CN115733763A

CN115733763A - Label propagation method and device for associated network and computer readable storage medium

Info

Publication number: CN115733763A
Application number: CN202211492068.7A
Authority: CN
Inventors: 刘红宝; 何朔; 高鹏飞; 郑建宾; 汤韬; 邱震尧
Original assignee: China Unionpay Co Ltd
Current assignee: China Unionpay Co Ltd
Priority date: 2022-11-25
Filing date: 2022-11-25
Publication date: 2023-03-03

Abstract

The invention provides a label propagation method, a device and a computer readable storage medium of an associated network, wherein the method comprises the following steps: constructing a first associated network based on the first party data, and constructing a second associated network based on the second party data; associating the first associated network and the second associated network based on a safety request protocol to obtain a federal associated network; iteratively performing multiple rounds of label propagation on nodes of the federated association network; wherein each round of tag propagation comprises: determining label propagation probability between adjacent nodes in the federal association graph; and aiming at each node, determining the label of each node in the current round according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node. By using the method, the label propagation of the cross-platform network can be realized on the premise of ensuring the privacy data.

Description

Label propagation method and device for associated network and computer readable storage medium

Technical Field

The invention belongs to the field of computers, and particularly relates to a label propagation method and device for an associated network and a computer readable storage medium.

Background

This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.

With the stricter privacy protection laws, data privacy protection issues need to be considered more and more in data cooperation among organizations. At present, the privacy calculation technology mainly focuses on the scenes of federal learning, safe intersection, hidden trace query and the like, and all the scenes aim at the union and the summation of single-point data. The data sources of the tag propagation algorithm in the current associated network are all local. Cross-organization data joint application on the premise of privacy protection cannot be realized. The updating of the label cannot utilize the associated network data of the two parties at the same time, and the data value is not efficiently utilized.

Therefore, how to implement the federal network label propagation under the premise of privacy protection is a problem to be solved urgently.

Disclosure of Invention

In view of the above problems in the prior art, a method, an apparatus, and a computer-readable storage medium for propagating labels of an associated network are provided.

The present invention provides the following.

In a first aspect, a label propagation method for an association network is provided, including: constructing a first associated network based on the first party data, and constructing a second associated network based on the second party data; associating the first associated network and the second associated network based on a safety request protocol to obtain a federal associated network; iteratively performing multiple rounds of label propagation on nodes of the federated association network; wherein each round of tag propagation comprises: determining label propagation probability between adjacent nodes in the federal association graph; and aiming at each node, determining the label of each node in the current round according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node.

In one embodiment, associating the first association network and the second association network based on a security agreement to obtain a federated association network, further includes: encrypting and intersecting the first party data and the second party data, determining a common node in a first associated network and a second associated network, and associating the first associated network and the second associated network according to the common node to obtain a federal associated network;

in one embodiment, determining label propagation probabilities between adjacent nodes in a federated association graph further comprises: determining edge weight w of edge ij between node i and neighbor node j in federal correlation network _ij (ii) a Determining edge weights and sigma between node i and all its neighbor nodes J _j w _ij (ii) a According to the edge weight w _ij Sum side weight sum Σ _j w _ij Determining label propagation probability P of the neighbor node j to the node i _ij 。

In one embodiment, if node i is a non-common node, all neighbor nodes J represent all neighbor nodes of the graph in which node i is located.

In one embodiment, if node i is a common node, all neighbor nodes J represent a set of all neighbor nodes a of node i in the first association network and all neighbor nodes b of node i in the second association network.

In one embodiment, if node i is a common node, the first and second parties interact with the sum of edge weights between node i and all neighboring nodes a of the first association network and the sum of edge weights between node i and all neighboring nodes b of the second association network.

In one embodiment, the method further comprises: if the node i is a common node, determining the edge weight and sigma by using the following formula _j w _ij ：∑ _j w _ij ＝∑ _a w _ia +∑ _b w _ib (ii) a Therein, sigma _a w _ia Is the sum of edge weights, sigma, between the node i and all the neighbor nodes a of the first correlation network _b w _ib Is the edge weight sum between node i and all neighbor nodes b of the second associative network.

In one embodiment, iteratively performing multiple rounds of tag propagation for nodes of a federated association network further comprises: determining marked nodes and unmarked nodes of the federated association network; updating the labels of the nodes which are not marked in turn until the labels of the nodes which are not marked are not changed and/or exceed the updating turn threshold; and keeping the label of the label node unchanged.

