CN118153954A

CN118153954A - Object relationship identification method and device and computer equipment

Info

Publication number: CN118153954A
Application number: CN202410348364.2A
Authority: CN
Inventors: 芦少斌; 俞泱; 马堃; 邱耿峰
Original assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Current assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Priority date: 2024-03-26
Filing date: 2024-03-26
Publication date: 2024-06-07

Abstract

The present application relates to a method, an apparatus, a computer device, a storage medium and a computer program product for identifying an object relationship. The method comprises the following steps: obtaining an object node relation diagram; determining a target class from the classes based on the weights of the directed edges corresponding to each class; under the condition that the target class does not meet the iteration ending condition, determining the object relation among the object nodes of the directed edges corresponding to the target class, determining the target object nodes of the target class, deleting the directed edges corresponding to the target class and the directed edges corresponding to other classes in the object node relation graph, and obtaining an updated object node relation graph by taking the starting point of the edges as the directed edges of the target object nodes; and under the condition that the target class meets the iteration ending condition, carrying out aggregation processing on the object relations among the object nodes determined in each iteration process to obtain the finally identified object relation. By adopting the method, the object relationship can be accurately identified.

Description

Object relationship identification method and device and computer equipment

Technical Field

The present application relates to the field of big data technology, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for identifying an object relationship.

Background

In the risk management process of the object, the risk factors associated with the object are paid attention to, and taking the risk management process of the enterprise as an example, the enterprise can be helped to identify potential risk factors by identifying the upstream and downstream enterprise information of the enterprise, such as the reliability of the provider of the enterprise, the change of the demand of the client of the enterprise, and the like. Therefore, it is particularly important to identify relationships between businesses.

In the conventional technology, taking an enterprise as an example, a common method for identifying relationships between enterprises includes: market research and visit are carried out through a client manager, and the relationship between suppliers and clients of the enterprise is manually analyzed and recorded; or, the bill information of the enterprise transaction is utilized to identify the upstream enterprise and the downstream enterprise, for example, if the bill amount between the two enterprises is greater than a certain threshold value, the enterprises are considered to have a supply relationship, and the upstream enterprise or the downstream enterprise is determined based on the supply relationship.

However, the existing relationship recognition method of the object relationship has a problem of being inaccurate, and a method for accurately recognizing the object relationship is needed.

Disclosure of Invention

In view of the foregoing, it is desirable to provide an accurate object relationship identification method, apparatus, computer device, computer readable storage medium, and computer program product.

In a first aspect, the present application provides a method for identifying an object relationship, including:

obtaining an object node relation diagram, wherein the object node relation diagram comprises object nodes and directed edges between the object node pairs, and the attributes of the directed edges comprise weights and categories;

determining a target category from the categories based on the weights of the directed edges corresponding to the categories;

Under the condition that the target class does not meet the iteration ending condition, determining an object relation among object nodes of directed edges corresponding to the target class, determining the target object nodes of the target class, deleting the directed edges corresponding to the target class and the directed edges corresponding to other classes in the object node relation graph, wherein the starting point of the edges is the directed edge of the target object nodes, and obtaining an updated object node relation graph;

and under the condition that the target class meets the iteration ending condition, carrying out aggregation processing on the object relations among the object nodes determined in each iteration process to obtain the finally identified object relation.

In one embodiment, the obtaining the object node relation graph includes:

determining objects to be identified, and acquiring a resource transfer relationship between the objects to be identified;

taking the object to be identified as an object node, obtaining a directed edge between object node pairs based on the resource transfer relationship, and determining the attribute of the directed edge;

And acquiring an object node relation graph based on the object node, the directed edge and the attribute of the directed edge.

In one embodiment, the determining the attribute of the directed edge includes:

acquiring first resource transfer data between the type of the object to be identified and the object to be identified;

determining the category of the directed edge based on the category of the object to be identified;

Determining weights of the directed edges based on the first resource transfer data;

and determining the attribute of the directed edge according to the weight and the category.

In one embodiment, the first resource transfer data includes a single transfer resource amount and a total number of resource transfers;

The determining the weight of the directed edge based on the first resource transfer data includes:

Determining the weight of the directed edge in unit time based on the single transfer resource amount and the total number of resource transfers in unit time;

Determining a first weight of the directed edge in a preset first time period and a second weight of the directed edge in a preset second time period based on the weight of the unit time;

And determining the weight of the directed edge according to the first weight and the second weight.

In one embodiment, the determining the target class from the classes based on the weight of the directed edge corresponding to each class includes:

Obtaining the sum of the weights of the directed edges corresponding to the categories;

And determining the category with the maximum sum of the weights as a target category.

In one embodiment, the determining the target object node of the target class includes:

Acquiring second resource transfer data of candidate nodes based on the directed edges corresponding to the target categories, wherein the candidate nodes comprise starting points of the directed edges corresponding to the target categories;

Acquiring third resource transfer data of the candidate node based on the directed edge corresponding to the object node relation diagram;

And determining the candidate node with the ratio of the second resource transfer data to the third resource transfer data being greater than a preset threshold value as a target object node.

In one embodiment, the iteration end condition includes: and the sum of the weights of the directed edges corresponding to the target categories is smaller than the average value of the weights of the directed edges in the initial object node relation diagram.

In a second aspect, the present application also provides an apparatus for identifying an object relationship, the apparatus comprising:

The node relation diagram acquisition module is used for acquiring an object node relation diagram, wherein the object node relation diagram comprises object nodes and directed edges between the object node pairs, and the attributes of the directed edges comprise weights and categories;

The target category determining module is used for determining a target category from the categories based on the weight of the directed edge corresponding to each category;

The iteration module is used for determining the object relation between the object nodes of the directed edges corresponding to the target categories under the condition that the target categories do not meet the iteration ending condition, determining the target object nodes of the target categories, deleting the directed edges corresponding to the target categories and the directed edges corresponding to other categories in the object node relation graph, and obtaining an updated object node relation graph, wherein the starting points of the edges are the directed edges of the target object nodes;

and the aggregation module is used for carrying out aggregation processing on the object relationships among the object nodes determined in each iteration process under the condition that the target category meets the iteration ending condition, so as to obtain the finally identified object relationship.

