints;

   The second obtaining unit is configured to obtain, according to the type constraint set S, a node type set L _S that meets the needs of the community;

   The first judgment unit is used to judge whether the type of each node v in the heterogeneous network G belongs to the node type set L _S , and whether each node v satisfies each type in the type constraint set S constraint;

   The second judgment unit is configured to: if the type of the node v in the heterogeneous network G does not belong to the node type set L _S , or the node v does not satisfy any type constraint in the type constraint set S, then Determining the node v as an invalid node;

   The deleting unit is used to add all the non-valid nodes in the heterogeneous network G to the node set H to be deleted and delete them, and compose the non-valid nodes in the node set H to be deleted with each neighbor node The corresponding message is added to the message queue Q;

   The iterative judgment unit is used to iteratively judge whether a new invalid node has been generated in the message queue Q according to each type constraint of the type constraint set S, and if so, delete the new invalid node so as to satisfy the query condition Community.
9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program as claimed in claims 1 to The heterogeneous network community detection method based on node constraint described in any one of 7.
10. A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes any one of claims 1 to 7 The above-mentioned heterogeneous network community detection method based on node constraints.

Images