Associated data processing method and device based on knowledge graph, equipment and medium

Abstract

The invention relates to the field of data request processing and the field of financial science and technology, and discloses a knowledge graph-based associated data processing method, which comprises the following steps of: calculating a first information score of a target node according to first attribute information of the target node; determining a plurality of association nodes associated with the target node, and calculating an association weight between the target node and each association node; calculating a second information score value and a score level corresponding to each associated node, and determining a node state corresponding to each associated node in a preset knowledge graph, the node state including a susceptible node, an infected node and an immune node; and substituting the association weight and the score level into an association data calculation differential equation for calculation to obtain a third information score value of the target node, and generating an association data mining report according to the first information score value and the third information score value. According to the method, the association relationship among the nodes in the knowledge graph is calculated, so that the accuracy of associated data mining is improved.

Description

Technical Field

[0001] The present invention relates to the field of data request processing and financial technology, and in particular to a method, device, electronic device and storage medium for processing associated data based on a knowledge graph. Background Art

[0002] Data mining is a highly automated process of analyzing large amounts of data in a decision support process based on artificial intelligence (such as machine learning) and pattern recognition technologies to mine potential related data. Currently, data mining has been widely used in business operations, transportation, medical health, market analysis, engineering design and other fields.

[0003] For example, in the enterprise field (such as financial institutions, etc.), structured data tables are usually used to represent the associations between related data in order to mine and evaluate business risks (i.e., to make risk predictions). However, in many cases, some complex related data that is difficult to clearly express using structured data tables may have an impact on business risks.

[0004] For example, in the field of transportation, data mining can be used for traffic flow prediction, route optimization, and accident prevention. However, the complex interactions between various dynamic factors or risk events in the transportation network (such as weather conditions, emergencies, driver behavior, etc.) are difficult to represent only through simple table structure data, which makes it more difficult to accurately predict traffic conditions.

[0005] Therefore, how to improve the accuracy of association data mining is a technical problem that needs to be solved urgently. Summary of the invention

[0006] In view of the above content, it is necessary to provide a method for processing associated data based on knowledge graphs. Its purpose is to utilize the capabilities of knowledge graphs in information expression, complex network construction and intelligent reasoning, calculate the association relationships between nodes in the knowledge graph, and improve the accuracy of associated data mining.

[0007] In a first aspect, a method for processing associated data based on a knowledge graph is provided, comprising:

[0008] After receiving a user's request to calculate associated data for a target node in a preset knowledge graph, first attribute information corresponding to the target node is obtained from a preset database, and a first information score value of the target node is calculated according to the first attribute information, where the first attribute information is basic data associated with the target node;

[0009] Acquire second attribute information corresponding to the target node from the preset database, determine multiple associated nodes associated with the target node in the preset knowledge graph according to the second attribute information, and calculate the association weight between the target node and each associated node according to the second attribute information, wherein the second attribute information is relationship data between the target node and each associated node;

[0010] Obtaining third attribute information corresponding to each associated node from the preset database, calculating the second information score value corresponding to each associated node and the score level of each associated node according to the third attribute information of each associated node, and determining the node state corresponding to each associated node in the preset knowledge graph according to the second information score value corresponding to each associated node, wherein the node state includes a susceptible node, an infected node, and an immune node, and the third attribute information is basic data associated with each associated node;

[0011] Obtain an association data calculation differential equation constructed in advance based on the susceptible node, the infected node and the immune node, substitute the association weight and the scoring level of each association node into the association data calculation differential equation for calculation, and obtain the third information scoring value of the target node; generate an association data mining report of the target node according to the first information scoring value and the third information scoring value and feed it back to the user.

[0012] In a second aspect, a knowledge graph-based associated data processing device is provided, comprising:

[0013] A first calculation module is used to obtain first attribute information corresponding to a target node in a preset knowledge graph from a preset database after receiving a user's request to calculate associated data for the target node, and calculate a first information score value of the target node according to the first attribute information, where the first attribute information is basic data associated with the target node;

[0014] A second calculation module is used to obtain second attribute information corresponding to the target node from the preset database, determine multiple associated nodes associated with the target node in the preset knowledge graph according to the second attribute information, and calculate the association weight between the target node and each associated node according to the second attribute information, wherein the second attribute information is the relationship data between the target node and each associated node;

[0015] A determination module, used to obtain the third attribute information corresponding to each associated node from the preset database, calculate the second information score value corresponding to each associated node and the score level of each associated node according to the third attribute information of each associated node, and determine the node state corresponding to each associated node in the preset knowledge graph according to the second information score value corresponding to each associated node, wherein the node state includes a susceptible node, an infected node, and an immune node, and the third attribute information is basic data associated with each associated node;

[0016] A generation module is used to obtain an association data calculation differential equation constructed in advance based on the susceptible node, the infected node and the immune node, substitute the association weight and the scoring level of each association node into the association data calculation differential equation for calculation, obtain the third information scoring value of the target node, generate an association data mining report of the target node according to the first information scoring value and the third information scoring value, and feed it back to the user.

[0017] In a third aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for processing associated data based on the knowledge graph when executing the computer program.

[0018] In a fourth aspect, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned method for processing associated data based on the knowledge graph are implemented.

[0019] Compared with the prior art, the present invention extracts the first attribute information of the target node from the database to score the target node to obtain the first information score value, determines the association weights between other nodes associated with the target node according to the second attribute information, uses the third attribute information of each associated node to calculate the second information score value of each associated node, and classifies the associated nodes into susceptible, infected or immune states, substitutes the association weights and the score levels of each associated node into the associated data calculation differential equation to obtain the first information score value of the target node, and generates an associated data mining report of the target node according to the first information score value and the third information score value. The present invention calculates the association relationship between each node in the knowledge graph through the knowledge graph's capabilities in information expression, complex network construction and intelligent reasoning and the associated data calculation differential equation constructed with "susceptible-infected-immune", thereby improving the accuracy of associated data mining. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of an application environment of a method for processing associated data based on a knowledge graph in one embodiment of the present invention;

[0021] Figure 2 A schematic diagram of a process flow of a method for processing associated data based on a knowledge graph provided in one embodiment of the present invention;

