Industrial chain risk prediction method and device, electronic equipment and storage medium

Abstract

Embodiments of the present application provide an industry chain risk prediction method and device, electronic equipment and storage medium, belonging to the field of artificial intelligence. The method comprises: obtaining product supply relationship data of an industry chain and constructing an initial industry chain relationship graph, the relationship graph comprising initial nodes, and the initial nodes forming simplexes. Grouping each simplex in the initial industry chain relationship graph according to the dimensions of the simplexes to obtain a combination of simplexes, and then constructing an inter-dimension dependency matrix according to the combination of simplexes of each dimension and the combination of adjacent simplexes. Next, a target industry chain relationship graph is obtained according to a preset sample risk event node, an initial node and the inter-dimension dependency matrix. Then, a target risk prediction model is obtained by training a preset initial risk prediction model according to the target industry chain relationship graph. Finally, the target risk prediction model is used to predict the risk of a target risk event. The embodiments of the present application can improve the accuracy of risk prediction of the industry chain.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and is applicable to the financial sector, particularly to a risk prediction method and device, electronic device, and storage medium for the industrial chain. Background Technology

[0002] In related technologies, Graph Convolutional Networks (GCNs) are typically used to learn multi-node network graphs that represent supply chain relationships. After training, a risk prediction model is obtained to assess the potential impact of risk events on different entities in the supply chain (such as raw material suppliers, manufacturers, distributors, and customers).

[0003] In reality, complex industrial chains often involve complex multi-to-many or many-to-one interactions between nodes. However, multi-node network graphs mainly represent simple binary relationships between nodes and edges, which cannot effectively represent the complex interactions within the industrial chain, thus affecting the accuracy of risk prediction.

[0004] Therefore, improving the accuracy of risk prediction in the industrial chain has become an urgent technical problem to be solved. Summary of the Invention

[0005] The main objective of this application is to propose a method, apparatus, electronic device, and storage medium for predicting risks in the industrial chain, aiming to improve the accuracy of risk prediction in the industrial chain.

[0006] To achieve the above objectives, a first aspect of this application proposes a risk prediction method for the industrial chain, the method comprising:

[0007] Obtain product supply relationship data from at least two entities within the target industry chain;

[0008] An initial supply chain relationship diagram is constructed based on the product supply relationship data of each object; wherein, the initial supply chain relationship diagram includes at least two initial nodes, and the at least two initial nodes form at least two simplexes;

[0009] Based on the dimensions of the simplex, the simplexes in the initial industrial chain relationship diagram are grouped to obtain at least two simplex combinations;

[0010] Based on the diffusion relationship constructed from the simplex combination and adjacent simplex combinations in each dimension, the inter-dimensional dependency matrix is ​​obtained; wherein, the adjacent simplex combination is the combination whose dimension is adjacent to the simplex combination.

[0011] Based on the preset sample risk event nodes, the initial nodes, and the inter-dimensional dependency matrix, the sample risk event nodes are embedded into the initial industry chain relationship graph to obtain the target industry chain relationship graph;

[0012] The target risk prediction model is obtained by training the preset initial risk prediction model based on the target industry chain relationship diagram.

[0013] The target risk event is obtained, and the risk of the target risk event is predicted using the target risk prediction model.

[0014] In some embodiments, the adjacent simplex combinations include a first adjacent simplex combination and a second adjacent simplex combination, and the dimension of the first adjacent simplex combination is smaller than the dimension of the second adjacent simplex combination; the step of constructing a diffusion relationship based on the simplex combination and adjacent simplex combinations of each dimension to obtain an inter-dimensional dependency matrix includes:

[0015] Based on the diffusion relationship between the simplex combination and the first adjacent simplex combination, and the diffusion relationship between the simplex combination and the second adjacent simplex combination, matrix calculations are performed to obtain the inter-dimensional dependency submatrix of the simplex combination;

[0016] The inter-dimensional dependency matrix is ​​obtained by fusing the inter-dimensional dependency submatrices of each simplex combination.

[0017] In some embodiments, the step of performing matrix calculations based on the diffusion relationship between the simplex combination and the first adjacent simplex combination, and the diffusion relationship between the simplex combination and the second adjacent simplex combination, to obtain the inter-dimensional dependency submatrix of the simplex combination includes:

[0018] Construct the relation matrix of the simplex combination spreading to the first adjacent simplex combination to obtain the first boundary matrix;

[0019] Construct the relation matrix of the diffusion from the second adjacent simplex combination to the simplex combination to obtain the second boundary matrix;

[0020] Transpose the first boundary matrix to obtain the first common boundary matrix;

[0021] The second boundary matrix is ​​transposed to obtain the second common boundary matrix;

[0022] Based on the first boundary matrix, the first common boundary matrix, the second boundary matrix, and the second common boundary matrix, the inter-dimensional dependency submatrix of the simplex combination is obtained.

[0023] In some embodiments, the step of embedding the sample risk event nodes into the initial industry chain relationship graph based on preset sample risk event nodes, the initial node, and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph includes:

[0024] Calculate the similarity between the sample risk event node and each of the initial nodes to obtain the initial node similarity data;

[0025] Candidate nodes are selected from the initial nodes based on the initial node similarity data, and the sample risk event nodes are connected to the candidate nodes to obtain candidate connection edges;

[0026] The initial propagation node sequence is determined based on the sample risk event nodes, the candidate nodes, and the candidate connection edges;

[0027] The initial industry chain relationship graph is updated based on the initial propagation node sequence and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph.

[0028] In some embodiments, updating the initial industry chain relationship graph based on the initial propagation node sequence and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph includes:

[0029] Select the starting propagation node from the initial propagation nodes in the initial propagation node sequence;

[0030] The neighboring nodes of the initial propagation node are determined based on the inter-dimensional dependency matrix, and the similarity between the neighboring nodes and each initial propagation node is calculated to obtain candidate node similarity data.

[0031] Based on the candidate node similarity data, a target node is selected from the initial propagation nodes, the target node is connected to the neighboring nodes, and the initial propagation node sequence is updated based on the connection edges between the target node and the neighboring nodes to obtain the target propagation node sequence.

[0032] The initial industry chain relationship diagram is updated based on the target propagation node sequence to obtain the target industry chain relationship diagram.

