WO2021174693A1 - 一种数据分析方法、装置、计算机系统及可读存储介质 - Google Patents

一种数据分析方法、装置、计算机系统及可读存储介质 Download PDF

Info

Publication number
WO2021174693A1
WO2021174693A1 PCT/CN2020/093201 CN2020093201W WO2021174693A1 WO 2021174693 A1 WO2021174693 A1 WO 2021174693A1 CN 2020093201 W CN2020093201 W CN 2020093201W WO 2021174693 A1 WO2021174693 A1 WO 2021174693A1
Authority
WO
WIPO (PCT)
Prior art keywords
risk
node
path
model
rate
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2020/093201
Other languages
English (en)
French (fr)
Inventor
段洪云
汪伟
彭琛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2021174693A1 publication Critical patent/WO2021174693A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G06F16/9024—Graphs; Linked lists
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/906—Clustering; Classification

Definitions

  • This application relates to the field of computer technology, which relates to data mining technology for big data, and in particular to a data analysis method, device, computer system, and readable storage medium.
  • the current knowledge graph only describes the node data and the relationship between the nodes, but the information alone can only provide the basic information of the node and the related information between the nodes, without deep mining of the information in the knowledge graph. As a result, it can only provide simple information, but cannot provide users with valuable information that can be used directly.
  • the purpose of this application is to provide a data analysis method, device, computer system, and readable storage medium, which are used to solve the problem of in-depth mining of the information in the knowledge graph in the prior art, resulting in that it can only provide simple information.
  • this application provides a data analysis method based on big data, including:
  • the creation server creates a directed graph used to describe the association relationship and asset relationship between nodes, and calculates the risk transmission coefficient of each path in the directed graph through the infectious disease model to obtain a scale-free model, and sends it to the risk server;
  • the node refers to the information owner
  • the association relationship is used to reflect the involvement and influence between the information owners
  • the asset relationship is used to reflect the asset association ratio between the information owners;
  • the risk server identifies the infected node in the scale-free model, and calculates the risk transmission rate of each path in the scale-free model according to the infected node and combined with the risk transmission coefficient to obtain the risk infection model and send it Computing server
  • the calculation server extracts the node in the risk infection model according to the node request sent by the user terminal and uses it as the target node, and calculates the risk transmission rate in the incoming direction and the risk transmission rate in the outgoing direction of the target node to obtain transmission Incoming risk rate and outgoing risk rate; wherein the node request includes the node name corresponding to the node in the risk infection model, which is used to extract the node in the risk infection model.
  • this application also provides a data analysis device based on big data, including:
  • Create a server used to create a directed graph describing the relationship between nodes and asset relationships, and calculate the risk transmission coefficient of each path in the directed graph through the infectious disease model to obtain a scale-free model, and send it Risk server; wherein the node refers to the information owner, the association relationship is used to reflect the involvement and influence between the information owners, and the asset relationship is used to reflect the asset association ratio between the information owners;
  • the risk server is used to identify infected nodes in the scale-free model, and calculate the risk transmission rate of each path in the scale-free model according to the infected nodes and in combination with the risk transmission coefficient to obtain the risk infection model and Send computing server;
  • the calculation server is used to extract the node in the risk infection model and use it as the target node according to the node request sent by the user terminal, and calculate the risk transmission rate of the target node in the incoming direction and the risk transmission rate in the outgoing direction to obtain Incoming risk rate and outgoing risk rate; wherein the node request includes a node name corresponding to the node in the risk infection model, which is used to extract the node in the risk infection model.
  • the present application also provides a computer system, which includes a plurality of computer devices, each computer device includes a memory, a processor, and a computer program stored in the memory and running on the processor.
  • each computer device includes a memory, a processor, and a computer program stored in the memory and running on the processor.
  • the processor of the device executes the computer program, the steps of the above-mentioned data analysis method are jointly implemented.
  • the present application also provides a computer-readable storage medium, which includes multiple storage media, each of which stores a computer program, and when the computer program stored in the multiple storage media is executed by a processor Jointly implement the steps of the above data analysis method.
  • the data analysis method, device, computer system, and readable storage medium provided by this application create a directed graph for describing the association relationship and asset relationship between nodes, and calculate each of the directed graphs through an infectious disease model.
  • the risk transmission coefficient of the path is obtained to obtain a scale-free model to describe the degree of association between the nodes; by identifying the infected node in the scale-free model, the non-standard model is calculated according to the infected node and combined with the risk transmission coefficient.
  • the risk transmission rate of each path in the degree model is used to obtain the risk infection model; therefore, according to the infection node, the risk transmission coefficient and the average transmission rate are used to calculate the risk transmission rate of each path in the scale-free model, which is used to express the risk from infection.
  • the risk transmission rate of a node to other nodes because the risk transmission rate is obtained based on the average conductivity and the risk transmission coefficient, so it can reflect the probability of risk transmission between the two related nodes in the most true manner.
  • the node in the risk infection model is extracted and used as the target node, and the risk transmission rate in the incoming direction and the outgoing direction of the target node are calculated to obtain the incoming risk rate and Outgoing risk rate; Achieve a comprehensive knowledge of the outgoing risk rate and incoming risk rate of the target node, in order to comprehensively evaluate the risk characteristics of the target node, so as to comprehensively convey the risk environment of the target node to users, which is conducive to users according to the link Make judgments.
  • the four-quadrant model is used to calculate the incoming risk rate and the outgoing risk rate to obtain the judgment result, and output it to the user terminal.
  • the four-quadrant model is used to evaluate the risk characteristics of the target node, and the characteristics are named Or icons are output to the user terminal, so that the user can quickly learn about the risk environment describing the impact of surrounding nodes on the target node, and the impact of the target node on surrounding nodes, and realize the in-depth mining of the information in the knowledge graph, thereby solving
  • the lack of in-depth mining of the information in the knowledge map results in a technical problem that it can only provide simple information, but cannot provide users with valuable information that can be directly used.
