Knowledge graph-based industrial capability docking technology method

Abstract

The application relates to an industrial capacity docking technology method based on a knowledge graph, and the method comprises the following steps: acquiring a demand, wherein the demand comprises a category, a yield, a deadline, a performance index and a demand flow direction; the demand flow direction is inflow or outflow, wherein the inflow is input from a target enterprise, and the outflow is output to the target enterprise; a subgraph satisfying the category, the yield, the deadline and the performance index in the demand is determined from the knowledge graph; and the target enterprise is determined from the subgraph according to the demand flow direction in the demand; wherein the knowledge graph comprises primary points and secondary points, the primary points correspond to positions of an industrial chain one by one, the secondary points correspond to enterprises one by one, edges between two primary points represent flow relationship of the two primary points in the industrial chain, edges between two secondary points represent business relationship of the two primary points corresponding to the enterprises, the attribute of each secondary point is the attribute of the corresponding enterprise, and the weight of the edge between each secondary point is determined according to the attribute of the secondary point corresponding to the enterprise.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a knowledge graph-based industrial capability matching technology method. Background Technology

[0002] With the development of industrial modernization and the improvement of markets, the number of productive industrial enterprises has surged, and the turnover rate is rapid. This makes it difficult to connect with target enterprises that meet the needs of these numerous industrial enterprises. Summary of the Invention

[0003] (a) Technical problems to be solved

[0004] In view of the above-mentioned shortcomings and deficiencies of the existing technology, the present invention provides an industrial capability docking technology method based on knowledge graph.

[0005] (II) Technical Solution

[0006] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0007] A knowledge graph-based industrial capability matching technology method, the method comprising:

[0008] S101, Obtain demand, the demand includes category, output, time limit, performance indicators, and demand flow direction; the demand flow direction is inflow or outflow, wherein inflow is input from the target enterprise, and outflow is output to the target enterprise;

[0009] S102, determine the subgraph from the knowledge graph that satisfies the category, output, time, and performance indicators of the requirements;

[0010] S103, Based on the demand flow direction in the demand, determine the target enterprise from the sub-graph;

[0011] The knowledge graph includes primary nodes and secondary nodes. Primary nodes correspond one-to-one with the positions in the industrial chain, and secondary nodes correspond one-to-one with enterprises. The edges between two primary nodes represent the flow relationship between the two primary nodes in the industrial chain. The edges between two secondary nodes represent the business relationship between the enterprises corresponding to the two primary nodes. The attributes of each secondary node are the attributes of its corresponding enterprise, and the weight of the edges between each secondary node is determined by the attributes of the enterprise corresponding to the secondary node.

[0012] Optionally, before step S102, the method further includes:

[0013] S201, Obtain the attributes of each enterprise; the attributes include: enterprise attributes, industrial attributes, and service attributes; the enterprise attributes include: enterprise identifier, enterprise position in the industrial chain, enterprise category, and basic enterprise information; the industrial attributes include: product category, number of product lines, and production capacity of each product line; the service attributes include: production cycle and product performance indicators;

[0014] S202: Enterprises are classified according to their position in the industrial chain. Each category corresponds to a primary point in the knowledge graph. The flow relationship between primary points is determined according to the industrial chain. An edge is connected between two primary points with a flow relationship, pointing from the outflowing primary point to the inflowing primary point.

[0015] S203: Each enterprise corresponds to a second-level point in the knowledge graph. An edge is connected between each second-level point and the first-level point corresponding to its enterprise category, pointing from the first-level point to the second-level point. The attributes of each enterprise are used as the attributes of the corresponding second-level point.

[0016] S204: Obtain the business relationships between enterprises, form an edge between the corresponding secondary points of the enterprises based on the business relationships, and make the direction of the edge the same as the business direction. Determine the weight of the edge based on the industrial attributes and service attributes of the two secondary points connected by the edge.

[0017] Optionally, for an edge L connecting secondary point A and secondary point B, and where edge L points from secondary point A to secondary point B, the weight of edge L is DataA / DataB.