[0033] In some embodiments, the step of training a preset initial risk prediction model based on the target industry chain relationship diagram to obtain a target risk prediction model includes:

[0034] The inter-dimensional dependency matrix is ​​updated based on the target industry chain relationship diagram to obtain the target dependency matrix;

[0035] Based on the target dependency matrix and each node in the target industry chain relationship graph, the node relationships are integrated to obtain the sample graph structure features;

[0036] The parameters of the initial risk prediction model are adjusted based on the structural characteristics of the sample graph to obtain the target risk prediction model.

[0037] In some embodiments, obtaining the target risk event and predicting the risk of the target risk event using the target risk prediction model includes:

[0038] Construct the original risk node based on the target risk event;

[0039] The initial industry chain relationship diagram is updated based on the original risk nodes to obtain the structural features of the target diagram;

[0040] Based on the target risk prediction model, the structural features of the target graph are predicted to obtain the reference risk feature value of each node in the target industrial chain relationship graph.

[0041] Select the target risk node based on the reference risk characteristic value;

[0042] The connection edges associated with the target risk node are filtered based on the degree of dependency between the target risk node and related risk nodes to obtain the target risk edges;

[0043] Based on the target risk node and the target risk edge, risk prediction information is determined.

[0044] To achieve the above objectives, a second aspect of this application provides a risk prediction device for the industrial chain, the device comprising:

[0045] The data acquisition module is used to acquire product supply relationship data of at least two objects in the target industry chain;

[0046] A relationship graph construction module is used to construct an initial industry chain relationship graph based on the product supply relationship data of each object; wherein, the initial industry chain relationship graph includes at least two initial nodes, and the at least two initial nodes form at least two simplexes;

[0047] A grouping module is used to group the simplexes in the initial supply chain relationship diagram according to the dimensions of the simplexes, so as to obtain at least two simplex combinations;

[0048] The matrix construction module is used to construct the diffusion relationship based on the simplex combination and adjacent simplex combinations in each dimension to obtain the inter-dimensional dependency matrix; wherein, the adjacent simplex combination is the combination whose dimension is adjacent to the simplex combination.

[0049] The relationship graph update module is used to embed the sample risk event nodes into the initial industry chain relationship graph according to the preset sample risk event nodes, the initial nodes, and the inter-dimensional dependency matrix, so as to obtain the target industry chain relationship graph.

[0050] The prediction model training module is used to train the preset initial risk prediction model based on the target industrial chain relationship diagram to obtain the target risk prediction model;

[0051] The risk prediction module is used to acquire target risk events and perform risk prediction on the target risk events using the target risk prediction model.

[0052] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0053] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0054] The proposed method, apparatus, electronic device, and storage medium for risk prediction in the industrial chain first construct an initial industrial chain relationship graph, including a simplex structure, by acquiring product supply relationship data of different objects in the target industrial chain. Then, the simplexes in the initial industrial chain relationship graph are grouped according to their dimensions, and the diffusion relationships between simplex combinations of different dimensions are analyzed to generate an inter-dimensional dependency matrix. This inter-dimensional dependency matrix can characterize high-dimensional interaction characteristics, thus overcoming the deficiency of multi-node network graphs in the prior art, which cannot effectively reflect the complex dimensional interactions in complex industrial chains. Subsequently, preset sample risk event nodes are embedded into the graph structure to simulate the propagation characteristics of actual risk events and dynamically update the target industrial chain relationship graph. Then, the initial risk prediction model is trained in conjunction with the target industrial chain relationship graph to obtain a target risk prediction model. Finally, the target risk prediction model is used to predict target risk events, accurately assessing the risk characteristic values ​​and potential propagation paths of each node in the industrial chain. Compared to traditional schemes that predict risks based on multi-node network graphs, the method in this embodiment introduces a simplex structure when constructing the industrial chain relationship, which can reflect the complex dimensional interaction relationships between various nodes in a complex industrial chain. By updating the structure of the initial industry chain relationship graph using the inter-dimensional dependency matrix, the resulting target industry chain relationship graph can reflect the impact of sample risk event nodes on the target industry chain. This enables the target risk prediction model trained based on the target industry chain relationship graph to accurately capture the characteristics of risk propagation, thereby improving the accuracy of risk prediction for the industry chain. Attached Figure Description

[0055] Figure 1 This is a flowchart of the risk prediction method for the industrial chain provided in the embodiments of this application;

[0056] Figure 2 yes Figure 1 The flowchart of step S104 in the process;

[0057] Figure 3 yes Figure 2 The flowchart of step S201 in the text;

[0058] Figure 4 yes Figure 1 The flowchart of step S105 in the process;

[0059] Figure 5 yes Figure 4 The flowchart of step S404 in the document;

[0060] Figure 6 This is a supply chain relationship diagram of the supply chain risk prediction method provided in the embodiments of this application;

[0061] Figure 7 yes Figure 1 The flowchart of step S106 in the process;

[0062] Figure 8 yes Figure 1 The flowchart of step S107 in the process;

[0063] Figure 9 This is a schematic diagram of the structure of the risk prediction device for the industrial chain provided in the embodiments of this application;

[0064] Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0066] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0067] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0068] First, let's analyze some of the terms used in this application:

[0069] Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0070] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, intent recognition, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0071] Simplex: In geometry, a simplex, or n-simplex, is an n-dimensional geometric solid similar to a triangle. More precisely, a simplex is the convex hull of a set of (n+1) affine-independent points in some Euclidean space of n dimensions or higher. For example, a 0-simplex is a point, a 1-simplex is a line segment, a 2-simplex is a triangle, a 3-simplex is a tetrahedron with each face being a triangle, and a 4-simplex is a pentacell.

[0072] Simplex complex: a concept in topology, referring to a topological object obtained by "gluing together" simplexes such as points, line segments, and triangles.

[0073] In related technologies, Graph Convolutional Networks (GCNs) are typically used to learn multi-node network graphs that represent supply chain relationships. After training, a risk prediction model is obtained to assess the potential impact of risk events on different entities in the supply chain (such as raw material suppliers, manufacturers, distributors, and customers).

[0074] In reality, risk propagation in complex supply chains often involves more than just simple nodes and edges (such as the relationship between suppliers and manufacturers). This is because complex supply chains typically involve intricate many-to-many or many-to-one interactions between nodes. For instance, multiple suppliers may simultaneously influence a single manufacturer, and multiple manufacturers may collectively influence a single distributor. Multi-node network graphs, primarily represented by simple binary relationships of nodes and edges, fail to effectively represent the complex interactions within the supply chain. Consequently, graph convolutional networks trained on this graph structure tend to capture one-to-one connections or overall layout features between nodes, failing to effectively reflect the complex collaborations and dependencies among multiple nodes, thus impacting the accuracy of risk prediction.