  • FIG. 1 is a flowchart of Embodiment 1 of the data analysis method of this application;
  • 3 is a tree diagram of the data association mode of Euler Atlas in the first embodiment of the data analysis method of this application;
  • Embodiment 4 is a directed graph with primary conductivity in Embodiment 1 of the data analysis method of this application;
  • FIG. 5 is a directed graph with a primary conduction coefficient in which the risk conduction coefficient is loaded on each path in the first embodiment of the data analysis method of this application;
  • Fig. 6 is a directed graph of the risk infection model in the first embodiment of the data analysis method of this application.
  • FIG. 7 is a schematic diagram of the four-quadrant model in Embodiment 1 of the data analysis method of this application;
  • FIG. 8 is a schematic diagram of program modules of Embodiment 2 of the data analysis device of this application.
  • FIG. 9 is a schematic diagram of the hardware structure of the computer equipment in the third embodiment of the computer system of this application.
  • the data analysis method, device, computer system, and readable storage medium provided in this application are suitable for the computer field and provide a data analysis based on a big data creation module, an infection model creation module, a risk calculation module, and a risk judgment module method.
  • This application creates a directed graph used to describe the association relationship and asset relationship between nodes, and calculates the risk transmission coefficient of each path in the directed graph through an infectious disease model to obtain a scale-free model; identify the scale-free model
  • the infection node in the model calculates the risk transmission rate of each path in the scale-free model according to the infection node and the risk transmission coefficient to obtain the risk infection model; extracts the risk infection model according to the node request sent by the client And use it as the target node, calculate the risk transmission rate in the incoming direction and the outgoing direction of the target node to obtain the incoming risk rate and the outgoing risk rate; calculate the transmission risk rate through the four-quadrant model Incoming risk rate and outgoing risk rate to obtain the judgment result, and output it to the user terminal
  • a data analysis method based on big data in this embodiment includes:
  • S1 Create a server to create a directed graph used to describe the relationship between nodes and asset relationships, and calculate the risk transmission coefficient of each path in the directed graph through the infectious disease model to obtain a scale-free model, and send it to risk Server; wherein, the node refers to the information owner, the association relationship is used to reflect the involvement and influence between the information owners, and the asset relationship is used to reflect the asset association ratio between the information owners;
  • the risk server identifies the infected node in the scale-free model, and calculates the risk transmission rate of each path in the scale-free model according to the infected node and combined with the risk transmission coefficient to obtain the risk infection model and It sends the computing server;
  • the calculation server extracts the node in the risk infection model according to the node request sent by the user and uses it as the target node, and calculates the risk transmission rate in the incoming direction and the outgoing direction of the target node to Obtain the incoming risk rate and the outgoing risk rate; wherein the node request includes a node name corresponding to the node in the risk infection model, which is used to extract the node in the risk infection model.
  • the calculation of the risk transmission rate in the incoming direction and the risk transmission rate in the outgoing direction of the target node to obtain the incoming risk rate and the outgoing risk rate includes:
  • S4 Enter the incoming risk rate and outgoing risk rate into a preset four-quadrant model to obtain risk points, identify the area where the risk point is located and use the name of the area as the judgment result, and output the judgment result to the user end.
  • a directed graph is created to express the association relationship and asset relationship between nodes, where the node can be an enterprise or a natural person. Therefore, the name or name of the enterprise or natural person can be used as the node.
  • Calculate the average conduction probability of the directed graph in equilibrium that is, the number of nodes affected by the infected node is equal to the number of nodes free from the impact of the infected node
  • the risk transmission probability of each path in the graph is loaded on each path to obtain a scale-free model; wherein the scale-free model is a directed graph that reflects the relationship between the nodes and the asset relationship Based on the infectious disease model, the risk transmission probability of each path in the directed graph is calculated and loaded on the data model of each path.
  • a directed graph is a mathematical method of representing the relationship between objects and objects. It consists of some small dots (called vertices or nodes) and straight lines or curves connecting these dots (called edges). ) Composed of; in this embodiment, the dots correspond to the nodes in this application, the straight lines or curves connecting these dots, the arrows connecting the dots or the curves are loaded on the straight line Or the information on the curve corresponds to the asset relationship.
  • the infectious disease model is a standard measurement model used to calculate the influence of nodes. The nodes in the infectious disease model have three states: susceptible, infected, and recovered.
  • a susceptible person can be infected by an infected person, an infected person is infected and has the ability to infect a susceptible person, and a recovered person is recovered from an infected person, who no longer has the ability to infect and will not be infected again.
  • the infectious disease model parameters include the recovery rate ⁇ , the infection probability ⁇ , the number of repeated simulations T, and the simulation time timespace.
  • the simulation process is as follows: select a node i in the network as the infected node, and spread the virus to the neighbor susceptible nodes connected to i with the infection probability ⁇ , and the infected nodes continue to infect their neighbor susceptible nodes with the probability ⁇ .
  • each diseased node transforms into a recovery node with a probability of ⁇ at each stage.
  • the blacklist system can be used to identify the infected nodes in the scale-free model, that is: if a node belongs to the blacklist, the node is determined to be an infected node; the risk of an infected node is set to 100%, and it will pass through the path with the infected node
  • the connected nodes are regarded as directly connected nodes, and the risk transmission rate of the risk from the infected node to the directly connected node is calculated according to the risk transmission coefficient on the path; then the nodes connected to the connected node and the connected nodes are identified And set the connected nodes as indirectly connected nodes, and calculate the risk conduction rate of the risk from the infected node to the indirectly connected node through the directly connected node through the risk transmission coefficient on the connected path, By analogy, until all nodes directly connected and indirectly connected to the infected node in the scale-free model are calculated, the risk infection model is obtained.
  • the node request includes the node name used to correspond to the node in the risk infection model, the node with the node name in the risk infection model is used as the target node, the path in the risk infection model is identified, and the path to the target node is identified As the incoming path, the path pointed out by the target node is taken as the outgoing path, the incoming risk rate in the incoming direction is obtained according to the risk conduction rate of the incoming path, and the outgoing risk rate is obtained according to the risk conduction rate of the outgoing path The outgoing risk rate of the direction.
  • the user can set an area in the coordinate system of the four-quadrant model and assign a name to the area; according to the incoming risk rate and the outgoing risk rate, the risk point is obtained in the four-quadrant model, and the name of the area where the risk point is located is used as the judgment As a result, the judgment result is output to the user terminal.