[0018] Where DataA = the sum of the first values ​​corresponding to all product categories of the enterprise corresponding to the secondary point A * log the maximum number of product lines corresponding to all product categories;

[0019] DataB = Sum of the second values ​​corresponding to all product categories of the enterprise corresponding to the second-level point B * log the maximum number of product lines corresponding to all product categories;

[0020] The first value corresponding to any product category of the enterprise corresponding to secondary point A = the sum of the production capacity of all product lines corresponding to any product category of the enterprise corresponding to secondary point A / the production cycle of any product category of the enterprise corresponding to secondary point A;

[0021] The first value corresponding to any product category of the enterprise corresponding to the second-level point B is the sum of the production capacity of all product lines corresponding to any product category of the enterprise corresponding to the second-level point B / the production cycle of any product category of the enterprise corresponding to the second-level point B.

[0022] Optionally, S102 specifically includes:

[0023] S102-1, Determine the target first-level points based on the categories in the requirements;

[0024] S102-2, Based on the performance indicators in the requirements, determine the first secondary point among all secondary points connected to the target primary point;

[0025] S102-3, In the first secondary point, the second secondary point is determined based on the output and timeframe in the demand;

[0026] S102-4, A sub-graph that satisfies the requirements is formed based on the second and second-level points.

[0027] Optionally, S102-1 specifically includes:

[0028] Based on the categories in the demand, determine the position of the enterprise corresponding to that category in the industrial chain;

[0029] The primary point corresponding to the position of the enterprise in this category in the industrial chain is determined as the target primary point.

[0030] Optionally, S102-3 specifically includes:

[0031] Determine the number of product lines, the production capacity of the corresponding product lines, and the corresponding production cycle for each first and second level point that corresponds to the category in the demand.

[0032] Select the first or second-level points where the corresponding production cycle is less than or equal to the timeframe in the demand;

[0033] The third value of the selected first and second-level points is calculated as: the sum of the production capacities of all corresponding product lines of the selected first and second-level points * the stable value of the selected first and second-level points.

[0034] The first and second-level points where the third value is greater than or equal to the output value in the demand are taken as the second and second-level points;

[0035] in,

[0036] The stable value of the selected first and second-level points = the number of product lines corresponding to the selected first and second-level points / the fourth value + the fifth value;

[0037] The fourth value = (maximum production capacity of the product line corresponding to the selected first and second-level points + minimum production capacity of the product line corresponding to the selected first and second-level points + 4 * average production capacity of the product line corresponding to the selected first and second-level points) / 6.

[0038] The fifth value = demand duration - production cycle corresponding to the selected first or second level point / demand duration.

[0039] Optionally, S102-4 specifically includes:

[0040] Treat all second-level points as points in the subgraph, and treat all edges between second-level points as edges between points in the subgraph to form a subgraph;

[0041] The attributes of each point in the subgraph are the same as the attributes of that point in the knowledge graph, and the weights of each edge in the subgraph are the same as the weights of that edge in the knowledge graph.

[0042] Furthermore, the in-degree and out-degree of each second-level point in the knowledge graph are used as unique in-degree and unique out-degree attributes of each point in the subgraph.

[0043] Optionally, S103 specifically includes:

[0044] S103-1, determine the out-degree (out) and in-degree (in) of each second-level point;

[0045] S103-2, Determine the unique in-degree and unique out-degree attributes of each second-level point;

[0046] S103-3, determine the target enterprise based on the demand flow direction in the demand, the unique in-degree and out-degree attributes of each second and second level point, out, in.

[0047] Optionally, S103-3 specifically includes:

[0048] Calculate the sixth value of each second-level point = out-degree weight * out + in-degree weight * in;

[0049] The companies corresponding to the second and third level points with the largest sixth value preset threshold are taken as target companies;

[0050] The out-degree weight and in-degree weight are determined based on the demand flow direction, unique in-degree attribute, and unique out-degree attribute in the demand.

[0051] Optionally,

[0052] If the demand flow direction in the demand is input, then the out-degree weight of any second-level point is the difference between the unique out-degree attribute of the second-level point and the out of the second-level point / the unique out-degree attribute of the second-level point, and the in-degree weight is 0.2 * out-degree weight.

[0053] If the demand flow direction in the demand is output, then the out-degree weight is 0.2 * out-degree weight, and the in-degree weight of any second-level point is the difference between the unique in-degree attribute of the second-level point and the in of the second-level point / the unique in-degree attribute of the second-level point.