In one embodiment, the node relation diagram obtaining module is further configured to determine an object to be identified, and obtain a resource transfer relation between the objects to be identified; taking the object to be identified as an object node, obtaining a directed edge between object node pairs based on the resource transfer relationship, and determining the attribute of the directed edge; and acquiring an object node relation graph based on the object node, the directed edge and the attribute of the directed edge.

In one embodiment, the node relation diagram obtaining module is further configured to obtain first resource transfer data between the kind of the object to be identified and the object to be identified; determining the category of the directed edge based on the category of the object to be identified; determining weights of the directed edges based on the first resource transfer data; and determining the attribute of the directed edge according to the weight and the category.

In one embodiment, the first resource transfer data includes a single transfer resource amount and a total number of resource transfers; the node relation diagram obtaining module is further configured to determine a weight per unit time of the directed edge based on the single transfer resource amount and the total number of resource transfers in unit time; determining a first weight of the directed edge in a preset first time period and a second weight of the directed edge in a preset second time period based on the weight of the unit time; and determining the weight of the directed edge according to the first weight and the second weight.

In one embodiment, the target class determining module is further configured to obtain a sum of weights of the directed edges corresponding to the classes; and determining the category with the maximum sum of the weights as a target category.

In one embodiment, the iteration module is further configured to obtain second resource transfer data of a candidate node based on the directed edge corresponding to the target class, where the candidate node includes a starting point of the directed edge corresponding to the target class; acquiring third resource transfer data of the candidate node based on the directed edge corresponding to the object node relation diagram; and determining the candidate node with the ratio of the second resource transfer data to the third resource transfer data being greater than a preset threshold value as a target object node.

In a third aspect, the present application also provides a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

In a fourth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

In a fifth aspect, the application also provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of:

The object relation identification method, the device, the computer equipment, the storage medium and the computer program product acquire an object node relation graph, determine target classes through the weights of the directed edges of the object node relation graph, determine the object relation among the object nodes of the directed edges corresponding to the target classes under the condition that the target classes do not meet the iteration ending condition, determine the target object nodes of the target classes, delete the directed edges corresponding to the target classes and the directed edges corresponding to other classes in the object node relation graph, and acquire the updated object node relation graph. In the whole process, the target class is accurately determined based on the weight of the directed edge in the object node relation graph, and in the case that the target class does not meet the iteration ending condition, the directed edge corresponding to the target class in the object node relation graph and the directed edge corresponding to other classes are deleted, the starting point of the edge is the directed edge of the target object node of the target class, the interference of the determined target object node in the target class to the subsequent iteration process is avoided, the accurate repeated iteration of the target object node is realized, and the object relation among the determined object nodes in each iteration process is aggregated when the iteration is ended, so that the object relation is accurately identified.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort for those skilled in the art.

FIG. 1 is an application environment diagram of a method of identifying object relationships in one embodiment;

FIG. 2 is a flow chart of a method for identifying object relationships in one embodiment;

FIG. 3 is a flowchart of a method for identifying object relationships in another embodiment;

FIG. 4 is a flow chart of a method for identifying object relationships in a specific application example;

FIG. 5 is a block diagram of an object relationship identification apparatus in one embodiment;

FIG. 6 is an internal block diagram of a computer device in one embodiment;

Fig. 7 is an internal structural view of a computer device in another embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are both information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to meet the related regulations.

The object relationship identification method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.

The object relation identification method provided by the embodiment of the application can also be used in an application environment where the terminal and the server act together.

For example, a user performs an operation on the terminal 102, sends an object relationship identification request, the terminal 102 sends the object relationship identification request to the server 104, and the server 104 obtains an object node relationship graph corresponding to the object relationship identification request from a database based on the object relationship identification request, wherein the object node relationship graph comprises object nodes and directed edges between object node pairs, and the attribute of the directed edges comprises weights and categories; determining a target class from the classes based on the weights of the directed edges corresponding to each class; under the condition that the target class does not meet the iteration ending condition, determining the object relation among the object nodes of the directed edges corresponding to the target class, determining the target object nodes of the target class, deleting the directed edges corresponding to the target class and the directed edges corresponding to other classes in the object node relation graph, and obtaining an updated object node relation graph by taking the starting point of the edges as the directed edges of the target object nodes; and under the condition that the target class meets the iteration ending condition, carrying out aggregation processing on the object relations among the object nodes determined in each iteration process to obtain the finally identified object relation. Further, the server 104 sends the final identified object relationships to the terminal 102 to present the final identified object relationships to the user on the terminal 102.

The object relationship identification method provided by the embodiment of the application can also be independently used in the application environment of the terminal or the server.

For example, a user performs an operation on the terminal 102, sends out an object relationship identification request, and the terminal 102 obtains an object node relationship graph corresponding to the object relationship identification request from a local place based on the object relationship identification request, wherein the object node relationship graph comprises object nodes and directed edges between object node pairs, and the attribute of the directed edges comprises weights and categories; determining a target class from the classes based on the weights of the directed edges corresponding to each class; under the condition that the target class does not meet the iteration ending condition, determining the object relation among the object nodes of the directed edges corresponding to the target class, determining the target object nodes of the target class, deleting the directed edges corresponding to the target class and the directed edges corresponding to other classes in the object node relation graph, and obtaining an updated object node relation graph by taking the starting point of the edges as the directed edges of the target object nodes; and under the condition that the target class meets the iteration ending condition, carrying out aggregation processing on the object relations among the object nodes determined in each iteration process to obtain the finally identified object relation.

