CN-122022972-A - Processing method and device of guest group information, electronic equipment and computer program product

CN122022972ACN 122022972 ACN122022972 ACN 122022972ACN-122022972-A

Abstract

The application discloses a method, a device, electronic equipment and a computer program product for processing guest group information. The method comprises the steps of obtaining guest group information, constructing an information network diagram according to user information of M financial users in the guest group information to obtain an initial information network diagram, wherein the initial information network diagram comprises M nodes, each node indicates user information of one financial user, determining a degree sequence of the initial information network diagram, sub-dividing the initial information network diagram according to the degree sequence to obtain a divided information network diagram, and adjusting the divided information network diagram according to node degree values of each node in the divided information network diagram to obtain a target information network diagram. The application solves the technical problems that the processing efficiency is low and the flow characteristics of ask a guest to stay group information cannot be ensured in the generated information network diagram when the privacy protection processing is carried out on the information network diagram constructed by the guest group information in the related technology.

Inventors

WANG YUANZHENG

Assignees

中国工商银行股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (10)

1. A method for processing guest group information, comprising: acquiring customer group information stored by a financial institution, and constructing an information network diagram according to user information of M financial users in the customer group information to obtain an initial information network diagram, wherein the initial information network diagram comprises M nodes, each node indicates user information of one financial user, each side in the initial information network diagram is a side between nodes corresponding to two financial users with social relations, and M is a positive integer; Determining a degree sequence of the initial information network diagram, and sub-dividing the initial information network diagram according to the degree sequence to obtain a divided information network diagram, wherein the divided information network diagram comprises N divided sub-information network diagrams, N is smaller than M, and N is a positive integer; And adjusting the partitioned information network graph according to the node degree value of each node in the partitioned information network graph to obtain a target information network graph, wherein the node degree value of each node refers to the number of edges directly connected with each node in the partitioned information network graph.
2. The method of claim 1, wherein determining the degree sequence of the initial information network graph comprises: For one node in the initial information network diagram, acquiring the number of edges directly connected with the node to obtain a node degree value; for each node in the initial information network diagram, calculating the similarity of the node degree values of the node and M-1 nodes respectively to obtain a group of similarity data, wherein the group of similarity data comprises M-1 similarity data; Performing similarity clustering on the M nodes to obtain Y node sets, and determining a degree value of each node set to obtain Y set degree values, wherein Y is a positive integer; and arranging the Y aggregation degree values in a descending order to obtain the degree sequence.
3. The method of claim 1, wherein sub-dividing the initial information network graph according to the degree sequence to obtain a divided information network graph comprises: acquiring M nodes associated with the degree sequence, taking each node as a subgraph, and acquiring M initial subgraphs; Acquiring node degree values of the M nodes in the initial information network diagram according to the degree sequence, and calculating a module degree value of the initial information network diagram according to the M node degree values to obtain an initial module degree value; calculating the modularity value of the initial information network graph after combining any two initial subgraphs to obtain R candidate modularity values, wherein R is a positive integer; dividing the M initial subgraphs according to the R candidate modularity values to obtain N network subgraphs, and forming the divided information network diagram by the N sub information network diagrams related to the N network subgraphs.
4. The method of claim 3, wherein partitioning the M initial subgraphs according to the R candidate modularity values to obtain N network subgraphs comprises: respectively calculating the difference values of the R candidate module degree values and the initial module degree values to obtain R increment values, and combining two initial subgraphs corresponding to the increment value with the largest value to obtain a combined subgraph; Obtaining an information network diagram after obtaining the combined subgraph, obtaining a first information network diagram, and calculating a modularity value of the first information network diagram to obtain an updated modularity value; And obtaining the number of the divided sub-images, dividing the combined sub-image and M-1 initial sub-images according to the updated module degree value when the number of the sub-images is smaller than the preset number until the number of the sub-images is equal to the preset number, and determining the combined sub-images as the N network sub-images.
5. The method of claim 1, wherein adjusting the partitioned information network graph according to the node degree value of each node in the partitioned information network graph comprises: For one sub-information network diagram in the divided information network diagram, G nodes in the sub-information network diagram are obtained, and node degree values associated with each node are obtained to obtain G node degree values, wherein G is a positive integer; For an ith node and a jth node in the G nodes, determining modification types of the ith node and the jth node according to the G node degree values, wherein i and j are positive integers; Modifying the ith node and the jth node according to the modification type, and modifying data of the node degree value of the ith node and the node degree value of the jth node to obtain a modified sub-information network diagram; and merging the N modified sub-information network diagrams to obtain the target information network diagram.
6. The method of claim 5, wherein determining the modification types for the ith node and the jth node based on the G node degree values comprises: Acquiring the node degree value of the ith node from the G node degree values to obtain the ith node degree value, and acquiring the node degree value of the jth node from the G node degree value to obtain the jth node degree value; calculating a circulation index value between the ith node and the jth node by using the ith node degree value and the jth node degree value; And acquiring an index mapping rule, and screening the modification types from the index mapping rule according to the circulation index value, wherein the index mapping rule comprises a plurality of candidate modification types and a numerical interval of the circulation index value corresponding to each candidate modification type.
7. The method of claim 5, wherein modifying the ith node and the jth node according to the modification type comprises: Deleting the edge between the ith node and the jth node under the condition that the ith node and the jth node are connected; Adding an edge between the ith node and the jth node in the case that the ith node and the jth node are not connected; Acquiring a kth node in the G nodes, deleting the edge between the ith node and the kth node and adding the edge between the jth node and the kth node under the condition that the ith node and the kth node are connected, wherein k is a positive integer; Acquiring a z-th node in the G nodes, deleting edges between the i-th node and the k-th node, deleting edges between the j-th node and the z-th node, and adding edges between the k-th node and the z-th node when the i-th node is connected with the k-th node and the j-th node is connected with the z-th node, wherein z is a positive integer; and under the condition that the modification type indicates the newly added type, acquiring a newly added node, and adding an edge between the ith node and the newly added node.
8. A guest group information processing apparatus, comprising: The system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring guest group information stored by a financial institution, constructing an information network diagram according to user information of M financial users in the guest group information, and obtaining an initial information network diagram, wherein the initial information network diagram comprises M nodes, each node indicates user information of one financial user, each side in the initial information network diagram is a side between nodes corresponding to two financial users with social relations, and M is a positive integer; A determining unit, configured to determine a degree sequence of the initial information network map, and sub-divide the initial information network map according to the degree sequence to obtain a divided information network map, where the divided information network map includes N divided sub-information network maps, N is smaller than M, and N is a positive integer; And the adjusting unit is used for adjusting the divided information network graph according to the node degree value of each node in the divided information network graph to obtain a target information network graph, wherein the node degree value of each node refers to the number of edges directly connected with each node in the divided information network graph.
9. An electronic device, comprising: A memory storing an executable program; a processor for executing the program, wherein the program executes the guest group information processing method according to any one of claims 1 to 7.
10. A computer program product comprising computer instructions which, when executed by a processor, implement the steps of the method of processing guest group information according to any one of claims 1 to 7.