[0075] Therefore, improving the accuracy of risk prediction in the industrial chain has become an urgent technical problem to be solved.

[0076] Based on this, embodiments of this application provide a method and apparatus for predicting risks in the industrial chain, an electronic device and a storage medium, aiming to improve the accuracy of risk prediction in the industrial chain.

[0077] The risk prediction method, apparatus, electronic device, and storage medium for the industrial chain provided in this application are specifically described through the following embodiments. First, the risk prediction method for the industrial chain in this application is described.

[0078] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0079] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0080] The supply chain risk prediction method provided in this application relates to the field of artificial intelligence technology. This supply chain risk prediction method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the supply chain risk prediction method, but is not limited to the above forms.

[0081] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0082] Figure 1 This is an optional flowchart of the risk prediction method for the industrial chain provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S107.

[0083] Step S101: Obtain product supply relationship data for at least two objects in the target industry chain.

[0084] Step S102: Construct an initial industry chain relationship diagram based on the product supply relationship data of each object.

[0085] Step S103: Group the simplexes in the initial supply chain relationship diagram according to their dimensions to obtain at least two simplex combinations.

[0086] Step S104: Construct diffusion relationships based on the simplex combination of each dimension and adjacent simplex combinations to obtain the inter-dimensional dependency matrix.

[0087] Step S105: Based on the preset sample risk event nodes, initial nodes, and inter-dimensional dependency matrix, embed the sample risk event nodes into the initial industrial chain relationship diagram to obtain the target industrial chain relationship diagram.

[0088] Step S106: Train the preset initial risk prediction model based on the target industrial chain relationship diagram to obtain the target risk prediction model.

[0089] Step S107: Obtain the target risk event and perform risk prediction on the target risk event using the target risk prediction model.

[0090] Steps S101 to S107 of this embodiment first involve acquiring product supply relationship data of different objects in the target industry chain and constructing an initial industry chain relationship graph including a simplex structure. Then, the simplexes in the initial industry chain relationship graph are grouped according to their dimensions, and the diffusion relationships between simplex combinations of different dimensions are analyzed to generate an inter-dimensional dependency matrix. This inter-dimensional dependency matrix can characterize high-dimensional interaction characteristics, thus overcoming the deficiency of multi-node network graphs in the background technology, which cannot effectively reflect complex dimensional interactions in complex industry chains. Subsequently, preset sample risk event nodes are embedded into the graph structure to simulate the propagation characteristics of actual risk events and dynamically update the target industry chain relationship graph. Then, the initial risk prediction model is trained in conjunction with the target industry chain relationship graph to obtain a target risk prediction model. Finally, the target risk prediction model is used to predict target risk events, accurately assessing the risk characteristic values ​​and potential propagation paths of each node in the industry chain. Compared to traditional schemes that predict risks based on multi-node network graphs, the method of this embodiment introduces a simplex structure when constructing the industry chain relationship, which can reflect the complex dimensional interaction relationships between various nodes in a complex industry chain. By updating the structure of the initial industry chain relationship graph using the inter-dimensional dependency matrix, the resulting target industry chain relationship graph can reflect the impact of sample risk event nodes on the target industry chain. This enables the target risk prediction model trained based on the target industry chain relationship graph to accurately capture the characteristics of risk propagation, thereby improving the accuracy of risk prediction for the industry chain.

[0091] In step S101 of some embodiments, the target supply chain refers to the supply chain for which risk analysis needs to be performed. The objects refer to the basic entities involved in production, transportation, or sales within the supply chain, such as raw material suppliers, manufacturers, distributors, and customers, but are not limited to these. Product supply relationship data is used to clarify the upstream and downstream relationships between various objects related to product supply within the target supply chain, and can be obtained from data such as inter-enterprise contract data, logistics data, and transaction records.

[0092] In step S102 of some embodiments, the initial supply chain relationship graph is a graph structure constructed based on the product supply relationship data of each object in the target supply chain. In this embodiment, the initial supply chain relationship graph is a simplex graph, including at least two initial nodes, and the at least two initial nodes form at least two simplexes. It can be understood that each initial node represents an object in the target supply chain. If there is a connecting edge between two objects, the connecting edge has a direction, which can represent the product supply or logistics path between different objects. In some embodiments, the weight data of the connecting edges in the initial supply chain relationship graph is defaulted to 1.

[0093] In step S103 of some embodiments, simplexes with different dimensions (such as nodes, edges, faces, etc.) are classified and grouped according to their dimensions to obtain simplex combinations of different dimensions. It should be noted that the simplex elements in the initial industry chain diagram are not mutually exclusive during the grouping process. That is, all initial nodes in the initial industry chain diagram can be classified into combinations of 0-dimensional simplexes, and all edges can be classified into combinations of 1-dimensional simplexes. Even if the two ends of an edge are classified into combinations of 0-dimensional simplexes, it does not preclude the connecting edges between them from being grouped into combinations of 1-dimensional simplexes. Similarly, even if every edge in a triangle is grouped into a combination of 1-dimensional simplexes, all triangles in the initial industry chain diagram can still be classified into combinations of 2-dimensional simplexes. For example, if the initial industry chain diagram is a structure where nodes A, B, and C are connected end-to-end, then the grouping result is a combination of simplexes with three dimensions. The 0-dimensional simplex combination includes nodes A, B, and C. The 1-dimensional simplex combination includes edges AB, BC, and CA. The 2-dimensional simplex combination includes face ABC.

[0094] In step S104 of some embodiments, it can be understood that adjacent simplex combinations are combinations whose dimensions are adjacent to those of other simplex combinations. Specifically, adjacent simplex combinations include first adjacent simplex combinations and second adjacent simplex combinations, and the dimension of the first adjacent simplex combination is smaller than the dimension of the second adjacent simplex combination. The inter-dimensional dependency matrix is ​​used to describe the boundary relationships between different simplex combinations in the supply chain relationship diagram, and to capture the transformations, connections, and dependencies between simplexes of different dimensions. Therefore, it can be used to describe the interaction relationships between multiple nodes. This matrix can be a Laplace matrix, a normalized Laplace matrix, or a Hodge Laplace matrix, etc., and is not limited to these.