[0022] Figure 3 A schematic diagram of a module of a knowledge graph-based associated data processing device provided by an embodiment of the present invention;

[0023] Figure 4 A schematic diagram of the structure of an electronic device for implementing a method for processing associated data based on a knowledge graph provided by an embodiment of the present invention;

[0024] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] It should be noted that the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0027] The method for processing associated data of the basic knowledge graph provided by the embodiment of the present invention can be applied in the following aspects: Figure 1In an application environment, the client communicates with the server through a network. After the server receives a user's request for calculating associated data for a target node in a preset knowledge graph through the client, it obtains first attribute information corresponding to the target node from a preset database, calculates a first information score value of the target node based on the first attribute information, and the first attribute information is basic data associated with the target node; obtains second attribute information corresponding to the target node from the preset database, determines multiple associated nodes associated with the target node in the preset knowledge graph based on the second attribute information, calculates the association weight between the target node and each associated node based on the second attribute information, and the second attribute information is the relationship data between the target node and each associated node; obtains the corresponding The third attribute information, according to the third attribute information of each associated node, calculates the second information score value corresponding to each associated node and the score level of each associated node, and according to the second information score value corresponding to each associated node, determines the node state corresponding to each associated node in the preset knowledge graph, and the node state includes susceptible nodes, infected nodes and immune nodes; obtains the associated data calculation differential equation constructed in advance according to the susceptible nodes, infected nodes and immune nodes, substitutes the associated weight and the score level of each associated node into the associated data calculation differential equation for calculation, obtains the third information score value of the target node, and generates the associated data mining report of the target node according to the first information score value and the third information score value to respond to the request. The present invention is aimed at financial enterprises, transportation, medical health, market analysis and engineering design, etc., and uses the knowledge graph in information expression, complex network construction and intelligent reasoning. The ability to calculate the association relationship between each node in the knowledge graph is used to improve the accuracy of associated data mining. Among them, the client can be but not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server can be implemented with an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0028] Reference Figure 2 FIG. 1 is a flow chart of a method for processing associated data based on a knowledge graph according to an embodiment of the present invention. The method is executed by an electronic device.

[0029] In this embodiment, the associated data processing method based on the knowledge graph includes:

[0030] S1. After receiving a user's request to calculate associated data for a target node in a preset knowledge graph, obtain first attribute information corresponding to the target node from a preset database, and calculate a first information score value of the target node based on the first attribute information, where the first attribute information is basic data associated with the target node.

[0031] In this embodiment, after receiving a request from a user or a development program to calculate associated data for a target node, the identification information of the target node contained in the analysis request is analyzed, and based on the identification information of the target node, the first attribute information of the target node is extracted from a preset database or other information source platform, and the first attribute information is basic data associated with the target node.

[0032] If the target node is a target node corresponding to a target enterprise in the enterprise field, the basic data associated with the target node includes news and public opinion, financial statements, operating information, and litigation records of the target enterprise.

[0033] If the target node is a target node corresponding to a target road in the field of transportation, the basic data associated with the target node includes the daily traffic volume, design capacity, and accident frequency of the target road.

[0034] Correlation data calculation refers to the process of estimating the possibility of a certain risk event occurring in the future through quantitative analysis.

[0035] According to each keyword of the first attribute information, features corresponding to each basic data are extracted from the basic data associated with the target node, the extracted features are input into a preset scoring prediction model, and each feature is calculated separately using multiple tree functions of the preset scoring prediction model to generate multiple leaf function values, and the first information scoring value of the target node is generated based on the multiple leaf function values.

[0036] For example:

[0037] Assume that the risk management department of a commercial bank receives a request to perform associated data calculation on a target node (target enterprise) called "Technology Company A" and obtains the first attribute information of "Technology Company A" from a preset database, for example, obtains positive and negative news about "Technology Company A" from news media, social media, industry reports and other channels. Positive news: Technology Company A recently received funding for a major scientific research project from the government. Negative news: A product of Technology Company A was recalled due to quality issues.

[0038] For example, extracted from the annual reports, quarterly reports and other financial statements submitted by A Technology Company to regulatory agencies, such as total assets: 800 million yuan, total liabilities: 300 million yuan, current ratio: 1.8, quick ratio: 1.2, net profit margin: 12%.

[0039] The features of news and public opinion, financial statements, operating information, and litigation records are extracted from the first attribute information, and all the extracted features are input into the pre-trained scoring prediction model for calculation. The model calculates that the first information score of "A Technology Company" is 0.35 (assuming that the model output range is 0 to 1, 0 indicates no risk level, and 1 indicates an extremely high risk level). According to the calculation results, the first information score of "A Technology Company" is 0.35, which is a medium-low score level. This means that "A Technology Company" has certain risks under the current circumstances, but they are generally controllable.

[0040] In one embodiment, calculating the first information score of the target node according to the first attribute information includes:

[0041] Extracting features corresponding to each basic data from basic data associated with the target node according to each keyword of the first attribute information;

[0042] Inputting the extracted features into a preset score prediction model, wherein the preset score prediction model is generated by an extreme gradient boosting decision tree model;

[0043] Utilizing the plurality of tree functions of the preset scoring prediction model to calculate each feature respectively, and generating a plurality of leaf function values;

[0044] A first information score value of the target node is generated based on a plurality of leaf function values.

[0045] If the target node is a target node corresponding to a target road in the field of transportation, then according to each keyword of the first attribute information (daily traffic flow, design capacity, accident frequency), the characteristics corresponding to each basic data are extracted from the first attribute information, such as the characteristics of daily traffic flow, the characteristics of design capacity, and the characteristics of accident frequency. Specifically, the average traffic flow, peak traffic flow, and minimum traffic flow of the target road in a preset time period are obtained as the characteristics of daily traffic flow, the number of lanes, width, and speed limit included in the design standard of the target road are obtained as the characteristics of design capacity, and the number and severity of accidents on the target road in a preset time period are obtained as the characteristics of accident frequency.

[0046] If the target node is the target node corresponding to the target enterprise in the enterprise field, then according to the keywords of the first attribute information (news sentiment, financial statements, business information, litigation records), the features corresponding to the basic data are extracted from the first attribute information, such as the features of news sentiment, the features of financial statements, the features of business information, and the features of litigation records.