In one embodiment, determining the label of each node according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node comprises: for each node, determining a label of each neighbor node of the node in the current round and the label propagation probability of each neighbor node to the node; calculating the sum of the label propagation probabilities corresponding to each label in all the neighbor nodes of the node to obtain the label propagation aggregation probability corresponding to each label; and updating the labels of the current round of the nodes according to the label with the maximum label propagation aggregation probability.

In one embodiment, if the node is a non-common node, the method further comprises: and calculating the label propagation aggregation probability corresponding to each type of neighbor node label of the node by the node.

In one embodiment, if the node is a common node, the method further comprises: a first party calculates the propagation aggregation probability of first party labels corresponding to all neighbor node labels of a node in a first correlation network; the second party calculates the second label propagation aggregation probability corresponding to all the neighbor node labels of the node in the second correlation network; the first party and the second party interact a first label propagation aggregation probability and a second label propagation aggregation probability; and the first party and the second party respectively carry out label propagation probability aggregation again based on the interactive information to obtain the label propagation aggregation probability corresponding to each label.

In one embodiment, the method further comprises: determining graph weights of the first correlation network and the second correlation network according to the node relation closeness degree of the first correlation network and the second correlation network; and introducing graph weight in the interaction process of the first association network and the second association network.

In one embodiment, introducing graph weights during interaction of a first associated network and a second associated network comprises: if the node i is a common node, the node i is determined by the following formulaSide weight sum Σ _j w _ij ：

∑ _j w _ij ＝θ _a ∑ _a w _ia +θ _b ∑ _b w _ib (ii) a Therein, sigma _a w _ia Is the sum of edge weights, sigma, between the node i and all the neighbor nodes a of the first correlation network _b w _ib Is the sum of the edge weights, θ, between node i and all the neighboring nodes b of the second associative network _a Is the graph weight, θ, of the first associative network _b Is the graph weight of the second associated network.

In one embodiment, introducing graph weight in the interaction process of the first association network and the second association network further comprises: and after the first party and the second party interact the first label propagation aggregation probability and the second label propagation aggregation probability, carrying out label propagation probability aggregation again based on the graph weights of the first association network and the second association network to obtain the label propagation aggregation probability corresponding to each label.

In one embodiment, the method further comprises: and if the first correlation network and the second correlation network are directed graph networks, only the inflow neighbor node of each node is used as a neighbor node.

In a second aspect, a label distribution apparatus associated with a network is provided, including: the graph building module is used for building a first associated network based on the first party data and building a second associated network based on the second party data; the federal network module is used for associating the first associated network and the second associated network based on a safety traffic protocol to obtain a federal associated network; the label propagation module is used for iteratively executing multiple rounds of label propagation on the nodes of the federated association network; wherein each round of tag propagation comprises: determining label propagation probability between adjacent nodes in the federal association graph; and aiming at each node, determining the label of each node in the current round according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node.

In a third aspect, a label propagation apparatus associated with a network is provided, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform: the method of the first aspect.

In a fourth aspect, there is provided a computer readable storage medium storing a program which, when executed by a multicore processor, causes the multicore processor to perform the method of the first aspect.

One of the advantages of the above embodiment is that label propagation across a platform network can be realized on the premise of ensuring private data. .

Other advantages of the present invention will be explained in more detail in conjunction with the following description and the accompanying drawings.

It should be understood that the above description is only an overview of the technical solutions of the present invention, so that the technical means of the present invention can be more clearly understood and implemented according to the content of the specification. In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.

Drawings

The advantages and benefits described herein, as well as other advantages and benefits, will be apparent to those of ordinary skill in the art upon reading the following detailed description of the exemplary embodiments. The drawings are only for purposes of illustrating exemplary embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

fig. 1 is a schematic structural diagram of a label propagation device associated with a network according to an embodiment of the present invention;

fig. 2 is a flowchart illustrating a label propagation method for an association network according to an embodiment of the invention;

FIG. 3 is a schematic diagram of a first association network and a second association network in accordance with an embodiment of the present invention;

FIG. 4 is a schematic diagram of a federated association network in accordance with an embodiment of the present invention;

FIG. 5 is a schematic diagram of determining a probability of tag propagation for a first association network and a second association network, according to an embodiment of the invention;

FIG. 6 is a schematic diagram of determining a tag propagation probability for a federated association network in accordance with an embodiment of the present invention;

FIG. 7 is a schematic diagram of tag propagation associated with a network according to one embodiment of the present invention;

FIG. 8 is a schematic diagram of tag propagation for an association network according to one embodiment of the invention;

fig. 9 is a schematic structural diagram of a tag distribution apparatus associated with a network according to an embodiment of the present invention.