[0095] Please see Figure 2 In some embodiments, step S104 may include, but is not limited to, steps S201 to S202:

[0096] Step S201: Perform matrix calculations based on the diffusion relationship between the simplex combination and the first adjacent simplex combination, and the diffusion relationship between the simplex combination and the second adjacent simplex combination, to obtain the inter-dimensional dependency submatrix of the simplex combination.

[0097] Step S202: The inter-dimensional dependency submatrices of each simplex combination are fused to obtain the inter-dimensional dependency matrix.

[0098] In step S201 of some embodiments, for each simplex combination of dimensions, it is necessary to construct the diffusion relationship between it and two simplex combinations of adjacent dimensions. The diffusion relationship between the simplex combination and the first adjacent simplex combination refers to the diffusion relationship between the current simplex combination and the lower-dimensional simplex combination, and the diffusion relationship between the simplex combination and the second adjacent simplex combination refers to the diffusion relationship between the current simplex combination and the higher-dimensional simplex combination. The dimensions of the first adjacent simplex combination, the simplex combination, and the second adjacent simplex combination increase sequentially, and the increments of dimensions can be equal or different.

[0099] The following examples illustrate the process of constructing the inter-dimensional dependency matrix when using the properties of the Hodge Laplace matrix to describe the diffusion relationship between simplex combinations of different dimensions. Please refer to [link to relevant documentation]. Figure 3 In this embodiment, the dimensions of the first adjacent simplex combination, the simplex combination, and the second adjacent simplex combination increase sequentially, with an increment of 1. Step S201 may include, but is not limited to, steps S301 to S305:

[0100] Step S301: Construct the relation matrix of simplex combination diffusion to the first adjacent simplex combination to obtain the first boundary matrix.

[0101] Step S302: Construct the relation matrix of the diffusion from the second adjacent simplex combination to the simplex combination to obtain the second boundary matrix.

[0102] Step S303: Transpose the first boundary matrix to obtain the first common boundary matrix.

[0103] Step S304: Transpose the second boundary matrix to obtain the second common boundary matrix.

[0104] Step S305: Based on the first boundary matrix, the first common boundary matrix, the second boundary matrix, and the second common boundary matrix, obtain the dimension-dependent submatrix of the simplex combination.

[0105] In step S301 of some embodiments, for example, the dimension of the current simplex combination is described by the symbol k, then the dimension of the first adjacent simplex combination is (k-1), and the first boundary matrix is ​​used to describe the mapping relationship from the k-dimensional simplex to the (k-1)-dimensional simplex.

[0106] In step S302 of some embodiments, for example, the dimension of the current simplex combination is described by the symbol k, then the dimension of the second adjacent simplex combination is (k+1), and the second boundary matrix is ​​used to describe the mapping relationship from the (k+1)-dimensional simplex to the k-dimensional simplex.

[0107] In step S303 of some embodiments, the first common boundary matrix is ​​the transpose of the first boundary matrix, used to describe the mapping relationship from the (k-1)-dimensional simplex to the k-dimensional simplex.

[0108] In step S304 of some embodiments, the second common boundary matrix is ​​the transpose of the second boundary matrix, used to describe the mapping relationship from the k-dimensional simplex to the (k+1)-dimensional simplex.

[0109] In step S305 of some embodiments, the first boundary matrix, the first common boundary matrix, the second boundary matrix, the second common boundary matrix, and the inter-dimensional dependency submatrix can satisfy the following analytical expression:

[0110]

[0111] Here, the dimension of the simplex combination is denoted as k, and the first boundary matrix is ​​denoted as B. k The first common boundary matrix is ​​represented as The second boundary matrix is ​​represented as B. k+1 The second common boundary matrix is ​​represented as The interdimensional dependency submatrix is ​​represented as L k .

[0112] It should be noted that within the same simplex graph, when dimension k represents the lowest and highest dimensions of the simplex, the specific structure of the corresponding inter-dimensional dependency submatrix will discard the boundary matrix corresponding to the non-existent dimension. For example, the initial supply chain relationship graph has 0-dimensional, 1-dimensional, and 2-dimensional simplexes. For the 0-dimensional simplex combination, there is no dimension lower than 0, so k is 0, and therefore the corresponding inter-dimensional dependency submatrix... For a 1-dimensional simplex combination, k is 1, so the corresponding inter-dimensional dependency submatrices... For a 2-dimensional simplex combination, there is no dimension higher than 2, so k is 2. Therefore, the corresponding inter-dimensional dependency submatrices... Furthermore, in calculating L1 and L2, the boundary matrix B2, which maps the 2D simplex combination to the 1D simplex combination, is used, but... and The meanings are different. It operates on a 1-dimensional simplex and is used to describe the connection between edges through triangles, such as how many triangles each edge is contained within. It operates on a 2D simplex and is used to describe the connection between triangles through edges, such as how many edges each triangle shares with itself, or how many common edges connect two triangles.

[0113] Steps S301 to S305 as illustrated in the embodiments of this application accurately describe the topological relationships of multidimensional structures in simplex graphs by precisely constructing the relationship matrix between simplex graphs of different dimensions. This is particularly suitable for risk analysis scenarios in complex industrial chains. In multidimensional data structures, it can capture dependency paths and propagation effects that cannot be described by single-dimensional analysis, providing an efficient tool for analyzing the diffusion behavior and dependency relationships of complex networks, thereby significantly improving the accuracy of the model.

[0114] In step S202 of some embodiments, the inter-dimensional dependency submatrices corresponding to each dimension are fused to obtain an inter-dimensional dependency matrix representing the entire initial industry chain relationship graph. In other embodiments, the inter-dimensional dependency submatrices are independent of each other, and the set of all inter-dimensional dependency submatrices can be used as the inter-dimensional dependency matrix.

[0115] Steps S201 to S202, as illustrated in this embodiment, effectively integrate the diffusion characteristics and dependencies of simplexes with different dimensions in the initial industrial chain relationship diagram by constructing and fusing inter-dimensional dependency sub-matrices. The method of this embodiment provides higher expressive power in topological modeling and lays a solid foundation for risk propagation analysis, feature extraction, and model optimization, significantly improving the system's prediction accuracy and reliability for complex industrial chain risks.

[0116] In step S105 of some embodiments, the sample risk event nodes are embedded into the initial industry chain relationship graph according to the preset sample risk event nodes, initial nodes, and inter-dimensional dependency matrix to obtain the target industry chain relationship graph. Specifically, the sample risk event nodes are the starting points for risk propagation constructed based on risk events. Risk events refer to events that impact or cause chain reactions to the product supply relationship of the target industry chain. Examples include, but are not limited to, adjustments to trade strategies between different regions, logistics delays, price fluctuations, and natural disasters.