[0047] In one embodiment, extracting the features of the news sentiment, the features of the financial statements, the features of the business information, and the features of the litigation records includes:

[0048] Performing sentiment analysis on the news public opinion, and using the number and influence of positive news and negative news of the target node obtained by the analysis as the characteristics of the news public opinion;

[0049] Identify key financial indicators in the financial statements and use the key financial indicators as features of the financial statements, wherein the key financial indicators include current ratio, quick ratio, debt-to-asset ratio, and net profit margin;

[0050] Extracting the product sales, product market share, and product competitiveness of the target node from the business information as features of the business information;

[0051] The number of litigation cases and the ratio of winning to losing of the target node within a preset time period are obtained from the litigation records as features of the litigation records.

[0052] All the feature data extracted above are input into a pre-trained score prediction model. The score prediction model is trained by machine learning methods, such as the extreme gradient boosting decision tree model (XGBoost). Within the score prediction model, multiple tree functions are used to calculate the input features. Each tree will split the features and ultimately generate a function value for a leaf node. The function values ​​of each leaf node are integrated to generate the first information score value of the final target node.

[0053] Among them, the extreme gradient boosting decision tree model (XGBoost) is improved on the basis of the GBDT algorithm. It is a model that integrates multiple trees and is a type of Ensemble Method. It uses distributed loading data and distributed training data, and performs a second-order Taylor expansion on the loss function. In addition, a regularization term is added to the objective function to find the optimal solution as a whole, which is used to balance the decrease of the objective function and the complexity of the model to avoid overfitting.

[0054] S2. Obtain second attribute information corresponding to the target node from the preset database, determine multiple associated nodes associated with the target node in the preset knowledge graph based on the second attribute information, and calculate the association weight between the target node and each associated node based on the second attribute information, where the second attribute information is the relationship data between the target node and each associated node.

[0055] In this embodiment, it is necessary to obtain the second attribute information of the target node. If the target node is the target node corresponding to the target enterprise in the enterprise field, the relationship data between the target node and each associated node includes the asset relationship and transaction relationship between the target node and its associated nodes. The asset relationship includes but is not limited to: equity relationship (for example, the equity holding situation of the target node and its associated nodes), personnel relationship (for example, the relationship between the target node and key figures such as executives and directors of the associated nodes), debt relationship (for example, the loan relationship between the target node and the bank or associated node), guarantee relationship (for example, the situation where the target node provides a guarantee for the associated node). Transaction relationship includes but is not limited to: related transactions (for example, business dealings such as purchase, sale, lease, and service between the target node and the associated nodes), bidding (for example, the bidding relationship between the target node and the associated nodes), and supply chain (for example, the supply chain relationship between the target node and suppliers, distributors, etc.).

[0056] If the target node is the target node corresponding to the target road in the field of transportation, the relationship data between the target node and each associated node includes the geographical location relationship and traffic flow relationship between the target node and its associated nodes. The geographical location relationship includes but is not limited to: the intersection or junction between the target road and other roads, the geographical coordinates of the starting point and the end point of the target road, and the distance between the target road and important landmarks (such as airports, railway stations, and commercial centers). The traffic flow relationship includes but is not limited to: the comparison of the average daily traffic volume of the target road and adjacent roads, the traffic exchange between the target road and other roads during peak hours, and the impact of the target road as a main road on the surrounding branches.

[0057] According to the second attribute information, multiple associated nodes associated with the target node are determined in the preset knowledge graph, and the association weight between the target node and each associated node is calculated according to the second attribute information.

[0058] In one embodiment, before receiving a user's request to calculate associated data for a target node in a preset knowledge graph, the method further includes:

[0059] Obtain a preset number of target objects from a preset database, and determine each node of the preset knowledge graph according to the preset number of target objects, wherein the target objects are entity objects corresponding to each node in the preset knowledge graph;

[0060] The edges between the nodes are generated according to the relationship data between the target objects, and the preset knowledge graph is constructed based on the nodes and the edges between the nodes. The relationship data includes the closeness, depth and type of the relationship between the target objects.

[0061] In the enterprise field, a knowledge graph is pre-built based on the asset relationship and transaction relationship between the above target node and each associated node. In the knowledge graph, each node represents an enterprise, and the lines between the nodes represent the asset relationship and transaction relationship between enterprises. The knowledge graph can be stored and queried using graph database technology to ensure efficient data processing capabilities. By building a knowledge graph, it is possible to clearly identify which enterprises have an associated relationship with the target node, as well as the nature of these relationships (asset relationship, transaction relationship, etc.). This helps to accurately identify the source and path of risk transmission.

[0062] Obtain the mapping relationship between asset relationships, transaction relationships and the edge information of the preset knowledge graph. This step is to map the enterprise relationships in the real world to the node connections in the graph to ensure the accuracy and completeness of the knowledge graph. Traverse all nodes of the preset knowledge graph according to the mapping relationship and find the enterprise nodes directly connected to the target node. These nodes represent multiple associated nodes associated with the target node. For each associated node, calculate the association weight between the target node and it. The calculation of the association weight is based on factors such as the closeness of the relationship, the depth of the relationship and the type of relationship.

[0063] For example: Get the second attribute information of "Technology Company A" from the bank's database, determine multiple associated nodes associated with Technology Company A in the preset knowledge graph based on the asset relationship and transaction relationship between Technology Company A and each associated node, and calculate the association weight between Technology Company A and each associated node. Traverse the knowledge graph based on the asset relationship and transaction relationship to determine the enterprise nodes directly connected to Technology Company A, which represent multiple associated nodes associated with Technology Company A. For each associated node, obtain the relationship closeness, relationship depth and relationship type between Technology Company A and each associated node from the knowledge graph and calculate the association weight between the target node and each associated node.

[0064] In one embodiment, according to the second attribute information, determining a plurality of associated nodes associated with the target node in the preset knowledge graph includes:

[0065] Obtaining the relationship data between the target node and each associated node and the mapping relationship between the edges of the preset knowledge graph;

[0066] All nodes of the preset knowledge graph are traversed according to the mapping relationship to obtain multiple associated nodes associated with the target node.