  • the creation of the directed graph used to describe the association relationship and asset relationship between nodes includes:
  • S101 Obtain node data and the association relationship between the nodes from the service system, and construct a directed graph for describing the association relationship between the nodes according to the association relationship.
  • the knowledge graph is obtained from the service system, the node data and the association relationship between the nodes are obtained from the knowledge graph, and if there is an association relationship between the two node data, then the two node data are Draw a path between the corresponding nodes. If node A is an investment relationship with node B, the path between node A and node B is A to B; among them, because the path in the knowledge graph has arrows, it can be learned from the knowledge The relationship between node A and node B is directly learned in the graph, so it will not be repeated in this application.
  • the service system stores a large amount of enterprise and personal data, and establishes a large amount of enterprise relationship data, equity relationship data, litigation relationship data, etc. based on the enterprise and personal data.
  • the service system can cover industry, commerce, finance and economics. Announcements, legal documents, social media, and overseas public opinion; the service system described in this application uses Euler Atlas, which is a type of storage for corporate and personal data, and is constructed based on corporate and personal data.
  • the corporate network of related persons and related companies from the knowledge map of the six major data relationships of shareholder stock relationship, foreign investment relationship, supply chain relationship, equity pledge relationship, financing guarantee relationship and corporate executives; this application is to be resolved
  • the technical problem is how to deeply dig the knowledge relationship of the current knowledge graph (Euler graph in this application) to obtain information of potential application value; and for obtaining node data and various information according to the knowledge graph (Euler graph in this application)
  • the association relationship between the nodes can be easily obtained by a person skilled in the art through the extracted Euler graph, so it will not be repeated here. Since this system is an existing technology, it will not be repeated here.
  • S102 Calculate the asset association ratio between the interconnected nodes in the directed graph, and load it on the path between the interconnected nodes to describe the asset relationship between the nodes in the directed graph.
  • the node data of two nodes that are related to each other are extracted, and the invested data and investment data in the node data are obtained, wherein the one of the two nodes that emits the arrow (path) is regarded as the investment node, and the The party pointed by the arrow (path) is used as the invested node, extract the investment data in the investment node, and the invested data associated with the investment data in the invested node, and divide the invested data by the investment data to obtain the asset association And load the asset correlation ratio on the path.
  • the node data of the two interconnected nodes is information describing that one of the nodes invests in the other node, so in the node data of the two nodes, one of the node data of the two nodes must be the node data of the investment node, and the other is the node data of the investment node.
  • the node data of the invested node refers to the ratio of the amount received by the invested node to the total amount of foreign investment of the investment node; the invested data associated with the investment data refers to the ratio of the investment data to the invested node.
  • the invested data generated by the investment for example, the investment data of the investment node is 1 million, the invested data of the invested node is 500,000, and the 200,000 of the invested data is based on the investment data of the investment node,
  • the other 300,000 is obtained based on the investment data of other investment nodes, so the asset correlation ratio between the investment node and the invested node is 20%.
  • the calculation of the risk transmission coefficients of the paths in the directed graph through the infectious disease model to obtain the scale-free model includes:
  • S111 Calculate the model index of the directed graph through the infectious disease model to obtain the average transmission probability.
  • the infectious disease model has an equilibrium condition setting, and the average conductivity of the model index of the directed graph under the equilibrium condition is calculated through the infectious disease model.
  • infectious disease model is a standard measurement model used to calculate the influence of nodes. Based on the principle of the infectious disease model, the process of risk transmission in the directed graph is that the susceptible person receives the infection of the risk node. The infected person will recover and become the recoverer, and the recovered person will become the infected person again due to the infection of the risk node again.
  • Probability refers to the probability of risk transmission that can maintain the equilibrium condition of the directed graph.
  • the objective function is as follows:
  • the method first defines the following:
  • K refers to the total number of individuals in the infectious disease model, corresponding to the total number of nodes; risk individuals are recorded as I, healthy individuals are recorded as S; S(t) is the number of healthy individuals at time t; ⁇ (t) is at time t K ⁇ is the ratio of the number of individuals at risk that can be infected per unit time to the total number of healthy individuals at the time, corresponding to the product of the node degree and the average degree; ⁇ refers to the average density of the infectious disease model, corresponding to The average density; risk individuals pass the risk ⁇ through probability ⁇ , and the risk individuals will recover health but will not be immune;
  • the equilibrium state is obtained (that is, the number of newly infected people in the current period is equal to the number of newly recovered healthy people in the current period.
  • Number is equal to the overall infection probability ⁇ , and the overall infection probability ⁇ is set as the average transmission probability.
  • the objective function of the commercial environment infectious disease model is only in a balanced state, and the medical infectious disease model dynamically simulates the change in the number of infected patients in the population to obtain the population's infectious proportion in the balanced state.
  • S112 The average conduction probability is respectively multiplied by the asset correlation ratio on the path of the directed graph to obtain the primary conduction coefficient, and the correlation degree between the interconnected nodes in the directed graph is identified through the random forest model , And load it on the path of the interconnected nodes;
  • the path between the two nodes is assigned a correlation value of 1; if it is judged that the node data of the two nodes are not related, then the The path between the two nodes is assigned a correlation value of 0; the risk conduction coefficient is loaded on the path of the directed graph.
  • a random forest is a classifier that contains multiple decision trees, and its output category is a model that is determined by the mode of the category output by an individual tree; because there are many decision trees in the random forest, and each There is no correlation between a decision tree, so after getting the forest, when a new input sample enters, let each decision tree in the forest make a judgment separately to see which sample should belong to One category (for the classification algorithm), and then see which category is selected the most, and then predict which category this sample belongs to.
  • the random forest is obtained by training in the following way:
  • N samples are prepared, and N samples are randomly selected for replacement (one sample is randomly selected each time, and then returned to continue selection).
  • the selected N samples are used to train a decision tree as the samples at the root node of the decision tree.
  • each node In the process of forming the decision tree, each node must be split according to step 2 until it can no longer be split.