[0117] Please see Figure 4 In some embodiments, step S105 may include, but is not limited to, steps S401 to S404:

[0118] Step S401: Calculate the similarity between the sample risk event node and each initial node to obtain the initial node similarity data.

[0119] Step S402: Select candidate nodes from the initial nodes based on the initial node similarity data, and connect the sample risk event nodes with the candidate nodes to obtain candidate connection edges.

[0120] Step S403: Determine the initial propagation node sequence based on the sample risk event nodes, candidate nodes, and candidate connection edges.

[0121] Step S404: Update the initial industry chain relationship graph based on the initial propagation node sequence and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph.

[0122] In step S401 of some embodiments, it should be noted that initial node features are assigned to the initial nodes when constructing the initial supply chain relationship graph. Each initial node feature includes the corresponding object's inventory level, production capacity, geographical location, and the degree of dependence on other objects, and is stored in the form of an array. In other embodiments, the degree of dependence between different objects can be set as the weight data of the connection edges between different nodes. Sample risk event nodes also have preset sample node features. Since the initial supply chain relationship graph includes at least two initial nodes, at least two similarity data can be obtained by calculating the similarity between the sample node features and each initial node feature. The similarity data can be cosine similarity, Euclidean distance, or dot product, etc., and is not limited to these.

[0123] In step S402 of some embodiments, the initial supply chain relationship graph is connected to the sample risk event node. A similarity threshold is set, and the corresponding initial nodes with similarity data greater than the similarity threshold are selected as candidate nodes. It can be understood that the candidate node can represent the initial node that is directly affected after the sample risk event node is added to the initial supply chain relationship graph, and the candidate connection edge used to connect the sample risk event node and the candidate node represents the direction of risk propagation.

[0124] In step S403 of some embodiments, the initial propagation node sequence represents the direct propagation path of the risk event in the target industry chain. For the node sequence, word embedding methods in natural language processing (such as Word2Vec) are used to learn the representation of the nodes, so that each node in the initial industry chain relationship graph is mapped to a low-dimensional vector space, and these vectors represent the structural features of the node in the graph.

[0125] In step S404 of some embodiments, after determining the initial propagation node sequence, the node features in the sequence need to be updated. The initial nodes adjacent to the candidate nodes in the initial industrial chain relationship graph are determined according to the inter-dimensional dependency matrix. Then, based on the updated node features, it is determined whether to add new connection edges. The above steps are repeated. After performing risk propagation simulation on all nodes that may be affected by risk, the resulting industrial chain relationship graph is the target industrial chain relationship graph.

[0126] Steps S401 to S404, as shown in the embodiments of this application, involve progressively screening, constructing, and updating risk propagation relationships to accurately simulate the diffusion path of risk events in the industrial chain. This process adjusts and expands the initial industrial chain relationship diagram, ultimately generating a target industrial chain relationship diagram. This diagram clearly depicts the propagation path of risk events in the industrial chain and their impact on upstream and downstream nodes.

[0127] In some embodiments, the initial propagation node sequence is updated using a random walk. See also... Figure 5 Step S404 may also include, but is not limited to, steps S501 to S504:

[0128] Step S501: Select the starting propagation node from the initial propagation nodes in the initial propagation node sequence.

[0129] Step S502: Determine the neighboring nodes of the initial propagation node based on the inter-dimensional dependency matrix, calculate the similarity between the neighboring nodes and each initial propagation node, and obtain candidate node similarity data.

[0130] Step S503: Select a target node from the initial propagation nodes based on the candidate node similarity data, connect the target node with its neighboring nodes, and update the initial propagation node sequence based on the connection edges between the target node and its neighboring nodes to obtain the target propagation node sequence.

[0131] Step S504: Update the initial industry chain relationship diagram according to the target propagation node sequence to obtain the target industry chain relationship diagram.

[0132] In some embodiments, in step S501, all nodes in the initial propagation node list are initial propagation nodes. In this embodiment, the initial propagation node is selected randomly. For example, please refer to... Figure 6 , Figure 6 This is a feasible industry chain relationship diagram in the embodiments of this application. Figure 6In the diagram, the sample risk event node is denoted as X, and nodes A, B, C, E, D, and F form the structure of the initial industry chain relationship graph. When executing steps S401 to S403 in the above embodiment, it is determined that the similarity data between the sample risk event node X and nodes A and B in the initial industry chain relationship graph exceeds the similarity threshold. Therefore, it is determined that the sample risk event node X will be connected to nodes A and B respectively, and the candidate connection edges are as follows: Figure 6 The dashed edges XA and XB are shown in the diagram. It should be noted that for each new connecting edge, the features of the two nodes connected by the new connecting edge are updated. The two nodes will add an upstream and downstream relationship to themselves. For example, the node feature of node A will add "is the downstream of node X", and the node feature of the sample risk event node X will add "is the upstream of node A".

[0133] At this point, the initial propagation nodes of the initial propagation node sequence are sample risk event node X, node A, and node B. When executing step S501, a node is randomly selected from sample risk event node X, node A, and node B as the initial propagation node.

[0134] In step S502 of some embodiments, combined with Figure 6 In step S501, node A is randomly selected as the starting propagation node. The neighboring nodes are determined as nodes D, E, F, and the sample risk event node X using the inter-dimensional dependency matrix. For node D, the candidate node similarity data between node D and each propagation node in the propagation node sequence is calculated; that is, the similarity between node D and node A, node D and node B, and node D and the sample risk event node X are calculated respectively. The calculation logic for the candidate node similarity of nodes E and F is the same as that for node D, and will not be elaborated here.

[0135] In step S503 of some embodiments, when the similarity data of candidate nodes exceeds a preset similarity threshold, the corresponding node is taken as the target node. For example, as... Figure 6 As shown, if the similarity between node D and node B exceeds the similarity threshold, then node B is taken as the target node, node B is connected to node D, the initial propagation node sequence is updated, and finally the target propagation node sequence is obtained: [sample risk event node X, node A, node D] and [sample risk event node X, node B, node D].

[0136] In step S504 of some embodiments, the node characteristics of relevant nodes are updated according to the target propagation node sequence. Steps S501 to S504 are repeated, and all nodes in the initial industrial chain relationship diagram that may be affected by risk are simulated for risk propagation. The final determined industrial chain relationship diagram is the target industrial chain relationship diagram.