[0054] (III) Beneficial Effects

[0055] This invention discloses a knowledge graph-based industrial capability matching technology method. It acquires demand, which includes category, output, timeframe, performance indicators, and demand flow direction. Demand flow direction is either inflow or outflow, where inflow refers to input from the target enterprise and outflow refers to output to the target enterprise. The method then determines subgraphs from the knowledge graph that satisfy the category, output, timeframe, and performance indicators of the demand. Based on the demand flow direction, it identifies the target enterprise from the subgraph. The knowledge graph includes primary and secondary nodes, where primary nodes correspond one-to-one with positions in the industrial chain, and secondary nodes correspond one-to-one with enterprises. Edges between two primary nodes represent the flow relationship between them in the industrial chain. Edges between two secondary nodes represent the business relationship between the enterprises corresponding to the two primary nodes. The attributes of each secondary node are the attributes of its corresponding enterprise, and the weights of the edges between secondary nodes are determined by the attributes of the corresponding enterprises, thus achieving accurate matching of current enterprises based on demand. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating an industrial capability docking technology method based on knowledge graphs, provided in an embodiment of the present invention.

[0057] Figure 2 This is a schematic diagram of the structure of a knowledge graph provided in an embodiment of the present invention. Detailed Implementation

[0058] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] With the development of industrial modernization and the improvement of markets, the number of productive industrial enterprises has surged, and the turnover rate is rapid. This makes it difficult to connect with target enterprises that meet the needs of these numerous industrial enterprises.

[0060] Based on this, the present invention provides a knowledge graph-based industrial capability matching technology method. This method includes: acquiring demand, which includes category, output, timeframe, performance indicators, and demand flow direction; the demand flow direction is either inflow or outflow, where inflow is input from a target enterprise and outflow is output to a target enterprise; determining subgraphs from the knowledge graph that satisfy the category, output, timeframe, and performance indicators of the demand; and determining target enterprises from the subgraphs based on the demand flow direction. The knowledge graph includes primary nodes and secondary nodes, where primary nodes correspond one-to-one with positions in the industrial chain, and secondary nodes correspond one-to-one with enterprises. Edges between two primary nodes represent the flow relationship between them in the industrial chain; edges between two secondary nodes represent the business relationship between the enterprises corresponding to the two primary nodes; the attributes of each secondary node are the attributes of its corresponding enterprise, and the weights of the edges between secondary nodes are determined by the attributes of the enterprises corresponding to the secondary nodes, thus achieving accurate matching of current enterprises based on demand.

[0061] See Figure 1 The implementation process of an industrial capability docking technology method based on knowledge graphs in this embodiment is as follows:

[0062] S101, Obtain Requirements.

[0063] Demand includes category, output, timeframe, performance indicators, and demand flow.

[0064] Demand flow can be either inflow or outflow. Inflow refers to input from the target enterprise (e.g., purchasing products from the target enterprise), while outflow refers to output to the target enterprise (e.g., selling products to the target enterprise).

[0065] S102, Identify subgraphs from the knowledge graph that satisfy the categories, output, timeframes, and performance metrics of the requirements.

[0066] The knowledge graph consists of primary nodes and secondary nodes. Primary nodes correspond one-to-one with positions within the industrial chain, while secondary nodes correspond one-to-one with enterprises. Edges between two primary nodes represent the flow relationship between them within the industrial chain. Edges between two secondary nodes represent the business relationship between the enterprises corresponding to those primary nodes. The attributes of each secondary node are the attributes of its corresponding enterprise, and the weights of the edges between secondary nodes are determined by the attributes of the corresponding enterprise.

[0067] In other words, a knowledge graph is constructed before executing S102. The construction process of this knowledge graph is as follows:

[0068] S201, obtain the attributes of each enterprise.

[0069] The attributes include: enterprise attributes, industry attributes, and service attributes.

[0070] Enterprise attributes include: enterprise identifier, enterprise position in the industrial chain (such as upstream, midstream, downstream), enterprise category (such as industry, transportation, service industry, etc.), and basic enterprise information.

[0071] Industrial attributes include: product category (such as bolts, nuts, etc.), number of product lines, and production capacity of each product line.