  • the user trains the initial random forest by setting the m attributes and classification strategies and using the above-mentioned method to obtain node data for identifying two nodes that are related to each other, whether the random forest is related Model; Because the training method of random forest belongs to the common sense of the public of those skilled in the art, the training process of training random forest by setting m attributes and classification strategies will not be repeated here.
  • S113 Multiply the primary conduction coefficient and the correlation degree on the path of the directed graph to obtain a risk conduction coefficient, and load the risk conduction coefficient on the path of the directed graph to obtain a scale-free model.
  • the directed graph with primary conductivity includes nodes labeled A-G, and its layout is shown in FIG. 4.
  • the overall average conductivity of the scale-free model is calculated to simulate the risk transmission probability of the entire scale-free model in a dynamic equilibrium state, which is used to provide users with ideal, A quantifiable average risk transmission probability of the full model.
  • the random forest model is used to identify the correlation between interconnected nodes in the directed graph, and based on this The correlation degree judges whether the risk will be transmitted between the two associated nodes, and the risk transmission method is described in a true and accurate manner. The goal is to identify the correlation degree between the two nodes.
  • the identifying the infected node in the scale-free model includes:
  • a blacklist with unit names in the blacklist; compare the names of each node in the scale-free model with those in the blacklist, and if the name of the node matches a unit in the blacklist If the names are consistent, it is determined that the node belongs to the blacklist.
  • S202 Set nodes belonging to the blacklist as infected nodes.
  • the nodes belonging to the blacklist in the scale-free model are set as infectious nodes.
  • the calculation of the risk conduction rate of each path in the scale-free model according to the infected node and the risk conduction coefficient includes:
  • S211 Identify a continuous path continuously associated with the infected node in the scale-free model, and sequentially number the sub-paths of the continuous path starting from the infected node; wherein the continuous path is Refers to the entire path of the associated nodes connected in series with the infected node in the scale-free model, and the sub path is the path between two adjacent nodes in the continuous path.
  • the continuous path in the aforementioned scale-free model includes: BDF and BDG; the sub-paths of BDF include BD and DF, and the sub-paths of BDG include BD and DG; because the infected node of the continuous path is Node B, so take Node B as the starting position; in the BDF continuous path, the number of the sub-path BD is 01, and the number of the sub-path DF is 02; in the BDG continuous path, the number of the sub-path BD is set to 11. Set the number of the sub-path DG to 12.
  • S212 Set any sub-path on the continuous path as a path to be calculated, identify the sub-path with a number smaller than the path to be calculated, extract the risk transmission coefficients therein, and then summarize them to form a coefficient set, and combine the risks of the coefficients
  • the conduction coefficient is multiplied by the risk conduction coefficient of the path to be calculated to obtain the true conduction coefficient, and it is loaded on the path to be calculated.
  • the path to be calculated is a sub-path BD, and its risk transmission coefficient is 0.2, because the sub-path does not have a sub-path with a number less than the path to be calculated in the continuous path BDF and BDG, therefore,
  • the true conductivity coefficient of the path to be calculated is set to 0.2 and loaded on the path to be calculated BD;
  • the path to be calculated is DF and its risk conduction coefficient is 0.4
  • the risk conduction coefficient in the sub-path BD is extracted and the coefficient set (0.2 ), multiply the risk transmission coefficient of 0.2 with the coefficient concentration and 0.4 of the risk transmission coefficient of the path to be calculated to obtain a risk transmission ratio of 0.08, and load it on the path to be calculated DF;
  • the path to be calculated is DG and its risk conduction coefficient is 0.5
  • the risk conduction coefficient in the sub-path BD is extracted and the coefficient set (0.2 )
  • the risk transmission rate of each path in the scale-free model is calculated, which is used to express the risk transmission rate of the risk from the infected node to other nodes, because of the risk transmission
  • the rate is obtained based on the average conductivity and the risk transmission coefficient, so it can most truly reflect the probability of risk transmission between the two related nodes.
  • the blacklist system is adopted to identify infected nodes, and the risk transmission rate is loaded on the corresponding path of the scale-free model to obtain the risk infection model, so as to quickly identify the incoming risk of any node in the risk infection model And outgoing risk.
  • the extracting and using the node in the risk infection model as the target node according to the node request sent by the user terminal includes:
  • S301 Receive a node request with a node name sent by the user terminal.
  • the node name in this step can be the company name or the company number.
  • S302 Compare the node names of all nodes in the risk infection model with the node request, and extract a node that matches the node request, and use the node as a target node.
  • the node request sent by the user terminal has a node name D
  • it will be compared with the nodes A-G in the scale-free model in turn, and the node D is obtained as a node matching the node name D.
  • the calculating the risk transmission rate in the incoming direction and the risk transmission rate in the outgoing direction of the target node to obtain the incoming risk rate and the outgoing risk rate includes:
  • S311 Set a node matching the node request as a target node, and extract a path connected to the target node in the scale-free model;
  • S312 Set the path pointing to the target node as an incoming path, and set the path pointed out from the target node as an outgoing path;
  • the path in this step has a direction.
  • the target node is set as the incoming path, which is used to describe the situation where the risk is passed from the outside to the target node; the path pointed out from the target node is set as the outgoing path, which is used to describe the risk The situation transmitted from the target node.
  • S313 Calculate the risk transmission rate of the incoming path of the target node by a weighted adjustment formula to obtain the incoming risk rate, and calculate the risk transmission rate of the outgoing path of the target node by the weighted adjustment formula to obtain the outgoing risk Rate.
  • the weighting adjustment formula is
  • T is the incoming risk rate or the outgoing risk rate, where the formula (1) in T is used to calculate the incoming risk rate, and the formula (2) is used to calculate the outgoing risk rate;
  • x is the risk transmission rate of the incoming path
  • a is the incoming coefficient
  • m is the incoming adjustment value
  • the incoming coefficient and the incoming adjustment value can be adjusted according to the needs of users
  • y is the risk transmission rate of the outgoing path
  • b Is the outgoing coefficient
  • n is the outgoing adjustment value
  • the outgoing coefficient and outgoing adjustment value can be adjusted according to the user's needs.