For another example, the server 104 obtains an object node relationship graph, where the object node relationship graph includes object nodes and directed edges between object node pairs, and attributes of the directed edges include weights and categories; determining a target class from the classes based on the weights of the directed edges corresponding to each class; under the condition that the target class does not meet the iteration ending condition, determining the object relation among the object nodes of the directed edges corresponding to the target class, determining the target object nodes of the target class, deleting the directed edges corresponding to the target class and the directed edges corresponding to other classes in the object node relation graph, and obtaining an updated object node relation graph by taking the starting point of the edges as the directed edges of the target object nodes; and under the condition that the target class meets the iteration ending condition, carrying out aggregation processing on the object relations among the object nodes determined in each iteration process to obtain the finally identified object relation.

In an exemplary embodiment, as shown in fig. 2, a method for identifying an object relationship is provided, and an example in which the method is applied to the computer device in fig. 1 is described, where the computer device includes a terminal or a server. Wherein:

S200, obtaining an object node relation diagram.

The object node relation graph comprises object nodes and directed edges between object node pairs, and the attributes of the directed edges comprise weights and categories.

In the application, the object node relation diagram is generally characterized by adopting the relation of a bipartite diagram, but can also be characterized by adopting other network structures. The bipartite graph refers to that an object node in the object node relation graph can be divided into two node subsets, nodes in the node subsets are mutually disjoint, and two nodes attached to each edge in the graph belong to the two mutually disjoint subsets. The directional side is a side having directivity.

Two node subsets in the object node relation diagram are respectively a resource transfer party and a resource receiving party. For example, in the case that the object node is an enterprise, that is, the object node relationship graph is a node relationship graph between enterprises, the existing enterprise A, B, C is a money transfer relationship, where a and B have a transfer relationship, a to B, A and C have a transfer relationship, a to C, B and C have a transfer relationship, and B to C have a transfer relationship, the obtained object node relationship graph is:

A→B

A→C

B→C

and the object node relation graph can be simplified, and the simplified object node relation graph is obtained. I.e. merging the same object node inside each subset.

It can be seen that the object node relationship graph includes object nodes A, B, and C; object node pairs AB, AC, BC; directed edges between pairs of object nodes represent transition relationships between pairs of object nodes. The directed edge itself has attributes including weights and categories. The weight refers to the weight of resource transfer, and in different application scenarios, the category can represent different information, taking an object as an enterprise node as an example, and the category refers to an industry pair between enterprises, for example, an industry pair between enterprise a and enterprise B. The classes of the directed edges of different object node pairs may or may not be identical, for example, the industry pair between enterprise a and enterprise B may be industry pair 1-2, the industry pair between enterprise a and enterprise C may be industry pair 1-2, and the industry pair between enterprise B and enterprise C may be industry pair 1-3.

Specifically, the computer device obtains an object node relation diagram, and analyzes the object node relation diagram to obtain object nodes in the object node relation diagram, directed edges between object node pairs, and weights and categories of the directed edges. In the case where the object node is an enterprise node, the computer device obtains the enterprise node, a directed edge between enterprises, and a transfer weight and industry pair of the directed edge in the object node relationship graph.

S400, determining a target category from the categories based on the weight of the directed edge corresponding to each category.

Specifically, the classes of the directed edges between the object node pairs may or may not be consistent. Thus, each class may correspond to a directed edge of at least one object node pair. And the directed edge of each object node pair has a weight, that is, each class may correspond to the weight of the directed edge of at least one object node pair. The weights of the directed edges corresponding to the categories are overlapped to obtain weights of the categories, the weights of the categories are processed to obtain a processing result, and the target category is determined from the categories based on the processing result. For example, when the industry pair between the enterprise a and the enterprise B is the industry pair 1-2, the industry pair between the enterprise a and the enterprise C is the industry pair 1-2, and the industry pair between the enterprise B and the enterprise C is 1-3, the sum of weights of the respective enterprise pairs of the industry pair 1-2 is the sum of weights of the enterprise a and the enterprise B, and the sum of weights of the enterprise a and the enterprise C, and the sum of weights of the respective enterprise pairs of the industry pair 1-2 is the weight of the enterprise B and the enterprise C.

S600, determining object relations among object nodes of directed edges corresponding to the target classes under the condition that the target classes do not meet iteration ending conditions, determining the target object nodes of the target classes, deleting the directed edges corresponding to the target classes and the directed edges corresponding to other classes in the object node relation graph, wherein the starting points of the edges are the directed edges of the target object nodes, and obtaining an updated object node relation graph.

The target object node is a starting point in the directed edge corresponding to the target class.

Specifically, the target class is analyzed, and if the target class does not meet the iteration ending condition, the description also needs to continue to iteratively identify the object relationship. First, the object relationship between object nodes of the directed edge corresponding to the target type obtained in the first iteration process can be determined, and the object relationship between object nodes of the directed edge corresponding to the target type is recorded.

After the object relationships among the object nodes of the directed edges corresponding to the target types are recorded, the object relationships among the object nodes of the directed edges corresponding to the target types can be separated from other types, and the object relationships among the object nodes of the directed edges corresponding to the target types are independently stored, namely, the directed edges corresponding to the target types in the object node relationship graph are deleted, and the object node relationship graph is updated. Further, the directional edges corresponding to the other categories may be selectively updated according to the weights of the directional edges corresponding to the target categories, specifically, the target object nodes satisfying the preset conditions are screened from the starting points in the directional edges corresponding to the target categories, and the target object nodes corresponding to the other categories are deleted, wherein the starting points of the edges are the directional edges of the target object nodes, which is substantially that the target object nodes in the target categories have satisfied the preset conditions, and the directional edges of the target object nodes in the other categories do not need to be considered any more, so that the deletion operation may be performed. And obtaining an updated object node relation diagram, wherein the updated object node relation diagram does not have directed edges corresponding to the target class and the directed edges corresponding to other classes, and the starting point of the edges is the directed edge of the target object node.

And in the updated object node relation diagram, continuously determining a target class from the classes based on the weights of the directed edges corresponding to each class, and determining the object relation between the object nodes of the directed edges corresponding to the target class until the iteration ending condition is met.