[0137] Steps S501 to S504, as illustrated in this embodiment, involve progressively screening and constructing propagation paths. This process determines the initial propagation node from the initial propagation nodes and expands its influence scope, ultimately dynamically updating the initial industry chain relationship diagram to generate a target industry chain relationship diagram representing risk propagation. This process filters nodes potentially affected by risk based on the inter-dimensional dependency matrix, then precisely filters nodes affected by the chain reaction of risk events through similarity calculations, thereby constructing propagation paths and simulating the gradual expansion of risk impact.

[0138] Please see Figure 7 In some embodiments, step S106 includes, but is not limited to, steps S701 to S703:

[0139] Step S701: Update the inter-dimensional dependency matrix according to the target industry chain relationship diagram to obtain the target dependency matrix.

[0140] Step S702: Integrate the node relationships based on the target dependency matrix and the target industry chain relationship diagram to obtain the sample graph structure features.

[0141] Step S703: Adjust the parameters of the initial risk prediction model according to the structural characteristics of the sample map to obtain the target risk prediction model.

[0142] In step S701 of some embodiments, the structure of the industry chain relationship graph changes with each added connection edge, and the corresponding simplex combination is also adjusted accordingly. Therefore, the inter-dimensional dependency matrix needs to be updated. The updated inter-dimensional dependency matrix is ​​the target dependency matrix. The process of constructing the target dependency matrix is ​​the same as that of constructing the inter-dimensional dependency matrix, and will not be described in detail here.

[0143] In step S702 of some embodiments, unlike the features of nodes, the sample graph structure features in this embodiment are vectors. The sample graph structure features satisfy the following analytical expression:

[0144]

[0145] Where i and j represent the node sequence, f i f represents the node characteristics of node i. j A represents the node characteristics of node j. ijThe elements of the adjacency matrix represent whether there is a connecting edge between node i and node j. If there is a connecting edge, A... ij A is 1 if there are no connecting edges. ij =0. L represents the target dependency matrix, f represents the sample graph structural features, f Τ Let f be the transpose of f.

[0146] In step S703 of some embodiments, the risk characteristic values ​​of all nodes in the target supply chain graph are initialized, while the risk characteristic values ​​of the sample risk event nodes are set to higher values. For example, the risk characteristic values ​​of the sample risk event nodes are set to be greater than 0, while the risk characteristic values ​​of all nodes except the sample risk event nodes are set to 0. The initial risk prediction model propagates the high-risk characteristic values ​​of the sample risk event nodes in the target supply chain graph according to the structural characteristics of the sample graph, updating the risk characteristic values ​​of other nodes.

[0147] A risk characteristic threshold is set; in this embodiment, the risk characteristic threshold can be 0. If the risk characteristic value of a node is greater than the risk characteristic threshold, the node is determined to be a high-risk node. If the risk characteristic value of a node is less than or equal to the risk characteristic threshold, the node is determined to be a low-risk node.

[0148] The risk feature values ​​of high-risk nodes obtained from the initial risk prediction model are compared with the true risk feature values ​​of the corresponding nodes using a preset loss function to obtain a loss value. The model parameters of the initial risk prediction model are then adjusted based on this loss value to reduce the loss. In this embodiment, the initial risk prediction model can be a graph convolutional network, and the loss value is reduced by adjusting the weight matrix of each layer during training. The steps of predicting risk nodes, calculating loss values, and adjusting parameters are repeated until the loss value is reduced to a preset value, or the number of repetitions reaches a preset maximum number of iterations. At this point, model training ends, and the target risk prediction model is obtained.

[0149] Steps S701 to S703, as illustrated in this embodiment, update the dependencies between dimensions in a timely manner to ensure that the target dependency matrix accurately reflects the latest topological changes in the industry chain. Then, sample graph structural features are constructed, fully integrating node features and adjacency information to vectorize the complex graph structural characteristics. By initializing node risk feature values ​​and propagating high-risk feature values ​​of sample risk event nodes, the diffusion path of risk in the industry chain is gradually simulated. Furthermore, by continuously adjusting model parameters to reduce loss values, the model training can dynamically adapt to structural changes in the industry chain, improving the accuracy of model predictions.

[0150] Please see Figure 8 In some embodiments, step S107 may include, but is not limited to, steps S801 to S806:

[0151] Step S801: Construct the original risk node based on the target risk event.

[0152] Step S802: Update the initial industrial chain relationship diagram based on the original risk nodes to obtain the structural features of the target diagram.

[0153] Step S803: Based on the target risk prediction model, perform risk prediction on the structural features of the target graph to obtain the reference risk characteristic value of each node in the target industrial chain relationship graph.

[0154] Step S804: Select the target risk node based on the reference risk characteristic value.

[0155] Step S805: Based on the degree of dependency between the target risk node and related risk nodes, filter the connection edges associated with the target risk node to obtain the target risk edge.

[0156] Step S806: Determine risk prediction information based on the target risk node and the target risk edge.

[0157] In step S801 of some embodiments, the target risk event refers to a risk event that requires analysis of the potential impact on the target industrial chain in the actual application scenario of risk prediction. The target risk event may include natural disasters, market fluctuations, supply strategies, technical failures, etc. In some embodiments, based on the nature of the target risk event, an object node with the same nature is constructed as the original risk node. For example, if the target risk event is that an object in the target industrial chain is unable to purchase or produce a certain raw material, and this event ultimately leads to the interruption of the supply of the corresponding raw material, then a node representing the bankrupt supplier is constructed as the original risk node. Assuming that the value range of the feature value is [0, 1], the node feature of the original risk node may be [inventory level: 0, production capacity: 0, downstream node dependence on this node: 1].

[0158] In step S802 of some embodiments, the original risk node is embedded into the initial industrial chain relationship diagram to obtain a new industrial chain relationship diagram. The process of embedding the node into the industrial chain relationship diagram has been explained in detail in the embodiment of step S105, and will not be repeated here. Then, the structural features of the target diagram are calculated according to the analytical formula (2).

[0159] In step S803 of some embodiments, the original risk node is preset with an original risk characteristic value. The target risk prediction model simulates the propagation of the original risk characteristic value in the industrial chain to obtain the risk characteristic value of each node after the occurrence of the target risk event. This risk characteristic value is a reference risk characteristic value.