In the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

In the description of the embodiments of the present application, it is to be understood that terms such as "including" or "having" are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof disclosed in the specification, and are not intended to preclude the presence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

A "/" indicates an OR meaning, for example, A/B may indicate A or B; "and/or" herein is merely an association relationship describing an associated object, and means that there may be three relationships, for example, a and/or B, and may mean: a exists alone, A and B exist simultaneously, and B exists alone.

The terms "first", "second", and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or to implicitly indicate the number of technical features indicated. Thus, a feature defined as "first," "second," etc. may explicitly or implicitly include one or more of that feature. In the description of the embodiments of the present application, "a plurality" means two or more unless otherwise specified.

For clarity of explanation of the embodiments of the present application, some concepts that may appear in subsequent embodiments will first be described.

The present invention will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Referring initially to FIG. 1, a schematic diagram of an environment 100 is schematically illustrated in which exemplary implementations according to the present disclosure may be used.

Fig. 1 shows a schematic diagram of an example of a computing device 100, according to an embodiment of the present disclosure. It should be noted that fig. 1 is a schematic structural diagram of a hardware operating environment of a tag propagation method for an associated network. The label spreading method based on the associated network of the embodiment of the invention can be a PC, a portable computer and other terminal equipment.

As shown in fig. 1, the label propagation method device of the associated network may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. The communication bus 1002 is used to implement connection communication among these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory such as a disk memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.

Those skilled in the art will appreciate that the label dissemination device architecture of the association network shown in fig. 1 does not constitute a limitation to the label dissemination method device of the association network and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.

As shown in fig. 1, a tag propagation method program of an operating system, a network communication module, a user interface module, and an associated network may be included in a memory 1005, which is a kind of computer storage medium. The operating system is a program for managing and controlling hardware and software resources of the tag distribution device associated with the network, and supports the operation of the tag distribution program associated with the network and other software or programs.

In the label propagation device of the associated network shown in fig. 1, the user interface 1003 is mainly used for receiving requests, data and the like sent by the first terminal, the second terminal and the supervision terminal; the network interface 1004 is mainly used for connecting the background server and performing data communication with the background server; and the processor 1001 may be configured to invoke a tag propagation program associated with the network stored in the memory 1005 and perform the following operations:

constructing a first associated network based on the first party data and constructing a second associated network based on the second party data; associating the first associated network and the second associated network based on a safety request protocol to obtain a federal associated network; iteratively performing multiple rounds of label propagation on nodes of the federated association network; wherein, each round of label propagation comprises: determining label propagation probability between adjacent nodes in the federal association graph; and aiming at each node, determining the label of each node in the current round according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node.

Therefore, both parties only need to interact non-private data such as tag propagation probability and the like, and cross-organization data association network joint calculation and application schemes can be carried out on the premise that the original data of both parties do not go out of the warehouse.

Fig. 2 shows a flow chart for performing a label propagation method for an associated network according to an embodiment of the present disclosure. The method may be performed, for example, by a computing device 100 as shown in FIG. 1. It should be understood that method 200 may also include additional blocks not shown and/or may omit blocks shown, as the scope of the present disclosure is not limited in this respect.

Step 210, constructing a first associated network based on the first party data, and constructing a second associated network based on the second party data;

for example, referring to fig. 3, parties a and B each form nodes and edges in the associated network based on their own data. Assuming that the A party is a bank, forming an A party associated network as a transfer associated network through transfer data among users, wherein the mobile phone number of the user is a node in the associated network, connecting edges of the nodes with the transfer relation, and taking the transfer amount as a side weight value among the nodes. The B side is an operator, a B side correlation network is formed through call data among users and is a call correlation network, the mobile phone number of the user is a node in the correlation network, the node with the call record is connected, and the call times are the weight value of the edge between the nodes. Optionally, the edge weight values of the respective associated networks may be normalized.