S800, under the condition that the target class meets the iteration ending condition, the object relations among the object nodes determined in each iteration process are aggregated, and finally the identified object relations are obtained.

Specifically, under the condition that the target class meets the iteration ending condition, the object relations among the object nodes determined in each iteration process are aggregated, and finally identified object relations are obtained. For example, let the object nodes be A, B, C, D, E, the object relationship between the object nodes determined in the first iteration process be a→ B, A →c, the object relationship between the object nodes determined in the second iteration process be b→ E, C →e, the object relationship between the object nodes determined in the third iteration process be d→c, after the third iteration process, the updated object category satisfies the iteration end condition, the object relationships a→ B, A → C, B → E, C → E, D →c are summarized, and the finally identified object relationship is a→b→ E, A →c→ E, D →c. That is, if A, B, C, D, E is enterprise and the resource is money, then the upstream enterprise of enterprise B is enterprise E and the downstream enterprise is enterprise a; the downstream enterprise of enterprise C is enterprise a or enterprise D, and the upstream enterprise is enterprise E.

In the method for identifying the object relationship, the object node relationship graph is obtained, the target class is determined through the weight of the directed edge of the object node relationship graph, the object relationship among the object nodes of the directed edge corresponding to the target class is determined under the condition that the target class does not meet the iteration ending condition, the target object node of the target class is determined, the directed edge corresponding to the target class in the object node relationship graph and the directed edge corresponding to other classes are deleted, the starting point of the edge is the directed edge of the target object node, and the updated object node relationship graph is obtained. In the whole process, the target class is accurately determined based on the weight of the directed edge in the object node relation graph, and in the case that the target class does not meet the iteration ending condition, the directed edge corresponding to the target class in the object node relation graph and the directed edge corresponding to other classes are deleted, the starting point of the edge is the directed edge of the target object node of the target class, the interference of the determined target object node in the target class to the subsequent iteration process is avoided, the accurate repeated iteration of the target object node is realized, and the object relation among the determined object nodes in each iteration process is aggregated when the iteration is ended, so that the object relation is accurately identified.

In one exemplary embodiment, as shown in fig. 3, S200 includes:

s220, determining objects to be identified, and acquiring resource transfer relations among the objects to be identified.

S240, taking the object to be identified as an object node, obtaining a directed edge between the object node pair based on the resource transfer relation, and determining the attribute of the directed edge.

S260, acquiring an object node relation graph based on the object nodes, the directed edges and the attributes of the directed edges.

The object to be identified is an object that needs to be identified in relation to each other, and in a specific field, the object to be identified may refer to different objects, for example, the object to be identified may refer to an enterprise. The resource transfer relationship refers to a relationship that a resource transfer party transfers a resource to a resource receiving party, that is, the resource transfer relationship includes the resource transfer party and the resource receiving party, and if the object to be identified is an enterprise, the object to be identified may be an enterprise belonging to the same industry chain.

Specifically, the objects to be identified are determined, and the resource transfer relationship between the objects to be identified is acquired. In order to construct the object node relation diagram, first, the object nodes in the object node relation diagram, and the directed edges between the object nodes need to be constructed. The application takes the object to be identified as the object node, and determines the resource transfer party and the resource receiving party through the resource transfer relation, and one object to be identified can be the resource transfer party in one resource transfer relation or the resource receiving party in another resource transfer relation.

Based on the resource transfer party and the resource receiving party, a directed edge is arranged between object nodes in the object node relation diagram, one end of the directed edge, which is close to an arrow, is the resource receiving party, and one end of the directed edge, which is far away from the arrow, is the resource transfer party. Further, the attribute of the directed edge may also be determined, i.e. the weight and class of the directed edge may be determined, wherein the class of the directed edge is essentially the class of the object node pair to which the directed edge corresponds.

In this embodiment, by acquiring the resource transfer relationships between the objects to be identified, the resource transfer party and the resource receiving party corresponding to each resource transfer relationship between the objects to be identified can be accurately determined, and the directed edges between the object node pairs can be accurately established based on the resource transfer party and the resource receiving party corresponding to each resource transfer relationship, so that the object node relationship graph can be accurately established.

In one exemplary embodiment, determining the properties of the directed edge includes:

acquiring the type of an object to be identified and resource transfer data between the objects to be identified; determining the class of the directed edge based on the class of the object to be identified; determining the weight of the directed edge based on the resource transfer data; and determining the attribute of the directed edge according to the weight and the category.

The resource transfer data is the resource data of the object to be identified, which is used as the resource transfer party, transferred to the object to be identified of the resource receiving party, and the resource data can be data of transaction amount and the like which can be transferred.

Specifically, the type of the object to be identified is obtained, the type of the object to be identified is taken as the type of the object node, and the type of the directed edge is determined based on the types of the two object nodes corresponding to the directed edge. For example, in the case that the object to be identified is an enterprise, there is one directed edge, two corresponding object nodes are enterprise a and enterprise B, the type of enterprise a is industry 1, the type of enterprise B is industry 2, and the type of directed edge is industry pair 12.

Further, resource transfer data between the objects to be identified are obtained, and the resource transfer data of two object nodes corresponding to the directed edges are obtained based on the resource transfer data between the objects to be identified. The weight of the directed edge is determined based on the resource transfer data of the two object nodes corresponding to the directed edge. It can be understood that the weight of the directed edge represents the resource transfer data of the two object nodes corresponding to the directed edge, so that the greater the weight of the directed edge, the more the resource transfer data of the two object nodes corresponding to the directed edge; the smaller the weight of a directed edge, the less resource transfer data it corresponds to two object nodes.

After the weights and the categories of the directed edges are determined, the weights and the categories of the directed edges can be used as attributes of the directed edges and are integrated with the object nodes and the directed edges to construct an object node relation graph.

In this embodiment, the weight of the directed edge can be accurately determined by processing the resource transfer data, and the category of the directed edge can be accurately determined by the category of the object to be identified, so that the attribute of the directed edge is obtained by the category and the weight of the directed edge, and the object node relation graph is efficiently and accurately constructed.