[0072] Service attributes include: production cycle and product performance indicators.

[0073] S202: Classify enterprises according to their position in the industrial chain. Each category corresponds to a first-level point in the knowledge graph. Determine the flow relationship between each first-level point according to the industrial chain. Connect two first-level points with a flow relationship with an edge pointing from the outflowing first-level point to the inflowing first-level point.

[0074] like Figure 2 As shown, enterprises are classified into upstream, midstream, and downstream, with upstream corresponding to... Figure 2 The first-level point P1 corresponds to the middle-stream point. Figure 2 The first-level point P2 in the middle corresponds to the downstream point. Figure 2 Given a first-level point P3, based on the upstream-midstream-downstream Liu Dong relationship, an edge is formed between P1 and P2 pointing from P1 to P2, and an edge is formed between P2 and P3 pointing from P2 to P3.

[0075] S203 assigns each enterprise to a secondary point in the knowledge graph, connects each secondary point to a primary point corresponding to its enterprise category with an edge pointing from the primary point to the secondary point, and uses the attributes of each enterprise as the attributes of the corresponding secondary point.

[0076] like Figure 2 The white circles represent secondary points, each corresponding to a company. Specifically, p11, p12, and p13 correspond to upstream companies, p21 to a midstream company, and p31 and p32 to downstream companies. Therefore, P1 forms an edge pointing to each of p11, p12, and p13. Similarly, P2 forms an edge pointing to p21. Finally, P3 forms an edge pointing to both p31 and p32.

[0077] S204: Obtain the business relationships between enterprises, form an edge between the corresponding secondary points of the enterprises based on the business relationships, and make the direction of the edge the same as the business direction. Determine the weight of the edge based on the industrial attributes and service attributes of the two secondary points connected by the edge.

[0078] If the company corresponding to p13 sells its products to the company corresponding to p21, then... Figure 2 As shown, an edge is formed between p13 and p21, pointing from p13 to p21.

[0079] The weight of the edge connecting any two secondary points (e.g., the weight of edge L connecting secondary point A and secondary point B) is calculated as follows: if edge L points from secondary point A to secondary point B, then the weight of edge L = DataA / DataB.

[0080] Where DataA = the sum of the first values ​​corresponding to all product categories of the enterprise corresponding to the secondary point A * log the maximum number of product lines corresponding to all product categories.

[0081] DataB = Sum of the second values ​​corresponding to all product categories of the enterprise corresponding to the second-level point B * log the maximum number of product lines corresponding to all product categories.

[0082] The first value corresponding to any product category of the enterprise corresponding to secondary point A is the sum of the production capacity of all product lines corresponding to any product category of the enterprise corresponding to secondary point A / the production cycle of any product category of the enterprise corresponding to secondary point A.

[0083] The first value corresponding to any product category of the enterprise corresponding to the second-level point B is the sum of the production capacity of all product lines corresponding to any product category of the enterprise corresponding to the second-level point B / the production cycle of any product category of the enterprise corresponding to the second-level point B.

[0084] Based on the knowledge graph described above, the implementation process of this step is as follows:

[0085] S102-1, Determine the target first-level point based on the category in the requirements.

[0086] For example: Based on the categories in the demand, determine the position of the enterprise corresponding to that category in the industrial chain. Then, define the primary point corresponding to the position of that enterprise in the industrial chain as the target primary point.

[0087] by Figure 2 For example, if the category is service industry, then the corresponding enterprise in the service industry is in the downstream position in the industrial chain, so P3 is determined as the target first-level point.

[0088] S102-2, Based on the performance indicators in the requirements, determine the first and second level points among all the second level points connected to the target first level point.

[0089] For example, determine that the attribute corresponding to p31 (specifically, the product performance index in the service attribute) meets the performance index in the requirement, and at the same time, determine that the attribute corresponding to p32 (specifically, the product performance index in the service attribute) meets the performance index in the requirement.

[0090] If both p31 and p32 meet the performance requirements, then both p31 and p32 are first- or second-level points.

[0091] If only p31 meets the performance requirements, then only p31 is a first or second-level point.

[0092] If only p33 meets the performance requirements, then only p33 is a first or second-level point.