Step 220, associating the first associated network with the second associated network based on a security request protocol to obtain a federal associated network;

in one embodiment, the step 220 further comprises: and encrypting and intersecting the first party data and the second party data, determining a common node in the first associated network and the second associated network, and associating the first associated network and the second associated network according to the common node to obtain the federal associated network.

For example, referring to fig. 4, a secure intersection algorithm (such as a privacy intersection algorithm based on RSA + HASH) may be used to perform secure intersection on data of two parties, and a common node is found without exposing original data, so as to form a virtual federal association network. As shown in fig. 4, the first association network and the second association network in fig. 3 may be associated to obtain the federated association network. Where Va represents the node on the a side, va represents the node on the B side, vab represents the node common to both sides, and Vab1, vab2, and Vab3 are the nodes common to both sides. Taking the example of node Vab1, there are 2 neighbor nodes of Vab1 from the perspective of the a-party association network alone. And from the perspective of global data, there are 4 neighbor nodes of Vab 1.

Step 230, performing multiple rounds of label propagation on nodes of the federated association network in an iterative manner;

specifically, the nodes of the federated association network may include marked nodes and unmarked nodes, for example, a part of unmarked nodes may exist in the first association network and the second association network, respectively. For another example, all the first associated networks are labeled nodes, and all the second associated networks are unlabeled nodes. And so on.

In an embodiment, for a case that the federal association network includes both marked nodes and unmarked nodes, the step 230 may further include the steps of: firstly, determining a marked node and an unmarked node of a federated association network; updating the labels of the nodes which are not marked in turn until the labels of the nodes which are not marked are not changed and/or exceed the updating turn threshold; and keeping the label of the label node unchanged. Therefore, the label of the original sample can be guaranteed to be unchanged, and the label propagation accuracy is guaranteed.

Optionally, the label of the labeled node may also be dynamically updated round by round, that is, the labels of all nodes in the federated association network are updated round by round until the labels of the nodes no longer change and/or exceed the update round threshold. Therefore, the original label can be corrected, and the hidden risk label can be mined.

In the above step 230, each round of tag propagation specifically includes the following steps 231-232:

step 231, determining label propagation probability between adjacent nodes in the federal association graph;

in an embodiment, the step 231 may specifically include:

(1) Determining edge weight w of edge ij between node i and neighbor node j in federal correlation network _ij ；

(2) Determining edge weights and sigma between node i and all its neighbor nodes J _j w _ij ；

Further, if node i is a common node, the first party and the second partyEdge weight sum sigma between two-party interaction node i and all neighbor nodes a of first correlation network _a w _ia And the edge weights and sigma between node i and all neighboring nodes b of the second correlation network _b w _ib Thus, the first party and the second party can respectively calculate the edge weight sum sigma between the node i and all the neighbor nodes J on the basis of the interactive two-party edge weight sum of the two parties _j w _ij 。

Further, in one embodiment, if node i is a common node, based on the mutual side weight sum of both sides, both the first side and the second side can determine the side weight sum Σ using the following formula _j w _ij ：

∑ _j w _ij ＝∑ _a w _ia +∑ _b w _ib ；

Therein, sigma _a w _ia Is the sum of edge weights, sigma, between the node i and all the neighbor nodes a of the first correlation network _b w _ib Is the edge weight sum between node i and all neighbor nodes b of the second associative network.

Optionally, in another embodiment, the influence of the service scenario on the closeness of the node relationship may be further considered. For example, in a financial scenario, the transfer relationship is a strong relationship, and the call relationship is a weak relationship, so when calculating the label propagation probability of an edge, the strength of the edge relationship between two parties in different business scenarios can be considered for performing weighted aggregation.

In this case, both the first party and the second party may determine the weighted edge weights and Σ using the following formulas _j w _ij ：

∑ _j w _ij ＝θ _a ∑ _a w _ia +θ _b ∑ _b w _ib ；

Wherein, theta _a Is the corresponding graph weight, θ, of the first correlation network _b Is the graph weight of the second associated network.

(3) According to the edge weight w _ij And the edge weight sum ∑ _j w _ij Determining the label propagation probability P of the neighbor node j to the node i _ij 。

In particular, label propagation probabilities on edges of a federated associative network

Wherein w _ij Representing the weight value of the edge ij. Here, for a non-common node, J represents a neighbor node of node i; for a common node, J represents all neighbor nodes of the node i on both sides a and B, respectively.