In one exemplary embodiment, the resource transfer data includes a single transfer resource amount and a total number of resource transfers;

determining weights for the directed edges based on the resource transfer data, comprising:

Determining the weight of the directed edge in unit time based on the single transfer resource amount and the total number of resource transfer times in unit time; determining a first weight of the directed edge in a preset first time period and a second weight of the directed edge in a preset second time period based on the unit time weight; and determining the weight of the directed edge according to the first weight and the second weight.

Wherein, the single transfer resource amount refers to the amount of resource transfer in one resource transfer process. The unit time is a custom time period, and for example, each month may be set as a unit time, or each day may be set as a unit time.

Specifically, taking a unit time as an example of each month, acquiring the single transfer resource amount and the total number of resource transfer times in each month, and determining the month weight of the directed edge. The method comprises the steps of obtaining the total times of resource transfer processes in each month and the resource quantity transferred in each resource transfer process among object nodes corresponding to the directed edges, and determining the month weight of the directed edges based on the total times of the resource transfer processes in each month and the resource quantity transferred in each resource transfer process. The resource transfer process refers to a process of transferring resources along the direction of the directed edge, and does not include a process of transferring resources in the direction of the reverse directed edge.

Specific expressions of the month weight of the directed edge include: f=Σlog [ (s/10000) +1- (1/C) ]. Wherein s is the resource quantity transferred in each resource transfer process in each month, C is the total number of resource transfer processes in each month, and F is the month weight of the directed edge.

Further, in the process of identifying the object relationship, the time span of resource transfer between objects needs to be considered, so the application comprehensively determines the weight of the directed edge through the weights in two preset time periods. The preset time period comprises a first preset time period and a second preset time period, the first preset time period can be a near-term time period, the second time period can be a long-term time period, and the weights of the directed edges are comprehensively determined through the two weights of the near-term time period and the long-term time period. For the object to be identified to be an enterprise, the first preset time period may be a recent time period, and the second preset time period may be a long-term time period, or may be a repayment period between enterprises, for example, when the repayment period between enterprises is a year, the second preset time period may be a year.

Let the first preset time period be t ₁, the second preset time period be t ₂, and the specific expression for determining the weight of the directed edge through the two weights in the first preset time period and the second preset time period includes: weight of directed edge = weight of directed edge in a x t ₁ + weight of directed edge in b x t ₂.

The weight of the directional edge in t ₁ is the weight obtained by adding the weights of the directional edges in unit time in t ₁; the weight of the directional edge in t ₂ is the weight obtained by adding the weights of the directional edges in unit time in t ₂; a is a first influence factor, b is a second influence factor, a+b=1, and the size relationship between a and b is not clearly defined, a > b when the traffic demand is more biased to the weight requiring the near-term period, a < b when the traffic demand is more biased to the weight requiring the far-term period, and a=b when the traffic demand is equal to the demand for the near-term period and the far-term period.

For example, the first preset time period may be within about six months, the second preset time period may be within about 12 months, the month weight of each month oriented edge within about 12 months is obtained, and the month weights of the 6 month oriented edges are added to obtain the month weight of the 6 month oriented edges; and adding the month weights of the directed edges in the last 12 months to obtain the month weights of the directed edges in the last 12 months.

In this embodiment, the weight of the directed edge in the unit time can be accurately calculated based on the single transfer resource amount and the total number of resource transfers, so as to obtain the weight in the preset time period, and the preset time period includes the first preset time period and the second preset time period, so that the recent weight and the distant weight can be comprehensively considered, and more accurate weight of the directed edge can be obtained, and interference caused by only considering the recent weight or the distant weight to the calculation of the weight of the directed edge is avoided.

In one exemplary embodiment, determining a target class from the classes based on the weights of the directed edges corresponding to each class includes:

obtaining the sum of the weights of the directed edges corresponding to the categories; and determining the category with the maximum sum of the weights as the target category.

Specifically, each class comprises at least one directed edge, the weights of all the directed edges under each class are obtained, and the weights of all the directed edges under each class are added to obtain the weight of each class, namely the sum of the weights of the directed edges corresponding to each class.

It can be understood that the larger the sum of the weights of the directed edges under each category is, the more the resource transfer data under the corresponding category is, and the category can be regarded as the category which obviously exists in the actual scene; conversely, the smaller the sum of the weights of the directed edges under each category, the less the resource transfer data under its corresponding category, and this category can be regarded as a category that does not need attention in the actual scenario.

For example, when the object to be identified is an enterprise and the resource transfer data is a transaction amount, if the sum of the weights of the directed edges between the enterprise nodes under a certain industry pair is larger, the transaction amount under the corresponding industry pair is larger, and the industry pair is an industry pair existing under a real economic scene; if the sum of the weights of the directed edges between the enterprise nodes under a certain industry pair is smaller, the corresponding transaction amount under the industry pair is smaller, and the industry pair is the industry pair which does not need to pay attention in a real economic scene.

Therefore, the application can screen through the weight sum of the directed edges corresponding to each class, determine the class with the largest weight sum, and determine the class with the largest weight sum as the target class, namely the class with the largest weight sum of the corresponding directed edges as the class under the real scene. If the object to be identified is an enterprise and the category is an industry pair, the industry pair with larger transaction amount between the enterprise pairs is determined to be a target industry pair existing in a real economic scene. The application can also adopt a method for sorting the weight sum of the directed edges corresponding to each class, sorts the weight sum of the classes from big to small, selects the class ranked at the forefront, and determines the class as the target class.

In one embodiment, the categories are ranked from large to small in the sum of weights, and the top N categories are selected and determined to be target categories. That is, the target class determined by the present application is not unique during each iteration.

In this embodiment, by acquiring the sum of the weights of the directed edges corresponding to each class and determining the class with the largest sum of the weights as the target class, the class in the real scene can be accurately determined based on the sum of the weights of the directed edges in each class, and thus a more accurate object relationship can be identified.