[0093] If neither p31 nor p32 meets the performance requirements, then there are no first or second-level points, and the connection will fail. In this case, the process will not be executed further, and the user will be notified of the connection failure.

[0094] S102-3, in the first and second level points, the second level point is determined based on the output and timeframe in the demand.

[0095] Specifically, determine the number of product lines, the production capacity of the corresponding product lines, and the corresponding production cycle for each first and second level point that corresponds to the category in the demand.

[0096] Select the first or second-level points where the corresponding production cycle is less than or equal to the timeframe in the demand.

[0097] The third value of the selected first and second-level points is calculated as: the sum of the production capacities of all corresponding product lines of the selected first and second-level points * the stable value of the selected first and second-level points.

[0098] The first and second-level points where the third value is greater than or equal to the output value in the demand are taken as the second and second-level points.

[0099] in,

[0100] The stable value of the selected first and second level points = the number of product lines corresponding to the selected first and second level points / the fourth value + the fifth value.

[0101] The fourth value = (maximum production capacity of the product line corresponding to the selected first and second level points + minimum production capacity of the product line corresponding to the selected first and second level points + 4 * average production capacity of the product line corresponding to the selected first and second level points) / 6.

[0102] The fifth value = demand duration - production cycle corresponding to the selected first or second level point / demand duration.

[0103] If no second or third-level points exist in this step, the connection will fail. In this case, the process will not be executed further, and the user will be notified of the connection failure.

[0104] S102-4, form a subgraph that meets the requirements based on the second and second level points.

[0105] This involves treating all second- and third-level points as points in the subgraph, and treating the edges between all second- and third-level points as edges between points in the subgraph, thus forming the subgraph.

[0106] The attributes of each point in the subgraph are the same as the attributes of that point in the knowledge graph, and the weights of each edge in the subgraph are the same as the weights of that edge in the knowledge graph.

[0107] Furthermore, the in-degree and out-degree of each second-level point in the knowledge graph are used as unique in-degree and out-degree attributes of each point in the subgraph.

[0108] In other words, a subgraph is a subgraph within a knowledge graph.

[0109] S103, Based on the demand flow in the demand, determine the target enterprise from the sub-graph.

[0110] The implementation process of this step is as follows:

[0111] S103-1, determine the out-degree (out) and in-degree (in) of each second-level point.

[0112] S103-2, determine the unique in-degree and unique out-degree attributes of each second-level point.

[0113] S103-3, determine the target enterprise based on the demand flow direction in the demand, the unique in-degree and out-degree attributes of each second and second level point, out, in.

[0114] For example:

[0115] Calculate the sixth value of each second-level point = out-degree weight * out + in-degree weight * in.

[0116] The companies corresponding to the second and third level points with the largest sixth value are selected as target companies.

[0117] The out-degree weight and in-degree weight are determined based on the demand flow direction, unique in-degree attribute, and unique out-degree attribute in the demand.

[0118] If the demand flow direction in the demand is input, then the out-degree weight of any second-level point is the difference between the unique out-degree attribute of any second-level point and the out of any second-level point / the unique out-degree attribute of any second-level point, and the in-degree weight is 0.2 * out-degree weight.

[0119] If the demand flow direction in the demand is output, then the out-degree weight is 0.2 * out-degree weight, and the in-degree weight of any second-level point is the difference between the unique in-degree attribute of any second-level point and the in of any second-level point / the unique in-degree attribute of any second-level point.

[0120] The knowledge graph-based industrial capability matching technology method in this embodiment describes the attributes of each enterprise through a knowledge graph, and then determines the target enterprise based on the relationship between the knowledge graph and the demand, which can achieve accurate matching of current enterprises based on demand.

[0121] The attributes include: enterprise attributes, industrial attributes, and service attributes. Enterprise attributes include: enterprise identifier, enterprise position in the industrial chain, enterprise category, and basic enterprise information. Industrial attributes include: product category, number of product lines, and production capacity of each product line. Service attributes include: production cycle and product performance indicators.