∑ _j w _ij The calculation logic of (1) is that the weight sum of the neighbor nodes of the local node i is sigma _a w _ia (ii) a B side calculates the weight sum of neighbor nodes of local node i as sigma _b w _ib . Mutual sigma _a w _ia And sigma _b w _ib Obtaining the final weight calculation denominator value as sigma _j w _ij ＝∑ _a w _ia +∑ _b w _ib 。

Referring to fig. 5, node Vab2 is taken as an example. In the local network of the a-party, there are 1 neighbor nodes of Vab2, and the label propagation probability of the self is calculated separately as P =0.1/0.1=1; in the local network on the side B, there are 3 neighbor nodes of Vab2, and the propagation probability of the label of the separately calculated own is P = 0.2/(0.2 +0.4+ 0.8) =1/7, P = 0.4/(0.2 +0.4+ 0.8) =2/7, P = 0.8/(0.2 +0.4+ 0.8) =4/7, respectively; further, both parties exchange the weight values of the target node Vab2 and the neighbor nodes, where the party a is 0.1, and the party b is 0.2+0.4+0.8=1.4. And updating the label propagation probability of the target node by combining the federal correlation network.

With reference to FIG. 6, through the above calculation, at party A, the label propagation probability of its neighbor node for this node i

On the B side, similarly, the label propagation probability of the neighbor nodes to the node i is 2/15,4/15,8/15 respectively.

Step 232, aiming at each node, determining the label of each node according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node.

With reference to fig. 7, the federal association network formed by Vab2 is taken as an example here, and is shown as follows. Node 5 is a risk node, shown as a white node, with the label set to "1"; the remaining nodes are unknown nodes, shown as gray nodes, with the label set to "0". In the label propagation process, the label of the node 5 is always '1', and the labels of the other nodes are updated round by round until the labels of all the nodes are not changed any more or exceed the update round threshold value.

In one embodiment, the step 232 further includes the following steps:

step 2321, for the node i, determining the label of each neighbor node of the node i in the current round and the label propagation probability of each neighbor node to the node i;

step 2322, in all the neighbor nodes of the node i, calculating the sum of the label propagation probabilities corresponding to each label to obtain the label propagation aggregation probability corresponding to each label;

specifically, if the node i is a non-shared node, the label propagation aggregation probability corresponding to each type of neighbor node label of the node i is calculated only by the party of the node i.

Specifically, if the node i is a common node, the following steps are executed: firstly, a first party calculates the propagation aggregation probability of first party labels corresponding to all neighbor node labels of a node i in a first correlation network; the second party calculates the second label propagation aggregation probability corresponding to all the neighbor node labels of the node i in the second correlation network; secondly, the first party and the second party interact a first label propagation aggregation probability and a second label propagation aggregation probability; and finally, the first party and the second party respectively carry out label propagation probability aggregation again based on the interactive information to obtain the label propagation aggregation probability corresponding to each label.

Step 2323, the labels of the current round of the node i are updated according to the label with the maximum label propagation aggregation probability.

The node tag update rule shown in the foregoing step 2321-step 2323 may include the following specific steps:

first, for the T-th round update of node i, let its neighbor node set be J ((J) ₁ ，L ₁ ，P _i1 )，(J ₂ ，L ₂ ，P _i2 )，(J _j ，L _j ，P _ij )...，(J _n ，L _n ，P _in )>Wherein J _j Is an identification of a neighbor node j, L _j Is a label of a neighbor node j, P _ij Is an edge<i，J _j >The propagation probability of (c).

Secondly, the aggregate propagation probability of all the labels of the neighbor node set is calculated. Specifically P (L) _j )＝∑P _ij Wherein, P _ij Is a label of L _j For the target node i. If the node i is a non-shared node, only the propagation probability of the local neighbor node label needs to be calculated. If the node i is a common node, the A party calculates the probability P (L) corresponding to all the neighbor node labels of the local party _aj )＝∑P _iaj And the B party calculates the probability P (L) corresponding to the local neighbor node label _bj )＝∑P _ibj Party A and party B interact P (L) _aj ) And P (L) _bj ) And respectively carrying out label propagation probability aggregation again in the local to obtain a label propagation aggregation probability P (L) finally combining the associated network information of the two parties _j )＝P(L _aj )+P(L _bj )。

Finally, the largest P (L) is selected _j ) The corresponding label L _j Is the label of the current round of node i. And repeating the steps until the labels of all the nodes are not changed any more.