  • the paths connected to node D in the scale-free model which are BD, DF, and DG
  • the layout diagram of the risk infection model shows that BD is the incoming path, and DF and DG are transmission paths.
  • the risk characteristics of the target node can be comprehensively evaluated, so that the risk environment of the target node can be fully communicated to the user, which is conducive to the user's judgment based on this link.
  • inputting the incoming risk rate and outgoing risk rate into a preset four-quadrant model to obtain a risk point includes:
  • S401 Enter the incoming risk rate and the outgoing risk rate into a preset four-quadrant model
  • the four-quadrant model is a coordinate system that describes the risk characteristics of the node based on the incoming risk rate and the outgoing risk rate of the node.
  • the origin of the coordinate system is the incoming risk rate and the outgoing risk rate, respectively.
  • the incoming risk rate increases along the extending direction of the vertical axis of the above-mentioned coordinate system
  • the outgoing risk rate increases along the extending direction of the horizontal axis of the above-mentioned coordinate system, as specifically shown in FIG. 7.
  • the coordinate system is divided into four parts by crossing the abscissa dividing line and the ordinate dividing line, which are used to express both the incoming risk rate and the outgoing risk rate.
  • Silent glacier area used to express the "expanse ocean” area with high incoming risk rate but low outgoing risk rate, and used to express "active volcano” area with low incoming risk rate but high outgoing risk rate , And used to express the "storm center” area where both the incoming risk rate and the outgoing risk rate are high.
  • the abscissa dividing line and the ordinate dividing line of the above four parts can be adjusted according to user requirements.
  • the incoming risk rate of 0.2 and the outgoing risk rate of 0.18 are respectively entered into the ordinate and abscissa of the four-quadrant model to obtain risk points.
  • the area where the risk point is located is used to describe the risk characteristics of the node corresponding to the node request sent by the user end, and the name of the area is used as the judgment result; for example, based on the above example, the horizontal axis dividing line and the vertical axis The intersection of the coordinate dividing lines is (0.5, 0.5), and it is identified that the risk point is located in the "silent glacier" area, so the "silent glacier" is used as the judgment result.
  • the four-quadrant model is used to evaluate the risk characteristics of the target node, and the characteristics are output to the user terminal in the form of a name or icon, so that the user can quickly learn the impact of the surrounding nodes on the target node, and the target node The technical effect of the risk environment on the impact of surrounding nodes.
  • a data analysis device 1 based on big data in this embodiment includes:
  • the creation server 11 is used to create a directed graph for describing the association relationship between nodes and asset relationships, and calculate the risk transmission coefficient of each path in the directed graph through an infectious disease model to obtain a scale-free model, and Send the risk server; wherein, the node refers to the information owner, the association relationship is used to reflect the involvement and influence between the information owners, and the asset relationship is used to reflect the asset association ratio between the information owners ;
  • the risk server 12 is configured to identify infected nodes in the scale-free model, and calculate the risk transmission rate of each path in the scale-free model according to the infected nodes and in combination with the risk transmission coefficient, so as to obtain the risk infection model and It sends the computing server;
  • the calculation server 13 is configured to extract the node in the risk infection model and use it as the target node according to the node request sent by the user terminal, and calculate the risk transmission rate of the target node in the incoming direction and the risk transmission rate in the outgoing direction. Obtain the incoming risk rate and the outgoing risk rate; wherein the node request includes a node name corresponding to the node in the risk infection model, which is used to extract the node in the risk infection model.
  • the data analysis device 1 further includes:
  • the judgment server 14 is configured to input the incoming risk rate and the outgoing risk rate into a preset four-quadrant model to obtain a risk point, identify the area where the risk point is located, and use the name of the area as the judgment result, and use the judgment The result is output to the user terminal.
  • this application also provides a computer system, which includes a plurality of computer devices 5.
  • the components of the data analysis apparatus 1 of the second embodiment can be dispersed in different computer devices, and the computer devices can be executable programs.
  • the computer equipment of this embodiment at least includes but is not limited to: a memory 21 and a processor 22 that can be communicatively connected to each other through a system bus, as shown in FIG. 9.
  • FIG. 9 only shows a computer device with components, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.
  • the memory 21 (ie, readable storage medium) includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), Read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disk, optical disk, etc.
  • the memory 21 may be an internal storage unit of a computer device, such as a hard disk or memory of the computer device.
  • the memory 21 may also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a smart memory card (Smart Media Card, SMC), and Secure Digital (SD).
  • the memory 21 may also include both an internal storage unit of the computer device and an external storage device thereof.
  • the memory 21 is generally used to store an operating system and various application software installed in a computer device, such as the program code of the data analysis device in the first embodiment.
  • the memory 21 can also be used to temporarily store various types of data that have been output or will be output.
  • the processor 22 may be a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, a microprocessor, or other data processing chips.
  • the processor 22 is generally used to control the overall operation of the computer equipment.
  • the processor 22 is used to run the program code or process data stored in the memory 21, for example, to run a data analysis device, so as to implement the data analysis method of the first embodiment.
  • this application also provides a computer-readable storage system, which includes multiple storage media.
  • the storage media may be non-volatile or volatile, such as flash memory, hard disk, multimedia card, and card.
  • Type memory for example, SD or DX memory, etc.
  • RAM random access memory
  • SRAM static random access memory
  • ROM read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • PROM programmable only
  • the read memory (PROM), magnetic memory, magnetic disk, optical disk, server, App application store, etc. have computer programs stored thereon, and the programs are executed by the processor 22 to realize corresponding functions.