[0122] This embodiment provides a knowledge graph-based industrial capability matching technology method. The method includes: acquiring demand, which includes category, output, timeframe, performance indicators, and demand flow direction; the demand flow direction is either inflow or outflow, where inflow is input from the target enterprise and outflow is output to the target enterprise; determining subgraphs from the knowledge graph that satisfy the category, output, timeframe, and performance indicators of the demand; and determining the target enterprise from the subgraphs based on the demand flow direction. The knowledge graph includes primary nodes and secondary nodes, where primary nodes correspond one-to-one with positions in the industrial chain, and secondary nodes correspond one-to-one with enterprises. Edges between two primary nodes represent the flow relationship between them in the industrial chain; edges between two secondary nodes represent the business relationship between the enterprises corresponding to the two primary nodes; the attributes of each secondary node are the attributes of its corresponding enterprise, and the weights of the edges between secondary nodes are determined by the attributes of the enterprises corresponding to the secondary nodes, thus achieving accurate matching of current enterprises based on demand.

[0123] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0124] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions.

[0126] It should be noted that any reference numerals placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In claims that enumerate several means, several of these means may be embodied by the same hardware. The use of the terms first, second, third, etc., is merely for convenience of expression and does not indicate any order. These terms can be understood as part of the component names.

[0127] Furthermore, it should be noted that in the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0128] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0129] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A knowledge graph-based industrial capability matching technology method, characterized in that, The method includes: S101, Obtain demand, the demand includes category, output, time limit, performance indicators, and demand flow direction; the demand flow direction is inflow or outflow, wherein inflow is input from the target enterprise, and outflow is output to the target enterprise; S102, determine the subgraph from the knowledge graph that satisfies the category, output, time, and performance indicators of the requirements; S103, Based on the demand flow direction in the demand, determine the target enterprise from the sub-graph; The knowledge graph includes primary nodes and secondary nodes. Primary nodes correspond one-to-one with the positions in the industrial chain, and secondary nodes correspond one-to-one with enterprises. The edges between two primary nodes represent the flow relationship between the two primary nodes in the industrial chain. The edges between two secondary nodes represent the business relationship between the enterprises corresponding to the two primary nodes. The attributes of each secondary node are the attributes of its corresponding enterprise, and the weight of the edges between each secondary node is determined by the attributes of the enterprise corresponding to the secondary node. For an edge L connecting secondary point A and secondary point B, and edge L points from secondary point A to secondary point B, then the weight of edge L = DataA / DataB. Where DataA = the sum of the first values ​​corresponding to all product categories of the enterprise corresponding to the second-level point A. log represents the maximum number of product lines corresponding to all product categories; DataB = the sum of the second values ​​corresponding to all product categories of the enterprise corresponding to the second-level point B. log represents the maximum number of product lines corresponding to all product categories; The first value corresponding to any product category of the enterprise corresponding to secondary point A = the sum of the production capacity of all product lines corresponding to any product category of the enterprise corresponding to secondary point A / the production cycle of any product category of the enterprise corresponding to secondary point A; The first value corresponding to any product category of the enterprise corresponding to the second-level point B = the sum of the production capacity of all product lines corresponding to any product category of the enterprise corresponding to the second-level point B / the production cycle of any product category of the enterprise corresponding to the second-level point B; In the knowledge graph, the first-level points include P1 corresponding to the upstream of the industrial chain, P2 corresponding to the midstream of the industrial chain, and P3 corresponding to the downstream of the industrial chain; an edge is formed between P1 and P2, pointing from P1 to P2, and an edge is formed between P2 and P3, pointing from P2 to P3. Secondary points include p11, p12, and p13 corresponding to upstream enterprises, p21 corresponding to midstream enterprises, and p31 and p32 corresponding to downstream enterprises; P1 forms an edge with p11, p12, and p13 respectively, pointing from P1 to p11, p12, and p13; P2 forms an edge with p21 respectively, pointing from P2 to p21; and P3 forms an edge with p31 and p32 respectively, pointing from P3 to p31 and p32. If the company corresponding to p13 sells its product to the company corresponding to p21, then an edge is formed between p13 and p21, pointing from p13 to p21. S102 specifically includes: S102-1, Determine the target first-level points based on the categories in the requirements; S102-2, Based on the performance indicators in the requirements, determine the first secondary point among all secondary points connected to the target primary point; S102-3, In the first secondary point, the second secondary point is determined based on the output and timeframe in the demand; S102-4, A sub-graph that satisfies the requirements is formed based on the second and second-level points; Specifically, S102-3 includes: Determine the number of product lines, the production capacity of the corresponding product lines, and the corresponding production cycle for each first and second level point that corresponds to the category in the demand. Select the first or second-level points where the corresponding production cycle is less than or equal to the timeframe in the demand; The third value of the selected first and second-level points is calculated as the sum of the production capacities of all corresponding product lines of the selected first and second-level points. The stable values ​​of the first and second level points selected; The first and second-level points where the third value is greater than or equal to the output value in the demand are taken as the second and second-level points; in, The stable value of the selected first and second-level points = the number of product lines corresponding to the selected first and second-level points / the fourth value + the fifth value; The fourth value = (maximum production capacity of the product line corresponding to the selected first and second-level points + minimum production capacity of the product line corresponding to the selected first and second-level points + 4) (Average production capacity of the product line corresponding to the selected first and second level points) / 6; Fifth value = (Demand duration - Production cycle corresponding to the selected first and second level points) / Demand duration.