[0160] In step S804 of some embodiments, a target risk node refers to a node that is predicted to have a high risk. A risk feature threshold is set. If the reference risk feature value of a node is greater than the risk feature threshold, the node is determined to be a target risk node. If the reference risk feature value of a node is less than or equal to the risk feature threshold, the node is determined to be a low-risk node.

[0161] In step S805 of some embodiments, the associated risk node refers to a node that has a connection relationship with the target risk node. Since each node's features record its own upstream and downstream relationships and the degree of dependency with upstream and downstream nodes, the degree of dependency is compared with a preset dependency feature threshold, and the connection edge with a dependency degree greater than the threshold is taken as the target risk edge. Figure 6 For example, if node A is determined to be the target risk node, then the associated risk nodes are node D, node E, node F, and risk event node X. The node characteristics of node A are [downstream of node X, degree of dependence with node X; upstream of node D, degree of dependence with node D; upstream of node E, degree of dependence with node E; upstream of node F, degree of dependence with node F], and the degree of dependence with node D is greater than the preset dependence characteristic threshold. Therefore, the connecting edge AD is determined to be the target risk edge.

[0162] In step S806 of some embodiments, risk prediction information is generated by comprehensively considering the object information represented by the target risk node and the event information represented by the target risk edge, for reference by risk analysts. For example, when the target risk event is a restriction on the export of a certain type of chip from a different location, local suppliers cannot provide that type of chip to manufacturers. Some manufacturers, lacking that type of chip and with no other suitable chip available on the market, are unable to produce the corresponding electronic products, leading to the removal of those products from shelves and ultimately preventing customers from purchasing them.

[0163] Steps S801 to S806, as illustrated in this embodiment, involve constructing original risk nodes, updating the initial industry chain relationship graph, and obtaining the structural features of the target graph. Using a target risk prediction model, the reference risk characteristic value of each node in the industry chain can be predicted. Next, target risk nodes are selected based on the reference risk characteristic values, and the dependence of these nodes on other nodes is assessed to filter out target risk edges. This helps identify critical paths and vulnerable links in the industry chain. Finally, by integrating the information from target risk nodes and target risk edges, risk prediction information is determined, providing decision-makers with a clear view of industry chain risks, improving the accuracy of risk assessment, and enhancing the understanding of risk propagation paths in complex industry chains, thereby helping to protect the industry chain from potential risks.

[0164] Please see Figure 9This application also provides a risk prediction device for the industrial chain, which can implement the above-mentioned risk prediction method for the industrial chain. The device includes:

[0165] Data acquisition module 901 is used to acquire product supply relationship data of at least two objects in the target industry chain;

[0166] The relationship graph construction module 902 is used to construct an initial industry chain relationship graph based on the product supply relationship data of each object; wherein the initial industry chain relationship graph includes at least two initial nodes, and the at least two initial nodes form at least two simplexes.

[0167] Grouping module 903 is used to group the simplexes in the initial supply chain relationship diagram according to the simplex dimension to obtain at least two simplex combinations.

[0168] The matrix construction module 904 is used to construct the diffusion relationship based on the simplex combination and adjacent simplex combinations of each dimension to obtain the inter-dimensional dependency matrix; wherein, the adjacent simplex combination is the combination whose dimension is adjacent to the simplex combination.

[0169] The relationship graph update module 905 is used to embed the sample risk event nodes into the initial industry chain relationship graph according to the preset sample risk event nodes, initial nodes, and inter-dimensional dependency matrix to obtain the target industry chain relationship graph.

[0170] The prediction model training module 906 is used to train the preset initial risk prediction model based on the target industrial chain relationship diagram to obtain the target risk prediction model.

[0171] The risk prediction module 907 is used to acquire target risk events and perform risk prediction on the target risk events through the target risk prediction model.

[0172] The specific implementation method of the risk prediction device for this industrial chain is basically the same as the specific implementation method of the risk prediction method for the above-mentioned industrial chain, and will not be described again here.

[0173] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned risk prediction method for the industry chain. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0174] Please see Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0175] The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0176] The memory 1002 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1002 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called and executed by the processor 1001 using the supply chain risk prediction method of the embodiments of this application.

[0177] Input / output interface 1003 is used to implement information input and output;

[0178] The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0179] Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004);

[0180] The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0181] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned risk prediction method for the industrial chain.

[0182] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0183] The supply chain risk prediction method, device, electronic device, and storage medium provided in this application first construct an initial supply chain relationship graph including a simplex structure by acquiring product supply relationship data of different objects in the target supply chain. Then, the simplexes in the initial supply chain relationship graph are grouped according to their dimensions, and the diffusion relationships between simplex combinations of different dimensions are analyzed to generate an inter-dimensional dependency matrix. This inter-dimensional dependency matrix can characterize high-dimensional interaction characteristics, thus overcoming the deficiency of multi-node network graphs in the background technology, which cannot effectively reflect the complex dimensional interactions in complex supply chains. Subsequently, preset sample risk event nodes are embedded into the graph structure to simulate the propagation characteristics of actual risk events and dynamically update the target supply chain relationship graph. Then, the initial risk prediction model is trained in conjunction with the target supply chain relationship graph to obtain a target risk prediction model. Finally, the target risk prediction model is used to predict target risk events, accurately assessing the risk characteristic values ​​and potential propagation paths of each node in the supply chain. Compared to traditional schemes that predict risks based on multi-node network graphs, the method in this embodiment introduces a simplex structure when constructing the supply chain relationship, which can reflect the complex dimensional interaction relationships between various nodes in a complex supply chain. By updating the structure of the initial industry chain relationship graph using the inter-dimensional dependency matrix, the resulting target industry chain relationship graph can reflect the impact of sample risk event nodes on the target industry chain. This enables the target risk prediction model trained based on the target industry chain relationship graph to accurately capture the characteristics of risk propagation, thereby improving the accuracy of risk prediction for the industry chain.