[0067] In the enterprise field, the asset relationship and transaction relationship between the target node and the associated nodes are obtained, and mapped with the edge information in the preset knowledge graph. These asset relationships and transaction relationships are mapped with the edge information in the preset knowledge graph to ensure that each asset relationship or transaction relationship can find the corresponding edge in the preset knowledge graph. All nodes of the preset knowledge graph are traversed according to the mapping relationship to determine the associated nodes related to the target node.

[0068] Specific steps: Set the starting node as the target node, and ensure that you have a unique identifier for the target node (such as the enterprise ID), use a graph traversal algorithm (such as depth-first search DFS or breadth-first search BFS) to traverse the graph and find nodes directly connected to the target node, for example: select a suitable traversal algorithm based on the scale and characteristics of the preset knowledge graph. If the preset knowledge graph is small and the connectivity is not high, you can choose DFS; if the preset knowledge graph is large and the connectivity is high, it is recommended to use BFS.

[0069] Starting from the target node, visit its adjacent nodes, for each adjacent node, check whether there is an asset relationship or transaction relationship in the mapping relationship, for each node that has an asset relationship or transaction relationship with the target node, record its identifier and regard it as an associated node of the target node, and finally, obtain multiple associated nodes associated with the target node based on the recorded identifiers of the associated nodes. In other embodiments, an identifier list can be created to record all enterprise identifiers that have a direct relationship with the target node.

[0070] In one embodiment, the calculating the association weight between the target node and each associated node according to the second attribute information includes:

[0071] According to the relationship data between the target node and each associated node, the closeness, depth and type of the relationship between the target node and each associated node are obtained from the preset knowledge graph;

[0072] According to the relationship closeness, the relationship depth and the relationship type, a preset association risk factor formula is substituted for calculation to obtain an association weight between the target node and each associated node.

[0073] In the corporate sector, the closeness of the relationship measures the strength of the connection between two companies, quantified from the following aspects (equity holding ratio, senior management positions, transaction frequency, guarantee relationship):

[0074] Equity holding ratio: If the target company holds shares in another company, the higher the holding ratio, the closer the relationship. For example, if the target company holds more than 50% of the shares of another company, the relationship is considered to be very close. Executive positions: If the executives of the target company also hold positions in the associated nodes, the relationship is also relatively close. For example, the chairman of the target company is also a member of the board of directors of another company. Transaction frequency: If there are frequent transaction records between the target company and another company, the relationship is also relatively close. For example, the target company purchases raw materials from another company every month, so a high transaction frequency indicates a close relationship. Guarantee relationship: If the target company has provided multiple guarantees for another company, this is also a manifestation of a close relationship. For example, the target company has provided loan guarantees for another company many times.

[0075] Relationship Depth: Measures the number of indirect relationships between two companies, involving indirect relationships formed through other companies. Quantified from the following aspects (direct relationship, indirect relationship, multi-layer indirect relationship): Direct relationship: If there is a direct equity, transaction or guarantee relationship between the target company and another company, the relationship depth is 1. For example, the target company directly holds shares in another company. Indirect relationship: If the relationship between the target company and another company needs to be established through a third company, the relationship depth is 2. For example, the target company has established a certain relationship with another company through an intermediary company. Multi-layer indirect relationship: If the relationship needs to be established through multiple companies, the relationship depth increases accordingly. For example, the target company has established a certain relationship with the final company through an intermediary company and then through another intermediary company, the relationship depth is 3.

[0076] The relationship type is used to identify the type of relationship between two companies, including but not limited to the following (competitive relationship, upstream and downstream supply chain, partnership, equity relationship, transaction relationship, guarantee relationship): Competitive relationship: The target company has a competitive relationship with another company in the same industry, such as competition for market share. The closeness of this competitive relationship can be measured by market share comparison and competitor analysis. Upstream and downstream supply chain relationship: There is a supply chain relationship between the target company and another company, such as the target company is the supplier or customer of the other company. The closeness of this upstream and downstream supply chain can be measured by the stability and degree of dependence of the supply chain. Partnership: There is a strategic partnership between the target company and another company, such as joint research and development, joint marketing, etc. The closeness of this partnership can be measured by the closeness of the cooperation agreement.

[0077] Calculating the association weights between the target node and each associated node can quantify the degree of influence of each associated node on the target enterprise during the risk transmission process. This is crucial for subsequent calculation and mining of associated data, and can help determine which associated nodes have the most significant impact on the target node.

[0078] The calculation results of the association weights can be used to further analyze the transmission effect of risks between enterprises. By calculating the influencing factors of different types of relationships (such as equity relationships and debt relationships), the possibility and intensity of risk transmission can be more accurately evaluated.

[0079] According to the relationship closeness, relationship depth and relationship type, the preset association risk factor formula is substituted for calculation to obtain the association weight between the target node and each associated node.

[0080] As time goes by, the relationships between enterprises will change. By regularly updating the knowledge graph and recalculating the association weights, it is possible to dynamically monitor the risk status of enterprises and promptly discover new risk transmission paths and sources of risk.

[0081] In one embodiment, the calculation of the association weight between the target node and each associated node by substituting the relationship closeness, the relationship depth and the relationship type into a preset association risk factor formula includes:

[0082] The relationship closeness, the relationship depth and the relationship type between the target node and any associated node are multiplied to obtain an association weight of any associated node to the target node;

[0083] Repeat the step of calculating the association weight of any associated node to the target node to obtain the association weight between the target node and each associated node.

[0084] The preset correlation risk factor formula is:

[0085] k(t,i)=∑w(i,j)*d(i,j)*p

[0086] Among them, k(t,i) is the association weight of node i corresponding to the target node at time t when affected by the associated node j, w(i,j) is the closeness of the relationship between the node i and the node j, d(i,j) is the depth of the relationship between the node i and the node j, and p is the type of relationship between the node i and the node j.

[0087] In the enterprise field, the closeness of the relationship can be quantified based on the actual situation, for example: Equity holding ratio: If the target node holds more than 50% of the shares of the associated node, the closeness of the relationship is high. Transaction frequency: If the target node and the associated node have dozens of transactions per year, the closeness of the relationship is high.