Description

Processing method and device of guest group information, electronic equipment and computer program product Technical Field The present application relates to the field of financial science and technology, and in particular, to a method, an apparatus, an electronic device, and a computer program product for processing guest group information. Background As economies develop, communications between financial institutions are becoming more frequent, and emerging financial products and services increase the diversity of financial institution collaboration. However, under this collaboration framework, how to process and share guest group data becomes a significant challenge for the industry. Currently, financial institutions need to take data desensitization measures when sharing guest group information, such as by using differential privacy or other techniques, so that personal identity details are not exposed when providing guest group statistics or behavioral profile analysis. Although these methods can guarantee user privacy to a certain extent, the above methods focus on guaranteeing individual data security, failing to effectively retain information reflecting the popularity of the guest group network, and easily causing the loss of key connectivity information in the original data, thereby reducing the value of the data. In addition, a social network model is constructed when the guest group information is shared, and the connection and interaction among individuals in the model form the fluxion, so that the fluxion not only can embody the effective transmission capability of the information in the network, but also can represent the evaluation of the value of potential partners. However, the privacy protection technology mainly focuses on desensitization of the representation data, but fails to fully consider the flow value in the graph attribute, so that financial institutions face the dilemma that the data privacy and the flow value are difficult to balance when information sharing is performed. Aiming at the technical problems that the processing efficiency is low and the flow characteristics of ask a guest to stay group information cannot be ensured when the information network diagram constructed by the guest group information is subjected to privacy protection processing in the related technology, no effective solution is proposed at present. Disclosure of Invention The application mainly aims to provide a processing method, a device, electronic equipment and a computer program product of guest group information, which are used for solving the technical problems that the processing efficiency is low and the flow characteristics of ask a guest to stay group information cannot be ensured when privacy protection processing is carried out on an information network diagram constructed by the guest group information in the related technology. In order to achieve the above object, according to one aspect of the present application, there is provided a method of processing guest group information. The method comprises the steps of obtaining guest group information stored by a financial institution, constructing an information network diagram according to user information of M financial users in the guest group information to obtain an initial information network diagram, wherein the initial information network diagram comprises M nodes, each node indicates user information of one financial user, each side in the initial information network diagram is a side between nodes corresponding to two financial users with social relations, M is a positive integer, determining a degree sequence of the initial information network diagram, sub-dividing the initial information network diagram according to the degree sequence to obtain a divided information network diagram, the divided information network diagram comprises N sub-information network diagrams, N is smaller than M, N is a positive integer, and the divided information network diagram is adjusted according to a node degree value of each node in the divided information network diagram to obtain a target information network diagram, wherein the node degree value of each node refers to the number of sides directly connected with each node in the divided information network diagram. Optionally, determining the degree sequence of the initial information network diagram comprises the steps of obtaining the number of edges directly connected with the nodes for one node in the initial information network diagram to obtain node degree values, calculating the similarity of the node degree values of the node and M-1 nodes for each node in the initial information network diagram to obtain a group of similarity data, wherein the group of similarity data comprises M-1 similarity data, carrying out similarity clustering on the M nodes to obtain Y node sets, determining the degree value of each node set to obtain Y set degree values, wherein Y is a positive integer, and carr