[0184] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0185] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0186] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0187] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0188] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0189] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0190] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0191] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0192] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0193] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0194] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for predicting risks of an industrial chain, characterized in that, The method includes: Obtain product supply relationship data from at least two entities within the target industry chain; An initial supply chain relationship diagram is constructed based on the product supply relationship data of each object; wherein, the initial supply chain relationship diagram includes at least two initial nodes, and the at least two initial nodes form at least two simplexes; Based on the dimensions of the simplex, the simplexes in the initial industrial chain relationship diagram are grouped to obtain at least two simplex combinations; Based on the diffusion relationship constructed from the simplex combination and adjacent simplex combinations in each dimension, the inter-dimensional dependency matrix is ​​obtained; wherein, the adjacent simplex combination is the combination whose dimension is adjacent to the simplex combination. Based on the preset sample risk event nodes, the initial nodes, and the inter-dimensional dependency matrix, the sample risk event nodes are embedded into the initial industry chain relationship graph to obtain the target industry chain relationship graph; The target risk prediction model is obtained by training the preset initial risk prediction model based on the target industry chain relationship diagram. The target risk event is acquired, and the risk of the target risk event is predicted using the target risk prediction model. The step of embedding the sample risk event nodes into the initial industry chain relationship graph based on the preset sample risk event nodes, the initial node, and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph includes: Calculate the similarity between the sample risk event node and each of the initial nodes to obtain the initial node similarity data; Candidate nodes are selected from the initial nodes based on the initial node similarity data, and the sample risk event nodes are connected to the candidate nodes to obtain candidate connection edges; The initial propagation node sequence is determined based on the sample risk event nodes, the candidate nodes, and the candidate connection edges; The initial industry chain relationship graph is updated based on the initial propagation node sequence and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph.

2. The method of claim 1, wherein, The adjacent simplex combinations include a first adjacent simplex combination and a second adjacent simplex combination, wherein the dimension of the first adjacent simplex combination is smaller than the dimension of the second adjacent simplex combination; the step of constructing a diffusion relationship based on the simplex combination and adjacent simplex combinations of each dimension to obtain an inter-dimensional dependency matrix includes: Based on the diffusion relationship between the simplex combination and the first adjacent simplex combination, and the diffusion relationship between the simplex combination and the second adjacent simplex combination, matrix calculations are performed to obtain the inter-dimensional dependency submatrix of the simplex combination; The inter-dimensional dependency matrix is ​​obtained by fusing the inter-dimensional dependency submatrices of each simplex combination.

3. The method of claim 2, wherein, The step of performing matrix calculations based on the diffusion relationships between the simplex combination and the first adjacent simplex combination, and between the simplex combination and the second adjacent simplex combination, to obtain the inter-dimensional dependency submatrix of the simplex combination, includes: Construct the relation matrix of the simplex combination spreading to the first adjacent simplex combination to obtain the first boundary matrix; Construct the relation matrix of the diffusion from the second adjacent simplex combination to the simplex combination to obtain the second boundary matrix; Transpose the first boundary matrix to obtain the first common boundary matrix; The second boundary matrix is ​​transposed to obtain the second common boundary matrix; Based on the first boundary matrix, the first common boundary matrix, the second boundary matrix, and the second common boundary matrix, the inter-dimensional dependency submatrix of the simplex combination is obtained.

4. The method of claim 1, wherein, The step of updating the initial industry chain relationship graph based on the initial propagation node sequence and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph includes: Select the starting propagation node from the initial propagation nodes in the initial propagation node sequence; The neighboring nodes of the initial propagation node are determined based on the inter-dimensional dependency matrix, and the similarity between the neighboring nodes and each initial propagation node is calculated to obtain candidate node similarity data. Based on the candidate node similarity data, a target node is selected from the initial propagation nodes, the target node is connected to the neighboring nodes, and the initial propagation node sequence is updated based on the connection edges between the target node and the neighboring nodes to obtain the target propagation node sequence. The initial industry chain relationship diagram is updated based on the target propagation node sequence to obtain the target industry chain relationship diagram.

5. The method of claim 1, wherein, The step of training a preset initial risk prediction model based on the target industry chain relationship diagram to obtain a target risk prediction model includes: The inter-dimensional dependency matrix is ​​updated based on the target industry chain relationship diagram to obtain the target dependency matrix; Based on the target dependency matrix and each node in the target industry chain relationship graph, the node relationships are integrated to obtain the sample graph structure features; The parameters of the initial risk prediction model are adjusted based on the structural characteristics of the sample graph to obtain the target risk prediction model.

6. The method according to any one of claims 1 to 5, characterized in that, The process of acquiring the target risk event and predicting the risk of the target risk event using the target risk prediction model includes: Construct the original risk node based on the target risk event; The initial industry chain relationship diagram is updated based on the original risk nodes to obtain the structural features of the target diagram; Based on the target risk prediction model, the structural features of the target graph are predicted to obtain the reference risk feature value of each node in the target industrial chain relationship graph. Select the target risk node based on the reference risk characteristic value; The connection edges associated with the target risk node are filtered based on the degree of dependency between the target risk node and related risk nodes to obtain the target risk edges; Based on the target risk node and the target risk edge, risk prediction information is determined. 7.A risk prediction device of an industry chain, characterized by comprising: The device includes: The data acquisition module is used to acquire product supply relationship data of at least two objects in the target industry chain; A relationship graph construction module is used to construct an initial industry chain relationship graph based on the product supply relationship data of each object; wherein, the initial industry chain relationship graph includes at least two initial nodes, and the at least two initial nodes form at least two simplexes; A grouping module is used to group the simplexes in the initial supply chain relationship diagram according to the dimensions of the simplexes, so as to obtain at least two simplex combinations; The matrix construction module is used to construct the diffusion relationship based on the simplex combination and adjacent simplex combinations in each dimension to obtain the inter-dimensional dependency matrix; wherein, the adjacent simplex combination is the combination whose dimension is adjacent to the simplex combination. The relationship graph update module is used to embed the sample risk event nodes into the initial industry chain relationship graph according to the preset sample risk event nodes, the initial nodes, and the inter-dimensional dependency matrix, so as to obtain the target industry chain relationship graph. The prediction model training module is used to train the preset initial risk prediction model based on the target industrial chain relationship diagram to obtain the target risk prediction model; The risk prediction module is used to acquire target risk events and perform risk prediction on the target risk events using the target risk prediction model. The step of embedding the sample risk event nodes into the initial industry chain relationship graph based on the preset sample risk event nodes, the initial node, and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph includes: Calculate the similarity between the sample risk event node and each of the initial nodes to obtain the initial node similarity data; Candidate nodes are selected from the initial nodes based on the initial node similarity data, and the sample risk event nodes are connected to the candidate nodes to obtain candidate connection edges; The initial propagation node sequence is determined based on the sample risk event nodes, the candidate nodes, and the candidate connection edges; The initial industry chain relationship graph is updated based on the initial propagation node sequence and the inter-dimensional dependency matrix to obtain the target industry chain relationship graph.

8. An electronic device, comprising: The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

Images