In one exemplary embodiment, determining a target object node of a target class includes:

Acquiring second resource transfer data of candidate nodes based on the directed edges corresponding to the target categories, wherein the candidate nodes comprise starting points of the directed edges corresponding to the target categories; acquiring third resource transfer data of the candidate node based on the directed edge corresponding to the object node relation diagram; and determining the candidate node with the ratio of the second resource transfer data to the third resource transfer data being greater than a preset threshold value as a target object node.

Specifically, when the target class does not meet the iteration end condition, the directed edges between the object nodes in other classes are updated based on the object nodes in the target class. First, the starting point of the directed edge under the target class is used as a candidate node, that is, the resource transfer party in each pair of resource transfer relations under the target class is used as a candidate node. And obtaining resource transfer data of the candidate nodes, namely second resource transfer data, taking the second resource transfer data as the resource transfer data of the candidate nodes under the target category, and obtaining total resource transfer data of the candidate nodes, namely the directed edges corresponding to the object node relation diagram, wherein the candidate nodes are taken as all resource transfer data of the resource transfer party. And acquiring the proportion of the second resource transfer data and the third resource transfer data, comparing the proportion with a preset threshold value, and determining the candidate node as a target object node if the proportion of the second resource transfer data and the third resource transfer data of the candidate node is larger than the preset threshold value.

It can be understood that the target class is the most likely class in the actual scenario, if the proportion of the resource transfer data of the target node in the total resource transfer data of the target node in the target class is greater than a preset threshold, the resource transfer data of the target node in the target class is represented to have reached the threshold, and even if the resource transfer data still exists in other classes, the resource transfer data in the target class can be considered only and the resource transfer data in other classes is not required to be paid attention. Further, after the target object node of the target class is determined, the target object node of the other class may be deleted, and the start point of the edge is the directed edge of the target object node, and the deleted directed edge is not required to be considered when the object relationship is identified.

For example, in the case where the object to be identified is the enterprise A, B, C, D, E, F, the target class is the industry pair 1, and each object node pair and directed edge under the industry pair 1 includes: other categories, a→ B, A → C, B →d, exist, such as industry pair 2, and the object node pairs and directed edges under industry pair 2 include: a→ F, B →E. The candidate nodes under the industry pair 1 are A and B, the resource transfer data of A-B is 100, the resource transfer data of A-C is 700, the resource transfer data of A-F is 100, and the preset threshold value is 750, and it can be seen that under the industry pair 1, when the resource transfer data of A serving as the starting point of the directed edge exceeds the preset threshold value, the A serving as the target node is taken as the target node, and at the moment, the resource transfer data of A serving as the starting point of the directed edge under the industry pair 2 does not need to be concerned, the directed edge of A-F is deleted, and the weight sum of the directed edge corresponding to the industry pair 2 is updated. Let the resource transfer data of b→d be 500, the resource transfer data of b→e be 400, and the preset threshold value be 750, it can be seen that, under the industry pair 1, if the resource transfer data of B as the starting point of the directed edge does not exceed the preset threshold value, B cannot be taken as the target object node, and b→e under the industry pair 2 is still reserved.

In other embodiments, not only the candidate node whose ratio of the second resource transfer data to the third resource transfer data is greater than the preset threshold may be determined as the target object node; since the weight of the directed edge can be determined based on the resource transfer data, a candidate node having a ratio of the sum of weights under the target category to the sum of total weights greater than a preset threshold can be determined as the target object node, wherein the sum of total weights includes the sum of weights of the candidate nodes in the directed edge corresponding to the object node relationship graph.

In the above example, if the object to be identified is the enterprise A, B, C, D, E, F, the target class is the industry pair 1, and each object node pair and directed edge in the industry pair 1 includes: other categories, a→ B, A → C, B →d, exist, such as industry pair 2, and the object node pairs and directed edges under industry pair 2 include: a→ F, B →E. The candidate nodes under the industry pair 1 are a and B, the weight of a→b is 10%, the weight of a→c is 70%, the weight of a→f is 10%, and the preset threshold is 75%, and it can be seen that under the industry pair 1, when the weight of a as the starting point of the directed edge has exceeded the preset threshold, the candidate node is the target node, and at this time, the weight of a as the starting point of the directed edge under the industry pair 2 is not required to be paid attention, the directed edge of a→f is deleted, and the sum of the weights of the directed edges corresponding to the industry pair 2 is updated. Let the weight of b→d be 50%, the weight of b→e be 40%, and the preset threshold be 75%, it can be seen that, under the industry pair 1, the weight of B as the starting point of the directed edge does not exceed the preset threshold, B cannot be taken as the target node, and still b→e under the industry pair 2 remains.

In this embodiment, by acquiring the second resource transfer data of the candidate node under the target category and the third resource transfer data of the candidate node in the object node relationship graph, and determining the candidate node with the ratio of the second resource transfer data to the third resource transfer data being greater than the preset threshold as the target object node, it is ensured that the resource transfer data of the target object node is the resource transfer data under the real scene, and if the resource transfer data of the target object node meets the preset condition, only the resource transfer data under the target category needs to be considered, and no influence of less resource transfer data under other categories on relationship identification of the target object node needs to be paid attention.

In one exemplary embodiment, the iteration end condition includes: the sum of the weights of the directed edges corresponding to the target categories is smaller than the average value of the weights of the directed edges in the initial object node relation graph.

Specifically, after repeated iteration, when the sum of the weights of the directed edges corresponding to the target classes in a certain iteration process is smaller than the average value of the weights of the directed edges in the initial object node relation graph, the iteration process is terminated. It can be understood that the sum of the weights of the directed edges corresponding to the target classes is smaller than the average value of the weights of the directed edges in the initial object node relation graph, which means that the class of the largest sum of weights, namely the class closest to the actual scene in the reserved class, cannot meet the average value of the weights of the directed edges in the object node relation graph in the initial iteration process, and at this time, the resource transfer weight of the directed edges in the class of the largest sum of weights is also smaller, so that the iteration can be directly terminated without paying attention to the class of the largest sum of weights in the subsequent iteration process.