[0088] The depth of a relationship can be expressed by the number of direct or indirect relationships. For example, direct relationship: d(i,j) = 1; indirect relationship through a third party: d(i,j) = 2.

[0089] Different types of relationships have different effects on risk transmission, so different influencing factors need to be set. For example: equity relationship: p_equity; transaction relationship: p_trade; guarantee relationship: p_guarantee; the setting of specific influencing factors can be determined based on historical data analysis or expert experience.

[0090] For example:

[0091] Assume that the relationship information between the target node A and the associated nodes B and C is as follows:

[0092] A and B: degree of relationship w(A,B) = 0.8 (high shareholding ratio); depth of relationship d(A,B) = 1 (direct relationship); relationship type is equity relationship, impact factor p_equity = 1.5.

[0093] A and C: relationship closeness w(A, C) = 0.5 (average transaction frequency); relationship depth d(A, C) = 2 (indirect relationship through a third party); relationship type is transaction relationship, impact factor p_trade = 1.0.

[0094] The association weights are calculated based on the above information: k(A,B)=0.8×1×1.5=1.2, k(A,C)=0.5×2×1.0=1.0. Through the above steps, the association weights between the target node and each associated node can be quantified, thereby providing accurate data support for subsequent association data calculation and mining.

[0095] S3. Obtain the third attribute information corresponding to each associated node from the preset database, calculate the second information score value corresponding to each associated node and the score level of each associated node according to the third attribute information of each associated node, and determine the node state corresponding to each associated node in the preset knowledge graph according to the second information score value corresponding to each associated node, wherein the node state includes susceptible nodes, infected nodes and immune nodes, and the third attribute information is basic data associated with each associated node.

[0096] In this embodiment, the method of obtaining the third attribute information of each associated node is the same as the method of obtaining the second attribute information of the target node. Each feature is extracted from the third attribute information of each associated node and input into a pre-trained scoring prediction model for calculation to obtain the second information scoring value of each associated node.

[0097] Determine the enterprise's rating level according to the pre-set rating level standards. Specific steps: Set different rating level standards according to the enterprise's second information rating value, for example: Low risk rating level: The second information rating value is low, indicating that the enterprise currently faces a small inherent risk. Medium risk rating level: The second information rating value is moderate, indicating that the enterprise has a certain inherent risk. High risk rating level: The second information rating value is high, indicating that the enterprise faces a high inherent risk. Determine the rating level of each associated node based on the calculated second information rating value.

[0098] According to the second information score of each associated node, its corresponding node status in the preset knowledge graph is determined. The node status includes susceptible nodes, infected nodes and immune nodes. Among them, the definition of susceptible nodes: the second information score is high, risk events or risk event infection sources are occurring, and the risk may be transmitted to other companies. The definition of infected nodes: companies with the second information score that may be infected by risk events and have close ties with companies in susceptible nodes. The definition of immune nodes: companies with low second information scores that are not easily affected by external risks and have fewer connections with companies in susceptible nodes or infected nodes.

[0099] Through the above steps, the second information score of each associated node can be accurately evaluated and divided into different node states, so as to better understand its position in the entire knowledge graph, identify potential risk transmission paths, and take corresponding risk management measures.

[0100] In the enterprise field, if an enterprise finds itself in the state of an infected node or an immune node, it can strengthen monitoring of risky enterprises and take measures to reduce contact with high-risk enterprises in susceptible nodes to reduce the risk of infection.

[0101] In the field of transportation, if the target road is found to be in the state of an infected node or an immune node, the risk of the target road is low and it is not easily affected by external factors, so the existing level of management and maintenance can be maintained. If the target road is found to be in a susceptible node (such as severe traffic congestion), immediate action is taken to resolve known problems, such as repairing damaged road surfaces, clearing obstacles, or strengthening law enforcement against traffic violations in the target road area.

[0102] In one embodiment, determining the node status corresponding to each associated node in the preset knowledge graph according to the second information score value of each associated node includes:

[0103] Obtaining a predefined classification standard for node status, wherein the classification standard is used to classify the node status into three types, and each type of node status is correspondingly provided with a different threshold;

[0104] The second information score value of each associated node is compared with a set threshold value of the classification standard, and the corresponding associated node state is defined according to the comparison result.

[0105] The classification standard of node status is predefined, and node status is divided into three types: susceptible nodes, infected nodes and immune nodes. Different thresholds are set for each type of node status.

[0106] In the enterprise field, different thresholds are set to distinguish different types of node states based on the second information score of each enterprise. For example:

[0107] Immune node: The second information score is low, the enterprise is relatively stable and not easily affected by risks. The threshold range is: [0, 0.3] (low risk interval). Infected node: The second information score is moderate. Although the current risk is low, it is easily affected by external risks. The threshold range is: (0.3, 0.6] (medium risk interval). Susceptible node: The second information score is high. The enterprise faces a high risk and may spread the risk to other enterprises. The threshold range is: (0.6, 1] (high risk interval).

[0108] The second information score value of each associated node is compared with the threshold value set above, and the type of node state is determined based on the comparison result. If the second information score value falls within the interval [0,0.3], the enterprise is defined as an immune node. If the second information score value falls within the interval (0.3,0.6], the enterprise is defined as an infected node. If the second information score value falls within the interval (0.6,1], the enterprise is defined as a susceptible node.

[0109] Through the above steps, the inherent risks of each associated node can be accurately assessed and divided into different node states, which can help enterprises better understand their position in the entire knowledge graph, identify potential risk transmission paths, and take corresponding risk management measures.

[0110] S4. Obtain an association data calculation differential equation constructed in advance based on the susceptible nodes, the infected nodes, and the immune nodes, substitute the association weights and the scoring levels of each association node into the association data calculation differential equation for calculation, and obtain a third information scoring value of the target node; generate an association data mining report of the target node based on the first information scoring value and the third information scoring value, and feed it back to the user.

[0111] In this embodiment, a pre-constructed association data calculation equation group is obtained. The association data calculation equation group is constructed based on the states of susceptible nodes, infected nodes, and immune nodes, and taking into account the association weights and rating levels between the nodes.