2. The method according to claim 1, characterized in that, Before S102, the following is also included: S201, Obtain the attributes of each enterprise; the attributes include: enterprise attributes, industrial attributes, and service attributes; the enterprise attributes include: enterprise identifier, enterprise position in the industrial chain, enterprise category, and basic enterprise information; the industrial attributes include: product category, number of product lines, and production capacity of each product line; the service attributes include: production cycle and product performance indicators; S202: Enterprises are classified according to their position in the industrial chain. Each category corresponds to a primary point in the knowledge graph. The flow relationship between primary points is determined according to the industrial chain. An edge is connected between two primary points with a flow relationship, pointing from the outflowing primary point to the inflowing primary point. S203: Each enterprise corresponds to a second-level point in the knowledge graph. An edge is connected between each second-level point and the first-level point corresponding to its enterprise category, pointing from the first-level point to the second-level point. The attributes of each enterprise are used as the attributes of the corresponding second-level point. S204: Obtain the business relationships between enterprises, form an edge between the corresponding secondary points of the enterprises based on the business relationships, and make the direction of the edge the same as the business direction. Determine the weight of the edge based on the industrial attributes and service attributes of the two secondary points connected by the edge.

3. The method according to claim 2, characterized in that, S102-1 specifically includes: Based on the categories in the demand, determine the position of the enterprise corresponding to that category in the industrial chain; The primary point corresponding to the position of the enterprise in this category in the industrial chain is determined as the target primary point.

4. The method according to claim 2, characterized in that, Specifically, S102-4 includes: Treat all second-level points as points in the subgraph, and treat all edges between second-level points as edges between points in the subgraph to form a subgraph; The attributes of each point in the subgraph are the same as the attributes of that point in the knowledge graph, and the weights of each edge in the subgraph are the same as the weights of that edge in the knowledge graph. Furthermore, the in-degree and out-degree of each second-level point in the knowledge graph are used as unique in-degree and unique out-degree attributes of each point in the subgraph.

5. The method according to claim 2, characterized in that, S103 specifically includes: S103-1, determine the out-degree (out) and in-degree (in) of each second-level point; S103-2, Determine the unique in-degree and unique out-degree attributes of each second-level point; S103-3, determine the target enterprise based on the demand flow direction in the demand, the unique in-degree and out-degree attributes of each second and second level point, out, in.

6. The method according to claim 2, characterized in that, Specifically, S103-3 includes: Calculate the sixth value of each second- and second-level point = out-degree weight out + in-degree weight in; The companies corresponding to the second and third level points with the largest sixth value preset threshold are taken as target companies; The out-degree weight and in-degree weight are determined based on the demand flow direction, unique in-degree attribute, and unique out-degree attribute in the demand.

7. The method according to claim 6, characterized in that, If the demand flow direction in the demand is input, then the out-degree weight of any second-level node is the difference between the unique out-degree attribute of that second-level node and the out value of that second-level node, divided by the unique out-degree attribute of that second-level node. The in-degree weight is 0.

2. Out-degree weight; If the demand flow direction in the demand is output, then the out-degree weight is 0.

2. The out-degree weight is the difference between the unique in-degree attribute of the second-level point and the in of the second-level point, divided by the unique in-degree attribute of the second-level point.

Images