In another embodiment, the iteration end condition is not limited to an average value of weights of the directed edges in the initial object node relation graph, but may be lower than a specific weight value, and the iteration end condition may be that a preset number of iterations is reached.

In this embodiment, by setting the iteration end condition, it can be accurately determined whether iteration is still required, and the iteration end condition is set as: the sum of the weights of the directed edges corresponding to the target categories is smaller than the average value of the weights of the directed edges in the initial object node relation graph, and when the sum of the weights of the directed edges of the target categories is smaller than the average value, iteration is stopped, so that the accurate objects in the categories of the real scene are identified, but the objects in the categories with smaller resource transfer data and no attention are not required to be paid to.

In a specific embodiment, as shown in fig. 4, the object to be identified is an enterprise, and the class of the directed edge is an industrial pair.

S1, a data preparation process.

And acquiring the account transfer and expenditure detail data, the industry and the industry chain of each month of the enterprise, wherein the industry chain of the enterprise is the same industry chain. And determining other enterprises which transact with the enterprises based on the account transfer expense detail data of each month of the enterprises, determining the expense relation of the transactions among the enterprises, forming enterprise pairs among the enterprises with the expense relation, and representing the enterprises by a binary pattern form to obtain an enterprise node relation diagram. For example, when enterprise A pays out to enterprise B, enterprise B is the upstream enterprise of enterprise A, and enterprise A is the downstream enterprise of enterprise B. Let the enterprise node relation diagram be:

A→B

A→C

A→D

B→C

D→E

Let the industry of enterprise A be 1, the industry of enterprise B, D be 2, the industry of enterprise C and enterprise E be 3, and the industry pair of enterprise pair A- & gt B, A- & gtD be 12; the industrial pair of enterprises A-C is 13; the industrial pair of enterprises B-C, D-E is 23.

The industry pair with larger transaction amount summarized by the enterprise pair relationship is necessarily the industry pair existing in the real economic scene, and conversely, when the transaction summary is very small, the industry pair with smaller proportion of economic transaction can be partially reflected, and the attention can be omitted.

S2, calculating the expenditure weight of the edge in the enterprise node relation graph.

Taking the monthly expense weight as the unit time expense weight, and acquiring the monthly expense weight between the enterprise pairs based on the expense detail data between the enterprise pairs, for example, the monthly expense weight between the enterprise pairs AB=Σlog [ (the monthly total expense times of the enterprise A expense to the enterprise B per expense amount/10000) +1- (1/the enterprise A expense to the enterprise B) ].

And because the expense weight of the edge in the enterprise node relation graph is used for measuring the industrial chain where the enterprise is located, the time span of enterprise expense needs to be considered, and the expense weight of the edge in the enterprise node relation graph is determined based on the expense conditions of half a year and one year of the enterprise, and the expense weight of the edge in the enterprise node relation graph=a is equal to the expense weight of the 6 months and +b is equal to the expense weight sum of 12 months.

Wherein a+b=1, the near 6 month payout weight sum reflects the near-industry chain condition, and the near 12 month payout weight sum considers that there is an order between enterprises to be a repayment period in a year.

Finally, the side of the enterprise to the AB (enterprise A, industry of enterprise B, industry chain, and expenditure weight of enterprise side) is obtained. And similarly, obtaining the edges of other enterprise pairs.

3. A target industry pair is selected.

And obtaining the weight of each industry pair according to the sum of the weights of all sides under the same industry pair, and determining the target industry pair with the maximum weight. Further, an N pair target industry pair may be determined in which the weight is topN. And determining an initial average of the weights for each industry pair.

4. And saving the industry pairs and the enterprise pairs under the industry pairs.

And separating the enterprise pairs under the target industry pairs from the enterprise node relation graph, and independently storing the enterprise pairs. Or may be to delete the edge between the business pair under the target industry pair directly. For example, if the target industry pair is the industry pair 12, the enterprise pair a→ B, A →d under the industry pair 12 is stored independently, and a→ B, A →d is deleted from the enterprise node relation diagram.

5. And judging the deletion enterprise.

If the ratio of the expenditure weight of one enterprise to the total expenditure weight of the enterprise in the target industrial pair is greater than a preset threshold, all edges of the enterprise serving as expenditure parties in other industrial pairs are removed. For example, since only the node below the industry pair 12 is the enterprise a, it is determined whether the weight of a→ B, A →d is greater than the preset threshold, and if so, other industry pairs, such as the enterprise a→ C, A →e in the industry pair 23, are deleted.

6. And (5) performing iterative processing.

Returning to step 3, the iteration process is continued on other industrial pairs, the weights of the industrial pairs are updated, in this embodiment, only the industrial pair 23 remains after the update, and it can be determined whether the weights of the industrial pair 23 after the update are lower than the initial average value. However, in a practical scenario, the number of enterprises and the number of industry pairs are more, and when the highest-weighted industry pair in a certain iteration process is lower than the initial average value, the iteration process is stopped.

7. And (5) identifying enterprise relations.

According to the obtained enterprise pairs under the target industry pairs in each iteration process, an upstream and downstream map of the enterprise under the same industry chain is constructed, for example, if the industry pair 23 is lower than the initial average value, the enterprise pair in the industry pair 23 does not need to be concerned, and the enterprise pairs A-B, A-D obtained before aggregation are collected to obtain the relationship between the enterprises, namely, the upstream enterprise of the enterprise A is the enterprise B and the enterprise D. The embodiment is simply described by way of example, and the acquired relationship between enterprises is richer under the condition that the number of enterprises and the number of industrial pairs are more.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides an object relationship identification device for realizing the above related object relationship identification method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the device for identifying one or more object relationships provided below may refer to the limitation of the method for identifying an object relationship hereinabove, and will not be described herein.