  • the computer-readable storage medium of this embodiment is used to store a data analysis device, and when executed by the processor 22, the data analysis method of the first embodiment is implemented.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Animal Behavior & Ethology (AREA)
  • Game Theory and Decision Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请公开了一种数据分析方法、装置、计算机系统及可读存储介质，基于大数据领域，包括：创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算有向图中各路径的风险传导系数以获得无标度模型；识别无标度模型中的感染节点，根据感染节点并结合风险传导系数计算无标度模型中各路径的风险传导率，以获得风险感染模型；根据用户端发送的节点请求提取风险感染模型中的节点并将其作为目标节点，计算目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率。本申请实现了对知识图谱中信息的深度挖掘，并提供深层次且有价值的信息的技术效果。

Description

一种数据分析方法、装置、计算机系统及可读存储介质
本申请要求于2020年3月5日提交中国专利局、申请号为CN 202010146003.1，发明名称为“一种数据分析方法、装置、计算机系统及可读存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，其涉及到大数据的数据挖掘技术，尤其涉及一种数据分析方法、装置、计算机系统及可读存储介质。
背景技术
当企业风险管理逐渐成为业内焦点，市面上出现了许多企业分析及风险监控类产品，如企查查、天眼查等，但是当前企业风险分析聚焦于企业自身各项风险，如工商信息、财务报表、法律诉讼、舆情风险等，然而当前的风险监控类产品却只能获知法律文件、账面数据以及风险评级等信息，并且这些信息通常采用知识图谱进行管理及展示。
然而当前的知识图谱仅说明了节点数据及各节点之间的关联关系，但是仅凭这些信息仅仅能够提供节点基本信息，以及节点之间的关联信息，而未对知识图谱中的信息进行深度挖掘，导致其仅能够提供简单的信息，而无法向使用者提供可直接使用的有价值的信息。
发明内容
本申请的目的是提供一种数据分析方法、装置、计算机系统及可读存储介质，用于解决现有技术存在的未对知识图谱中的信息进行深度挖掘，导致其仅能够提供简单的信息，而无法向使用者提供可直接使用的有价值的信息的技术问题。
为实现上述目的，本申请提供一种基于大数据的数据分析方法，包括：
创建服务器创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
所述风险服务器识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
所述计算服务器根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
为实现上述目的，本申请还提供一种基于大数据的数据分析装置，包括：
创建服务器，用于创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
风险服务器，用于识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
计算服务器，用于根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
为实现上述目的，本申请还提供一种计算机系统，其包括多个计算机设备，各计算机 设备包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序，所述多个计算机设备的处理器执行所述计算机程序时共同实现上述数据分析方法的步骤。
为实现上述目的，本申请还提供一种计算机可读存储介质，其包括多个存储介质，各存储介质上存储有计算机程序，所述多个存储介质存储的所述计算机程序被处理器执行时共同实现上述数据分析方法的步骤。
本申请提供的一种数据分析方法、装置、计算机系统及可读存储介质，通过创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，以描述各节点之间的关联程度；通过识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型；因此根据感染节点，并通过风险传导系数和平均传导率，计算无标度模型中各路径的风险传导率，用于表达风险从感染节点传导至其他节点的风险传导率，因该风险传导率是基于平均传导率和风险传导系数所获得，因此能够最真实的反应出相关的两个节点之间风险传导的概率。根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；实现了全面的获知目标节点的传出风险率和传入风险率，以全面评价目标节点的风险特征，以便于向用户全面传递目标节点的风险环境，有利于用户根据该环节进行判断。通过四象限模型计算所述传入风险率和传出风险率以获得判断结果，并将其输出至所述用户端，实现了采用四象限模型评价目标节点的风险特征，并将该特征以名称或图标的方式输出至用户端，以便于用户快速获知用于描述周围节点对目标节点影响，以及目标节点对周围节点的影响的风险环境，实现了对知识图谱中信息的深度挖掘，进而解决了现有技术中因未对知识图谱中的信息进行深度挖掘，导致其仅能够提供简单的信息，而无法向使用者提供可直接使用的有价值的信息的技术问题。
附图说明
图1为本申请数据分析方法实施例一的流程图；
图2为本申请数据分析方法实施例一中欧拉图谱的数据关联方式的有向图；
图3为本申请数据分析方法实施例一中欧拉图谱的数据关联方式的树状图；
图4为本申请数据分析方法实施例一中具有初级传导系数的有向图；
图5为本申请数据分析方法实施例一中将风险传导系数其加载在各路径上的具有初级传导系数的有向图；
图6为本申请数据分析方法实施例一中风险感染模型的有向图；
图7为本申请数据分析方法实施例一中四象限模型的示意图；
图8为本申请数据分析装置实施例二的程序模块示意图；
图9为本申请计算机系统实施例三中计算机设备的硬件结构示意图。
附图标记：
1、数据分析装置 2、计算机设备 11、创建服务器
12、风险服务器 13、计算服务器 14、判断服务器
21、存储器 22、处理器
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本申请，并不用于限定本申请。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