In one specific example, a specific calculation example of tag update is given with reference to fig. 7 and 8.

Referring to fig. 7, for the first round of propagation, for node Vab2, the following calculation is performed:

(1) And calculating the label propagation aggregation probability of the A-side neighbor node as < "0",1/15>, wherein "0" represents a risk-free label, and 1/15 represents the label propagation aggregation probability corresponding to the label "0". It can be understood that, since the a-side Vab2 has only one neighbor node 1 and its initial label value is "0", the label propagation probability of node 1 to node Vab2 has been calculated as 1/15 in the above, so that for the a-side Vab2 node, there is only one propagable label "0", and the label propagation aggregation probability corresponding to the propagable label "0" is 1/15.

(2) And calculating label propagation aggregation probability of the B-side neighbor node to be < "0",6/15>, < "1",8/15>, wherein "0" represents a risk-free label, and 6/15 represents label propagation aggregation probability corresponding to the label "0". "1" represents a risk label and 8/15 represents the label propagation aggregation probability corresponding to label "1". Since the B-party Vab2 has three neighbor nodes (3, 4, 5), and the initial label value of the

nodes

3, 4 is "0", and the initial label value of the node 5 is "1". It has been calculated above that the label propagation probability of node 3 to node Vab2 is 2/15, the label propagation probability of node 4 to node Vab2 is 4/15, and the label propagation probability of node 5 to node Vab2 is 8/15, so that for the B-side Vab2 node, there are 2 propagable labels "0" and "1", and the label propagation aggregation probability corresponding to the propagable label "0" is 6/15=2/15+4/15, and the label propagation aggregation probability corresponding to the propagable label "1" is 8/15.

(3) The two parties exchange the label propagation aggregation probabilities and accumulate the label propagation aggregation probabilities corresponding to the same label, so that the label propagation aggregation probabilities of the nodes Vab2 are < "0",7/15>, < "1", and 8/15> through respective calculation, that is, the label propagation aggregation probability corresponding to the propagatable label "0" is 7/15, and the label propagation aggregation probability corresponding to the propagatable label "1" is 8/15.

(4) And selecting the label "1" corresponding to the maximum label propagation aggregation probability < "1" and 8/15> as the label of the node Vab2 in the current round. Other nodes are similar to the above steps.

After the first round of propagation, the updated node label distribution map of the federal correlation network is shown in fig. 8, in which nodes Vab1 and Vab2 are both updated to label "1". And continuing to propagate the label in the next round until the label of the node is not changed any more or the propagation round is more than a certain threshold value.

In one embodiment, the graph weight of the first association network and the second association network is determined according to the node relation closeness degree of the first association network and the second association network; and introducing the graph weight in the interaction process of the first association network and the second association network.

For example, it may be determined that the first association network and the second association network are strong association or weak association according to a service scenario, and then graph weights of the first association network and the second association network may be introduced when calculating the label propagation probability of an edge. Of course, the graph weights of the first association network and the second association network may also be introduced when calculating the label propagation aggregation probability of each label, which is not specifically limited in the present application.

In one embodiment, the graph weight is introduced in the interaction process of the first association network and the second association network, and the introduction method at least comprises the following two introduction methods:

(1) In the above step 231, if the node i is a common node, the edge weight sum Σ is determined by the following formula _j w _ij ：

(2) In step 232, after the first party and the second party interact with the first label propagation aggregation probability and the second label propagation aggregation probability, label propagation probability aggregation is performed again based on the graph weights of the first association network and the second association network, so as to obtain the label propagation aggregation probability corresponding to each label. For example, if node i is a common node, party a calculates the probability P (L) corresponding to all the labels of its own neighbor nodes _aj )＝∑P _iaj And the B party calculates the probability P (L) corresponding to the local neighbor node label _bj )＝∑P _ibj Party A and party B interact P (L) _aj ) And P (L) _bj ) And respectively carrying out label propagation probability aggregation again in the local to obtain the information finally combined with the associated network information of the two partiesTag propagation aggregation probability P (L) _j )＝θ _a P(L _aj )+θ _b P(L _bj )。

In one embodiment, the method further comprises: and if the first correlation network and the second correlation network are directed graph networks, only the inflow neighbor node of each node is used as a neighbor node. For example, for a directed graph, only the inflow neighbor nodes of the target node may be considered when calculating the propagation probability of the node. The judgment can be specifically carried out by combining the service scene.