[0112] In the enterprise field, there are many situations in the process of calculating related data:

[0113] a) Taking the target node as the center, radiate the corresponding risk transmission probability, where the risk is the scoring type of the target node itself;

[0114] b) With the infected associated node as the center, there may be multiple source risk radiations, and the superposition and offset effect between enterprise risks needs to be further optimized;

[0115] c) Risk contagion caused by primary risks that are not caused by corporate risks, such as macroeconomic downturn or industry cyclical risks, needs to be considered.

[0116] According to the various situations existing in the above risk transmission probability process, the enterprise nodes are divided into the states of susceptible nodes, infected nodes and immune nodes to construct a specific set of correlation data calculation equations. The specific assumptions are as follows:

[0117]

[0118] The associated data calculation equations are:

[0119]

[0120] Among them, S(t) is the proportion of susceptible nodes, I(t) is the proportion of infected nodes, R(t) is the proportion of infected nodes, t is time t, which expresses the proportion of associated nodes in the three types of states at time t. Then S(t)+I(t)+R(t)=1, and λ is the risk transmission probability that all associated nodes transmit risk events to the target node.

[0121] The probability of risk transmission from associated node j to the target node is calculated separately: λ(i)=k(t,i)T(j), where k(t,i) is the association weight of node i corresponding to the target node at time t under the influence of associated node j, and T(j) is the rating of associated node j. k(t,i) is affected by factors such as the closeness of the relationship, the degree of association, and the type of relationship. The closer the relationship, the more direct the association, and the deeper the influence of the relationship type on the risk type, the greater k(t,i), which in turn leads to a greater probability of risk transmission.

[0122] Select enterprises that have a direct relationship with the target node to form node pairs, and calculate the unilateral risk transmission probability between the enterprise node pairs. Considering that in reality, the target node will be affected by risk events of multiple associated nodes, constructing a group of associated data calculation equations can perform multilateral probability calculations in a one-time and comprehensive manner, thereby obtaining the comprehensive probability of the target node being infected, which can improve the efficiency of associated data calculations.

[0123] Substitute the association weights between the target node and each associated node and the rating level of each associated node into the risk transmission calculation formula to obtain the risk transmission probability. For example, the risk transmission probability of the associated node j to the target node is calculated separately: λ(i)=k(t,i)T(j).

[0124] The calculated risk transmission probability is substituted into the associated data calculation equation group for calculation to obtain the third information score value of the target node, and an associated data mining report (for example, a data analysis view) of the target node is generated according to the first information score value and the third information score value.

[0125] The associated data calculation equations used in the present invention are modified and applied based on the classic SIR epidemic model, and can play the following roles:

[0126] 1. The associated data calculation equation group can simultaneously consider the first information score of the target node and the third information score brought by the associated node, making the associated data calculation more comprehensive. This method not only focuses on the health status of the target node itself, but also considers the impact of the external environment on the target node.

[0127] 2. The associated data calculation equation group can simulate the dynamic propagation process of risk events between nodes. Since the relationship and risk status between nodes are constantly changing, it can help nodes monitor the evolution of risks in real time and adjust risk management strategies in a timely manner.

[0128] 3. By constructing a group of related data calculation equations, the impact of multiple related nodes on the risk of the target node can be calculated at one time, avoiding the calculation of the probability of risk transmission between each node one by one, and improving the calculation efficiency.

[0129] To summarize, in the above steps S1-S4, the present invention extracts the first attribute information of the target node from the database to score the target node to obtain the first information score value, and determines the association weights between other nodes associated with the target node according to the second attribute information, and uses the third attribute information of each associated node to calculate the second information score value of each associated node, and classifies the associated nodes into susceptible, infected or immune states, and substitutes the association weights and the score levels of each associated node into the associated data calculation differential equation to obtain the first information score value of the target node, and generates an associated data mining report of the target node according to the first information score value and the third information score value. The present invention calculates the association relationship between each node in the knowledge graph through the knowledge graph's capabilities in information expression, complex network construction and intelligent reasoning and the associated data calculation differential equation constructed with "susceptible-infected-immune", thereby improving the accuracy of associated data mining.

[0130] like Figure 3 As shown, it is a module schematic diagram of an association data processing device based on a knowledge graph provided by one embodiment of the present invention.

[0131] The associated data processing device 100 based on the knowledge graph of the present invention can be installed in an electronic device. According to the functions to be implemented, the associated data processing device 100 based on the knowledge graph can include a first calculation module 110, a second calculation module 120, a determination module 130 and a generation module 140. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, which are stored in the memory of the electronic device.

[0132] In this embodiment, the functions of each module / unit are as follows:

[0133] A first calculation module 110 is used to obtain first attribute information corresponding to a target node in a preset knowledge graph from a preset database after receiving a user's request to calculate associated data for the target node, and calculate a first information score value of the target node according to the first attribute information, where the first attribute information is basic data associated with the target node;

[0134] A second calculation module 120 is used to obtain second attribute information corresponding to the target node from the preset database, determine multiple associated nodes associated with the target node in the preset knowledge graph according to the second attribute information, and calculate the association weight between the target node and each associated node according to the second attribute information, where the second attribute information is the relationship data between the target node and each associated node;

[0135] The determination module 130 is used to obtain the third attribute information corresponding to each associated node from the preset database, calculate the second information score value corresponding to each associated node and the score level of each associated node according to the third attribute information of each associated node, and determine the node state corresponding to each associated node in the preset knowledge graph according to the second information score value corresponding to each associated node, wherein the node state includes a susceptible node, an infected node, and an immune node, and the third attribute information is the basic data associated with each associated node;

[0136] The generation module 140 is used to obtain the association data calculation differential equation constructed in advance based on the susceptible node, the infected node and the immune node, substitute the association weight and the scoring level of each association node into the association data calculation differential equation for calculation, obtain the third information scoring value of the target node, generate the association data mining report of the target node according to the first information scoring value and the third information scoring value, and feed it back to the user.