本申请提供的一种数据分析方法、装置、计算机系统及可读存储介质，适用于计算机领域，为提供一种基于大数据创建模块、感染模型创建模块、风险计算模块和风险判断模块的数据分析方法。本申请通过创建用于描述节点间关联关系及资产关系的有向图，并通 过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型；识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型；根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；通过四象限模型计算所述传入风险率和传出风险率以获得判断结果，并将其输出至所述用户端。
实施例一
请参阅图1，本实施例的一种基于大数据的数据分析方法，包括：
S1：创建服务器创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
S2：所述风险服务器识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
S3：所述计算服务器根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
在一个优选的实施例中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率之后包括：
S4：将所述传入风险率和传出风险率录入预设的四象限模型获得风险点，识别所述风险点所在区域并将该区域的名称作为判断结果，将所述判断结果输出至用户端。
在示例性的实施例中，通过创建有向图以表达节点之间的关联关系和资产关系，其中，节点可以为企业，也可以为自然人，因此，可将企业或自然人的名称或姓名作为节点的名称；通过传染病模型计算有向图在均衡状态下(即：受到感染节点影响的节点数与摆脱感染节点影响的节点数相等)时的平均传导概率，根据所述平均传导概率计算有向图中各路径的风险传导概率，并将其加载于各路径上，以获得无标度模型；其中，所述无标度模型是以反应了各节点之间关联关系和资产关系的有向图为基础，通过传染病模型对有向图中各路径风险传导概率进行计算并加载于各路径上的数据模型。
需要说明的是，有向图在数学上是表示物件与物件之间的关系的方法，其由一些小圆点(称为顶点或结点)和连结这些圆点的直线或曲线(称为边)组成的；于本实施例中，所述圆点则对应本申请中的节点，所述连接这些圆点的直线或曲线，所述连接圆点的直线或曲线的箭头及加载于所述直线或曲线上的信息对应所述资产关系。传染病模型是一种用于计算节点影响力的标准测量模型，所述传染病模型中的节点有三个状态：易感者(susceptible)，感染者(infected)，恢复者(recovered)。易感者能够被感染者染病，感染者染病且有感染易感者的能力，恢复者由感染者恢复而成，不再具备感染能力同时也不会再被感染。传染病模型参数有恢复率μ，传染概率β，重复仿真次数T，仿真时间timespace。仿真过程如下：选择网络中一个节点i作为感染者节点，以传染概率β向i连接的邻居易感者节点传染病毒，被染病的节点继续以β的概率向他们的邻居易感者节点传染。同时每个染病节点在每阶段以μ的概率转变成恢复节点。
可采用黑名单制度识别所述无标度模型中的感染节点，即：若某一节点属于黑名单则判定该节点为感染节点；设定感染节点的风险为100％，将与感染节点通过路径连接的节点作为直接连接的节点，根据该路径上的风险传导系数计算风险从感染节点传导至所述直接连接的节点的风险传导率；再识别与所述连接的节点相连的节点及其连接的路径，并将所述相连的节点设为间接连接的节点，通过所述连接的路径上的风险传导系数计算风险从感 染节点经直接连接的节点传导至所述间接连接的节点的风险传导率，依次类推，直至将所述无标度模型中所有与感染节点直接连接和间接连接的节点计算完成后，获得风险感染模型。
所述节点请求中包括有用于对应风险感染模型中节点的节点名称，将所述风险感染模型中具有所述节点名称的节点作为目标节点，识别风险感染模型中的路径，将指向目标节点的路径作为传入路径，将从目标节点指出的路径作为传出路径，根据所述传入路径的风险传导率获得传入方向的传入风险率，根据所述传出路径的风险传导率获得传出方向的传出风险率。
用户可在四象限模型的坐标系中设置区域，并对该区域赋以名称；根据传入风险率和传出风险率在四象限模型中获得风险点，将风险点所在的区域的名称作为判断结果，将所述判断结果输出至用户端。
在一个优选的实施例中，所述创建用于描述节点间关联关系及资产关系的有向图包括：
S101：从服务系统中获取节点数据以及各节点之间的关联关系，根据所述关联关系构建用于描述节点间关联关系的有向图。
示例性地，从服务系统中获取知识图谱，从所述知识图谱中获取节点数据以及各节点之间的关联关系，若两个节点数据之间具有关联关系，则在所述两个节点数据所对应的节点之间绘制路径，若节点A对节点B为投资关系，则节点A与节点B之间的路径为A指向B；其中，因所述知识图谱中的路径具有箭头，故可从知识图谱中直接获知节点A与节点B之间的关系，故在本申请中不做赘述。
需要说明的是，所述服务系统是储存有大量企业和个人数据、并基于企业和个人数据建立有大量的企业关系数据、股权关系数据、诉讼关系数据等，所述服务系统可覆盖工商、财经公告、法律文书、社交媒体和海外舆情等各大维度；于本申请中所述服务系统采用的是欧拉图谱，其为一种用于储存企业和个人数据，并根据企业和个人数据构建包含关联人物、关联企业的企业关联网络，从股东持股关系、对外投资关系、供应链关系、股权出质关系、融资担保关系和企业高管六大方面的数据关系的知识图谱；本申请所要解决的技术问题是如何深度挖掘当前的知识图谱(本申请中的欧拉图谱)的知识关系以获取潜在应用价值的信息；而对于根据知识图谱(本申请中的欧拉图谱)获得节点数据及各节点之间的关联关系，本领域技术人员很容易通过提取的欧拉图谱获得，故在此不做赘述。因该系统为现有技术，故在此不做赘述。
其中，所述欧拉图谱的数据关联方式如图2和图3所示。
S102：计算所述有向图中相互关联的节点之间的资产关联比值，并将其加载于所述相互关联节点之间的路径上以描述所述有向图中节点间资产关系。
示例性地，提取相互关联的两个节点的节点数据，获取所述节点数据中的被投资数据和投资数据，其中，将所述两个节点中发出箭头(路径)的一方作为投资节点，将箭头(路径)指向的一方作为被投资节点，提取投资节点中的投资数据，以及被投资节点中与所述投资数据关联的被投资数据，将所述被投资数据与投资数据相除获得资产关联比值，并将所述资产关联比值加载在所述路径上。
其中，所述相互关联的两个节点的节点数据是描述其中一个节点向另一节点进行投资的信息，故所述两个节点的节点数据中，必然一个为投资节点的节点数据，另一个为被投资节点的节点数据。需要说明的是，所述资产关联比值是指被投资节点接收到的金额与投资节点对外投资总金额之比；所述与投资数据关联的被投资数据是指，因投资数据对被投资节点进行投资所生成的被投资数据；例如，投资节点的投资数据为100万，被投资节点的被投资数据为50万，而其中，被投资数据中20万为基于所述投资节点的投资数据所得，其他30万是基于其他投资节点的投资数据所得，那么所述投资节点和被投资节点之间的资产关联比值为20％。