In the description of the present specification, reference to a description of the term "some possible embodiments," "some embodiments," "examples," "specific examples," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the various embodiments or examples and features of the various embodiments or examples described in this specification can be combined and combined by those skilled in the art without contradiction.

With regard to the method flow diagrams of embodiments of the present application, certain operations are described as different steps performed in a certain order. Such flow diagrams are illustrative and not restrictive. Certain steps described herein may be grouped together and performed in a single operation, may be divided into multiple sub-steps, and may be performed in an order different than that shown herein. The various steps shown in the flowcharts may be implemented in any way by any circuit structure and/or tangible mechanism (e.g., by software running on a computer device, hardware (e.g., logical functions implemented by a processor or chip), etc., and/or any combination thereof).

Based on the same technical concept, the embodiment of the present invention further provides a label propagation apparatus for an associated network, configured to execute the label propagation method for an associated network provided in any of the above embodiments. Fig. 9 is a schematic structural diagram of a tag propagation apparatus of an association network according to an embodiment of the present invention.

As shown in fig. 9, the apparatus 900 includes:

a graph construction module 910, configured to construct a first associated network based on the first party data, and construct a second associated network based on the second party data;

a federation network module 920, configured to associate the first associated network and the second associated network based on a security negotiation protocol to obtain a federation associated network;

a label propagation module 930 configured to iteratively perform multiple rounds of label propagation on nodes of the federated association network; wherein each round of the tag propagation comprises: determining label propagation probability between adjacent nodes in the federal correlation graph; and aiming at each node, determining the label of each node according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node.

It should be noted that the apparatus in the embodiment of the present application can implement each process of the foregoing embodiment of the method, and achieve the same effect and function, which is not described herein again.

According to some embodiments of the present application, there is provided a non-transitory computer storage medium of a tag propagation method of an associated network, having stored thereon computer-executable instructions configured to, when executed by a processor, perform: the method of the above embodiment.

The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for apparatus, device, and computer-readable storage medium embodiments, the description of which is simplified since it is substantially similar to the method embodiments, and where relevant, reference may be made to some descriptions of the method embodiments.

The apparatus, the device, and the computer-readable storage medium provided in the embodiments of the present application correspond to the method one to one, and therefore, the apparatus, the device, and the computer-readable storage medium also have similar beneficial technical effects to the corresponding method.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, apparatus (device or system), or computer-readable storage medium. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the invention may take the form of a computer-readable storage medium implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices or systems), and computer-readable storage media according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Further, while operations of the methods of the invention are depicted in the drawings in a particular order, this does not require or imply that these operations must be performed in this particular order, or that all of the illustrated operations must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions.

While the spirit and principles of the invention have been described with reference to several particular embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, nor is the division of aspects, which is for convenience only as the features in such aspects cannot be combined to advantage. The invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A label propagation method for an associated network, comprising:

constructing a first associated network based on the first party data, and constructing a second associated network based on the second party data;

associating the first associated network and the second associated network based on a security request protocol to obtain a federal associated network;

iteratively performing multiple rounds of tag propagation on nodes of the federated association network;

wherein each round of the tag propagation comprises: determining label propagation probability between adjacent nodes in the federal correlation graph; and aiming at each node, determining the label of each node according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node.

2. The method of claim 1, wherein associating the first association network and the second association network based on a security claim agreement to obtain a federated association network, further comprises:

encrypting and intersecting the first party data and the second party data, determining a public node in the first association network and the second association network, and associating the first association network and the second association network according to the public node to obtain a federal association network;

3. the method of claim 1, wherein determining label propagation probabilities between adjacent nodes in the federal association graph further comprises:

determining an edge weight w of an edge ij between a node i and a neighbor node j in the federal correlation network _ij ；

Determining edge weights and Σ between the node i and all its neighbor nodes J _j w _ij ；

According to the edge weight w _ij And the side weight sum Σ _j w _ij Determining the label propagation probability P of the neighbor node j to the node i _ij 。

4. The method of claim 3,

and if the node i is a non-common node, all the neighbor nodes J represent all the neighbor nodes of the graph where the node i is located.

5. The method of claim 3,

if the node i is a common node, the all neighbor nodes J represent a set of all neighbor nodes a of the node i in the first associated network and all neighbor nodes b of the node i in the second associated network.