[0137] In one embodiment, the first calculation module 110 is specifically configured to:

[0138] Extracting features corresponding to each basic data from basic data associated with the target node according to each keyword of the first attribute information;

[0139] Inputting the extracted features into a preset score prediction model, wherein the preset score prediction model is generated by an extreme gradient boosting decision tree model;

[0140] Utilizing the plurality of tree functions of the preset scoring prediction model to calculate each feature respectively, and generating a plurality of leaf function values;

[0141] A first information score value of the target node is generated based on a plurality of leaf function values.

[0142] In one embodiment, the first calculation module 110 is specifically configured to:

[0143] Obtain a preset number of target objects from a preset database, and determine each node of the preset knowledge graph according to the preset number of target objects, wherein the target objects are entity objects corresponding to each node in the preset knowledge graph;

[0144] The edges between the nodes are generated according to the relationship data between the target objects, and the preset knowledge graph is constructed based on the nodes and the edges between the nodes. The relationship data includes the closeness, depth and type of the relationship between the target objects.

[0145] In one embodiment, the second calculation module 120 is specifically configured to:

[0146] Obtaining the relationship data between the target node and each associated node and the mapping relationship between the edges of the preset knowledge graph;

[0147] All nodes of the preset knowledge graph are traversed according to the mapping relationship to obtain multiple associated nodes associated with the target node.

[0148] In one embodiment, the second calculation module 120 is specifically configured to:

[0149] According to the relationship data between the target node and each associated node, the closeness, depth and type of the relationship between the target node and each associated node are obtained from the preset knowledge graph;

[0150] According to the relationship closeness, the relationship depth and the relationship type, a preset association risk factor formula is substituted for calculation to obtain an association weight between the target node and each associated node.

[0151] In one embodiment, the second calculation module 120 is specifically configured to:

[0152] The relationship closeness, the relationship depth and the relationship type between the target node and any associated node are multiplied to obtain an association weight of any associated node to the target node;

[0153] Repeat the step of calculating the association weight of any associated node to the target node to obtain the association weight between the target node and each associated node.

[0154] In one embodiment, the determination module 130 is specifically configured to:

[0155] Obtaining a predefined classification standard for node status, wherein the classification standard is used to classify the node status into three types, and each type of node status is correspondingly provided with a different threshold;

[0156] The first information score value of each associated node is compared with a set threshold value of the classification standard, and the node state of the corresponding associated node is defined according to the comparison result.

[0157] like Figure 4 As shown, it is a schematic diagram of the structure of an electronic device for implementing a method for processing associated data based on a knowledge graph provided by one embodiment of the present invention.

[0158] In this embodiment, the electronic device 1 includes, but is not limited to, a memory 11, a processor 12, and a network interface 13, which can be interconnected and communicated via a system bus. The memory 11 stores an associated data processing program 10 based on a knowledge graph, and the associated data processing program 10 based on a knowledge graph can be executed by the processor 12. Figure 4 Only the electronic device 1 having components 11-13 and the associated data processing program 10 based on the knowledge graph is shown. It can be understood by those skilled in the art that Figure 4 The structure shown does not constitute a limitation on the electronic device 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0159] Among them, the memory 11 includes a memory and at least one type of readable storage medium. The memory provides a cache for the operation of the electronic device 1; the readable storage medium may be a non-volatile storage medium such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the readable storage medium may be an internal storage unit of the electronic device 1; in other embodiments, the non-volatile storage medium may also be an external storage device of the electronic device 1, such as a plug-in hard disk equipped on the electronic device 1, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), etc. In this embodiment, the readable storage medium of the memory 11 is generally used to store an operating system and various application software installed on the electronic device 1, such as storing the code of the associated data processing program 10 based on the knowledge graph in an embodiment of the present invention. In addition, the memory 11 can also be used to temporarily store various types of data that have been output or are to be output.

[0160] The processor 12 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 12 is generally used to control the overall operation of the electronic device 1, such as performing control and processing related to data interaction or communication with other devices. In this embodiment, the processor 12 is used to run the program code stored in the memory 11 or process data, such as running the associated data processing program 10 based on the knowledge graph.

[0161] The network interface 13 may include a wireless network interface or a wired network interface, and the network interface 13 is used to establish a communication connection between the electronic device 1 and a terminal (not shown in the figure).

[0162] Optionally, the electronic device 1 may further include a user interface, which may include a display (Display), an input unit such as a keyboard (Keyboard), and the optional user interface may also include a standard wired interface and a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0163] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0164] The associated data processing program 10 based on the knowledge graph stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 12, it can achieve:

[0165] After receiving a user's request to calculate associated data for a target node in a preset knowledge graph, first attribute information corresponding to the target node is obtained from a preset database, and a first information score value of the target node is calculated according to the first attribute information, where the first attribute information is basic data associated with the target node;

[0166] Acquire second attribute information corresponding to the target node from the preset database, determine multiple associated nodes associated with the target node in the preset knowledge graph according to the second attribute information, and calculate the association weight between the target node and each associated node according to the second attribute information, wherein the second attribute information is relationship data between the target node and each associated node;

[0167] Obtaining third attribute information corresponding to each associated node from the preset database, calculating the second information score value corresponding to each associated node and the score level of each associated node according to the third attribute information of each associated node, and determining the node state corresponding to each associated node in the preset knowledge graph according to the second information score value corresponding to each associated node, wherein the node state includes a susceptible node, an infected node, and an immune node, and the third attribute information is basic data associated with each associated node;

[0168] Obtain an association data calculation differential equation constructed in advance based on the susceptible node, the infected node and the immune node, substitute the association weight and the scoring level of each association node into the association data calculation differential equation for calculation, and obtain the third information scoring value of the target node; generate an association data mining report of the target node according to the first information scoring value and the third information scoring value and feed it back to the user.

[0169] Specifically, the specific implementation method of the processor 12 for the above-mentioned knowledge graph-based associated data processing program 10 can refer to Figure 2 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0170] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable medium can be non-volatile or non-volatile. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0171] The computer-readable storage medium stores an associated data processing program 10 based on a knowledge graph, and the associated data processing program 10 based on a knowledge graph can be executed by one or more processors. The specific implementation method of the computer-readable storage medium of the present invention is basically the same as the above-mentioned embodiments of the associated data processing method based on the knowledge graph, and will not be repeated here.