In an exemplary embodiment, as shown in fig. 5, there is provided an object relationship identifying apparatus, including: a node relation diagram acquisition module 200, a target class determination module 400, an iteration module 600, and an aggregation module 800, wherein:

The node relation diagram obtaining module 200 is configured to obtain an object node relation diagram, where the object node relation diagram includes object nodes and directed edges between object node pairs, and attributes of the directed edges include weights and categories;

A target class determining module 400, configured to determine a target class from the classes based on the weights of the directed edges corresponding to each class;

The iteration module 600 is configured to determine an object relationship between object nodes of directed edges corresponding to the target class, determine the target object node of the target class, delete the directed edge corresponding to the target class from the object node relationship graph, and obtain an updated object node relationship graph, where a starting point of the edge is the directed edge of the target object node, if the target class does not satisfy the iteration end condition;

The aggregation module 800 is configured to aggregate the object relationships between the object nodes determined in each iteration process to obtain a final identified object relationship when the target class meets the iteration end condition.

In one embodiment, the node relation diagram obtaining module is further configured to determine an object to be identified, and obtain a resource transfer relation between the objects to be identified; taking the object to be identified as an object node, obtaining a directed edge between the object node pair based on a resource transfer relationship, and determining the attribute of the directed edge; based on the object nodes, the directed edges, and the attributes of the directed edges, an object node relationship graph is obtained.

In one embodiment, the node relation diagram obtaining module is further configured to obtain first resource transfer data between a type of the object to be identified and the object to be identified; determining the class of the directed edge based on the class of the object to be identified; determining a weight of the directed edge based on the first resource transfer data; and determining the attribute of the directed edge according to the weight and the category.

In one embodiment, the first resource transfer data includes a single transfer resource amount and a total number of resource transfers; the node relation diagram acquisition module is also used for determining the weight of the directed edge in unit time based on the single transfer resource quantity and the total number of resource transfer times in unit time; determining a first weight of the directed edge in a preset first time period and a second weight of the directed edge in a preset second time period based on the unit time weight; and determining the weight of the directed edge according to the first weight and the second weight.

In one embodiment, the target class determining module is further configured to obtain a sum of weights of the directed edges corresponding to each class; and determining the category with the maximum sum of the weights as the target category.

In one embodiment, the iteration end condition includes: the sum of the weights of the directed edges corresponding to the target categories is smaller than the average value of the weights of the directed edges in the initial object node relation graph.

The respective modules in the above-described object relation recognition apparatus may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one exemplary embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 6. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer equipment is used for storing data such as object node relation diagrams and the like. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a method of identifying object relationships.

In one exemplary embodiment, a computer device is provided, which may be a terminal, and an internal structure diagram thereof may be as shown in fig. 7. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program, when executed by a processor, implements a method of identifying object relationships. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structures shown in fig. 6 or 7 are merely block diagrams of portions of structures associated with aspects of the application and are not intended to limit the computer device to which aspects of the application may be applied, and that a particular computer device may include more or fewer components than those shown, or may combine certain components, or may have a different arrangement of components.

In an embodiment, there is also provided a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method embodiments described above when the computer program is executed.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.

In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims

1. A method of identifying an object relationship, the method comprising:

2. The method of claim 1, wherein the obtaining the object node relationship graph comprises:

3. The method of claim 2, wherein the determining the attribute of the directed edge comprises:

4. The method of claim 3, wherein the first resource transfer data comprises a single transfer resource amount and a total number of resource transfers;

5. The method of any one of claims 1 to 4, wherein determining a target class from the classes based on the weight of the directed edge corresponding to each of the classes comprises:

6. The method according to any one of claims 1 to 4, wherein said determining a target object node of said target class comprises:

7. The method of claim 1, wherein the iteration end condition comprises: and the sum of the weights of the directed edges corresponding to the target categories is smaller than the average value of the weights of the directed edges in the initial object node relation diagram.

8. An apparatus for identifying an object relationship, the apparatus comprising:

9. The apparatus of claim 8, wherein the node relationship graph acquisition module is further configured to determine objects to be identified and acquire resource transfer relationships between the objects to be identified; taking the object to be identified as an object node, obtaining a directed edge between object node pairs based on the resource transfer relationship, and determining the attribute of the directed edge; and acquiring an object node relation graph based on the object node, the directed edge and the attribute of the directed edge.

10. The apparatus of claim 9, wherein the node relation graph acquisition module is further configured to acquire first resource transfer data between the type of the object to be identified and the object to be identified; determining the category of the directed edge based on the category of the object to be identified; determining weights of the directed edges based on the first resource transfer data; and determining the attribute of the directed edge according to the weight and the category.

11. The apparatus of claim 10, wherein the first resource transfer data comprises a single transfer resource amount and a total number of resource transfers; the node relation diagram obtaining module is further configured to determine a weight per unit time of the directed edge based on the single transfer resource amount and the total number of resource transfers in unit time; determining a first weight of the directed edge in a preset first time period and a second weight of the directed edge in a preset second time period based on the weight of the unit time; and determining the weight of the directed edge according to the first weight and the second weight.

12. The apparatus according to any one of claims 8 to 11, wherein the target class determining module is further configured to obtain a sum of weights of the directed edges corresponding to the classes; and determining the category with the maximum sum of the weights as a target category.

13. The apparatus according to any one of claims 8 to 11, wherein the iteration module is further configured to obtain second resource transfer data of a candidate node based on the directed edge corresponding to the target class, where the candidate node includes a starting point of the directed edge corresponding to the target class; acquiring third resource transfer data of the candidate node based on the directed edge corresponding to the object node relation diagram; and determining the candidate node with the ratio of the second resource transfer data to the third resource transfer data being greater than a preset threshold value as a target object node.

14. The apparatus of claim 8, wherein the iteration end condition comprises: and the sum of the weights of the directed edges corresponding to the target categories is smaller than the average value of the weights of the directed edges in the initial object node relation diagram.

15. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.

16. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.

17. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.