6. The method of claim 3,

if the node i is a common node, the first party and the second party interact the sum of the edge weights between the node i and all the neighbor nodes a of the first association network and the sum of the edge weights between the node i and all the neighbor nodes b of the second association network.

7. The method of claim 3, further comprising:

if the node i is a common node, determining the edge weight and sigma by using the following formula _j w _ij ：

∑ _j w _ij ＝∑ _a w _ia +∑ _b w _ib ；

Therein, sigma _a w _ia Is the sum of edge weights between the node i and all neighboring nodes a of the first correlation network, the sigma _b w _ib Is the sum of the edge weights between the node i and all the neighboring nodes b of the second correlation network.

8. The method of claim 1, wherein iteratively performing multiple rounds of label propagation on nodes of the federated association network further comprises:

determining marked nodes and unmarked nodes of the federated association network;

updating the labels of the nodes which are not marked in turn until the labels of the nodes which are not marked are not changed and/or exceed an updating turn threshold value; and the number of the first and second groups,

keeping the label of the label node unchanged.

9. The method of claim 1, wherein determining the label of the current round of each node according to the label of the current round of the neighboring node and the label propagation probability of the neighboring node for the node comprises:

for each node, determining a label of each neighbor node of the node in the current round and a label propagation probability of each neighbor node for the node;

calculating the sum of the label propagation probabilities corresponding to each label in all the neighbor nodes of the node to obtain the label propagation aggregation probability corresponding to each label;

and updating the label of the current round of the node according to the label with the maximum label propagation aggregation probability.

10. The method of claim 9, wherein if the node is a non-common node, the method further comprises:

and the node side calculates the label propagation aggregation probability corresponding to each type of neighbor node label of the node.

11. The method of claim 9, wherein if the node is a common node, the method further comprises:

the first party calculates the propagation aggregation probability of the first party labels corresponding to all the neighbor node labels of the node in the first correlation network;

the second party calculates the second label propagation aggregation probability corresponding to all the neighbor node labels of the node in the second correlation network;

the first party and the second party interact the first label propagation aggregation probability and the second label propagation aggregation probability;

and the first party and the second party perform label propagation probability aggregation again based on the mutual information respectively to obtain the label propagation aggregation probability corresponding to each label.

12. The method of claim 1, further comprising:

determining graph weights of the first association network and the second association network according to the node relation closeness degree of the first association network and the second association network; and (c) a second step of,

and introducing the graph weight in the interaction process of the first associated network and the second associated network.

13. The method of claim 12, wherein introducing the graph weight during the interaction between the first and second associated networks comprises:

∑ _j w _ij ＝θ _a ∑ _a w _ia +θ _b ∑ _b w _ib ；

Therein, sigma _a w _ia Is the sum of edge weights between the node i and all neighboring nodes a of the first correlation network, the sigma _b w _ib Is the sum of the edge weights, θ, between the node i and all neighboring nodes b of the second correlation network _a Is a graph weight, θ, of the first correlation network _b A graph weight for the second associated network.

14. The method of claim 12, wherein introducing the graph weight during the interaction between the first and second associated networks comprises:

and after the first party and the second party interact the first label propagation aggregation probability and the second label propagation aggregation probability, performing label propagation probability aggregation again based on the graph weights of the first association network and the second association network to obtain the label propagation aggregation probability corresponding to each label.

15. The method of claim 1, further comprising:

and if the first correlation network and the second correlation network are directed graph networks, only the inflow neighbor node of each node is taken as the neighbor node.

16. A label propagation apparatus associated with a network, comprising:

the graph building module is used for building a first associated network based on the first party data and building a second associated network based on the second party data;

the federal network module is used for associating the first associated network and the second associated network based on a safety traffic protocol to obtain a federal associated network;

the label propagation module is used for iteratively executing multiple rounds of label propagation on the nodes of the federated association network; wherein each round of the tag propagation comprises: determining label propagation probability between adjacent nodes in the federal association graph; and aiming at each node, determining the label of each node according to the label of the neighbor node in the current round and the label propagation probability of the neighbor node to the node.

17. A label propagation apparatus associated with a network, comprising:

at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform: the method of any one of claims 1-15.

18. A computer-readable storage medium storing a program that, when executed by a multi-core processor, causes the multi-core processor to perform the method of any of claims 1-15.