[0172] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0173] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0174] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0175] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0176] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0177] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claim can also be implemented by one unit or device through software or hardware. The second and other words are used to indicate names, but not to indicate any particular order.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention. If software tools or components other than those of the company appear in the embodiments of the present application, they are only used for example introduction and do not represent actual use.

Claims

1. A method for processing associated data based on knowledge graph, characterized in that: The method comprises: After receiving a user's request to calculate associated data for a target node in a preset knowledge graph, first attribute information corresponding to the target node is obtained from a preset database, and a first information score value of the target node is calculated according to the first attribute information, where the first attribute information is basic data associated with the target node; Acquire second attribute information corresponding to the target node from the preset database, determine multiple associated nodes associated with the target node in the preset knowledge graph according to the second attribute information, and calculate the association weight between the target node and each associated node according to the second attribute information, wherein the second attribute information is relationship data between the target node and each associated node; Obtaining third attribute information corresponding to each associated node from the preset database, calculating the second information score value corresponding to each associated node and the score level of each associated node according to the third attribute information of each associated node, and determining the node state corresponding to each associated node in the preset knowledge graph according to the second information score value corresponding to each associated node, wherein the node state includes a susceptible node, an infected node, and an immune node, and the third attribute information is basic data associated with each associated node; Obtain an association data calculation differential equation constructed in advance based on the susceptible node, the infected node and the immune node, substitute the association weight and the scoring level of each association node into the association data calculation differential equation for calculation, and obtain the third information scoring value of the target node; generate an association data mining report of the target node according to the first information scoring value and the third information scoring value and feed it back to the user.

2. The method for processing associated data based on knowledge graph according to claim 1, characterized in that: The calculating the first information score value of the target node according to the first attribute information includes: Extracting features corresponding to each basic data from basic data associated with the target node according to each keyword of the first attribute information; Inputting the extracted features into a preset score prediction model, wherein the preset score prediction model is generated by an extreme gradient boosting decision tree model; Utilizing the plurality of tree functions of the preset scoring prediction model to calculate each feature respectively, and generating a plurality of leaf function values; A first information score value of the target node is generated based on a plurality of leaf function values.

3. The method for processing associated data based on knowledge graph according to claim 1, characterized in that: Before receiving a user's request to calculate associated data for a target node in a preset knowledge graph, the method further includes: Obtain a preset number of target objects from a preset database, and determine each node of the preset knowledge graph according to the preset number of target objects, wherein the target objects are entity objects corresponding to each node in the preset knowledge graph; The edges between the nodes are generated according to the relationship data between the target objects, and the preset knowledge graph is constructed based on the nodes and the edges between the nodes. The relationship data includes the closeness, depth and type of the relationship between the target objects.

4. The method for processing associated data based on knowledge graph according to claim 1, characterized in that: Determining, according to the second attribute information, a plurality of associated nodes associated with the target node in the preset knowledge graph includes: Obtaining the relationship data between the target node and each associated node and the mapping relationship between the edges of the preset knowledge graph; All nodes of the preset knowledge graph are traversed according to the mapping relationship to obtain multiple associated nodes associated with the target node.

5. The method for processing associated data based on knowledge graph according to claim 1 or 3, characterized in that: The calculating the association weight between the target node and each associated node according to the second attribute information includes: According to the relationship data between the target node and each associated node, the closeness, depth and type of the relationship between the target node and each associated node are obtained from the preset knowledge graph; According to the relationship closeness, the relationship depth and the relationship type, a preset association risk factor formula is substituted for calculation to obtain an association weight between the target node and each associated node.

6. The method for processing associated data based on knowledge graph according to claim 5, characterized in that: Substituting the relationship closeness, the relationship depth and the relationship type into a preset association risk factor formula for calculation to obtain an association weight between the target node and each associated node includes: The relationship closeness, the relationship depth and the relationship type between the target node and any associated node are multiplied to obtain an association weight of any associated node to the target node; Repeat the step of calculating the association weight of any associated node to the target node to obtain the association weight between the target node and each associated node.

7. The method for processing associated data based on knowledge graph according to claim 1, characterized in that: The determining, according to the second information score value of each associated node, the node state corresponding to each associated node in the preset knowledge graph includes: Obtaining a predefined classification standard for node status, wherein the classification standard is used to classify the node status into three types, and each type of node status is correspondingly provided with a different threshold; The first information score value of each associated node is compared with a set threshold value of the classification standard, and the node state of the corresponding associated node is defined according to the comparison result.

8. A device for processing associated data based on knowledge graph, characterized in that: The device comprises: A first calculation module is used to obtain first attribute information corresponding to a target node in a preset knowledge graph from a preset database after receiving a user's request to calculate associated data for the target node, and calculate a first information score value of the target node according to the first attribute information, where the first attribute information is basic data associated with the target node; A second calculation module is used to obtain second attribute information corresponding to the target node from the preset database, determine multiple associated nodes associated with the target node in the preset knowledge graph according to the second attribute information, and calculate the association weight between the target node and each associated node according to the second attribute information, wherein the second attribute information is the relationship data between the target node and each associated node; A determination module, used to obtain the third attribute information corresponding to each associated node from the preset database, calculate the second information score value corresponding to each associated node and the score level of each associated node according to the third attribute information of each associated node, and determine the node state corresponding to each associated node in the preset knowledge graph according to the second information score value corresponding to each associated node, wherein the node state includes a susceptible node, an infected node, and an immune node, and the third attribute information is basic data associated with each associated node; A generation module is used to obtain an association data calculation differential equation constructed in advance based on the susceptible node, the infected node and the immune node, substitute the association weight and the scoring level of each association node into the association data calculation differential equation for calculation, obtain the third information scoring value of the target node, generate an association data mining report of the target node according to the first information scoring value and the third information scoring value, and feed it back to the user.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a knowledge graph-based associated data processing program that can be executed by the at least one processor, and the knowledge graph-based associated data processing program is executed by the at least one processor so that the at least one processor can execute the knowledge graph-based associated data processing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an associated data processing program based on a knowledge graph, and the associated data processing program based on a knowledge graph can be executed by one or more processors to implement the associated data processing method based on a knowledge graph as described in any one of claims 1 to 7.

Images