在一个优选的实施例中，所述通过传染病模型计算所述有向图中各路径的风险传导系 数以获得无标度模型包括：
S111：通过传染病模型计算所述有向图的模型指标以获得平均传导概率。
本步骤中，所述传染病模型中具有均衡条件设定，通过所述传染病模型计算所述有向图的模型指标在均衡条件下的平均传导率。
需要说明的是，传染病模型是一种用于计算节点影响力的标准测量模型，基于传染病模型的原理，风险在有向图中的传递过程是易感者收到风险节点的感染会成为感染者，感染者会恢复并成为恢复者，而恢复者也会因再次收到风险节点的感染而再次成为感染者。
因此，通过传染病模型计算有向图的模型指标，获得所述有向图中易感者、感染者和恢复者达到均衡条件下时，有向图中所有节点的总体传染概率，这个总体传染概率是指能够保持有向图均衡条件的风险传导概率。
具体地，提取所述有向图的模型指标中的节点总数、节点度、平均度和平均密度，并将其录入所述传染病模型的目标函数中进行计算，以获得平均传导概率。
其中，所述目标函数如下所示：
Figure PCTCN2020093201-appb-000001
该方法首先做出如下定义：
K是指传染病模型中个体总数，对应于节点总数；风险个体记为I，健康个体记为S；S(t)为健康个体在时间t时的数量；ρ(t)是在时间t时的健康个体比例；kθ是单位时间内一个风险个体能传染的数量与当时健康个体总数的比例，对应于所述节点度与平均度的乘积；η是指传染病模型的模型平均密度，对应于所述平均密度；风险个体通过概率γ将风险传出，并且风险个体会恢复健康但不会免疫；
通过对上述偏微分方程的动态求解得到均衡状态下(即：当期新染病的人数与当期新恢复健康的人数相等，于本实施例中为受到感染节点影响的节点数与摆脱感染节点影响的节点数相等)的总体传染概率γ，将所述总体传染概率γ设为平均传导概率。
需要说明的是，商业环境传染病模型的目标函数只有在均衡状态下，医学传染病模型即动态模拟人群中染病人数的变化，得到在均衡状态下群体的传染比例。
S112：将所述平均传导概率分别与所述有向图的路径上的资产关联比值相乘以获得初级传导系数，通过随机森林模型识别所述有向图中各相互关联节点之间的相关度，并将其加载于所述相互关联的节点的路径上；
示例性地，提取所述有向图各路径上的资产关联比值，并将其与平均传导概率相乘以获得初级传导系数，通过所述随机森林模型提取任一相互关联的两个节点的节点数据，并识别其之间的关联性；
若判断所述两个节点的节点数据相互关联，则在所述两个节点之间的路径赋以值为1的相关度；若判断所述两个节点的节点数据不关联，则在所述两个节点之间的路径赋以值为0的相关度；将所述风险传导系数加载于所述有向图的路径上。
需要说明的是，随机森林是一个包含多个决策树的分类器，并且其输出的类别是由个别树输出的类别的众数而定的模型；因随机森林里面有很多的决策树，且每一棵决策树之间是没有关联的，因此在得到森林之后，当有一个新的输入样本进入的时候，就让森林中的每一棵决策树分别进行一下判断，看看这个样本应该属于哪一类(对于分类算法)，然后看看哪一类被选择最多，就预测这个样本为那一类。
所述随机森林通过以下方式训练获得：
1)准备N个样本，则有放回的随机选择N个样本(每次随机选择一个样本，然后返回继续选择)。这选择好了的N个样本用来训练一个决策树，作为决策树根节点处的样本。
2)当每个样本有M个属性时，在决策树的每个节点需要分裂时，随机从这M个属性中选取出m个属性，满足条件m<<M。然后从这m个属性中采用分类策略(比如说信息增益) 来选择1个属性作为该节点的分裂属性。
3)决策树形成过程中每个节点都要按照步骤2来分裂，一直到不能够再分裂为止。
4)按照步骤1～3建立大量的决策树，这样就构成了随机森林了。
于本实施例中，用户通过设定所述m个属性以及分类策略并通过上述方法来训练初始随机森林，以获得用于识别相互关联的两个节点的节点数据，是否具有关联性的随机森林模型；因随机森林的训练方法属于本领域技术人员的公职常识，因此通过设定m个属性及分类策略训练随机森林的训练过程在此不做赘述。
S113：将所述有向图的路径上的初级传导系数和相关度相乘以获得风险传导系数，将所述风险传导系数加载于所述有向图的路径上以获得无标度模型。
例如：所述具有初级传导系数的有向图中包括标号为A—G的节点，其布局如图4所示。
若通过随机森林模型识别所述有向图的结果是：除节点C与节点E的相关度为0，其他的相互关联的节点之间的相关度均为1；则将各路径上的初级传导系数及其相关度相乘以获得风险传导系数，并将其加载在各路径上，如图5所示。
综上，基于所述传染病模型在均衡状态下计算无标度模型整体的平均传导率，以模拟整个无标度模型在动态平衡的状态下风险的传导概率，用于向用户提供理想的，能够量化的全模型平均风险传导概率。同时，由于有些相互关联的节点之间会因为某些因素而不发生风险传导，例如，有限责任等，因此利用随机森林模型识别有向图中相互关联的节点之间的相关度，并根据该相关度判断该风险是否会从关联的两个节点之间传递，以真实准确的描述风险传递方式，其目标在于识别两个节点之间的相关度。
在一个优选的实施例中，所述识别所述无标度模型中的感染节点包括：
S201：判断所述无标度模型的节点是否属于预设的黑名单；
示例性地，创建黑名单，所述黑名单中具有单位名称；将无标度模型中各节点的名称与所述黑名单的进行比对，若节点的名称与所述黑名单中某一单位名称一致，则判定该节点属于黑名单。
S202：将属于所述黑名单的节点设为感染节点。
示例性地，将所述无标度模型中属于黑名单的节点设为传染节点。
在一个优选的实施例中，所述根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率包括：
S211：在所述无标度模型中识别与所述感染节点连续关联的连续路径，并以所述感染节点为起始依次对所述连续路径的子路径进行编号；其中，所述连续路径是指在所述无标度模型中以感染节点为输出源，依次与其串联的关联节点的整条路径，所述子路径为所述连续路径中相邻的两个节点之间的路径。
例如：若节点B为感染节点，上述无标度模型中的连续路径包括：B-D-F和B-D-G；B-D-F的子路径包括B-D和D-F，B-D-G的子路径包括B-D和D-G；因该连续路径的感染节点为节点B，因此以节点B为起始位置；在B-D-F连续路径中，将子路径B-D的编号为01，将子路径D-F的编号为02；在B-D-G连续路径中，将子路径B-D的编号设为11，将子路径D-G的编号设为12。
S212：将所述连续路径上的任一子路径设为待算路径，识别编号小于所述待算路径的子路径并提取其中的风险传导系数后汇总形成系数集，将所述系数集中的风险传导系数与所述待算路径的风险传导系数相乘获得真实传导系数，并将其加载于所述待算路径上。
例如，基于上述举例，若待算路径为子路径B-D，其风险传导系数为0.2时，因该子路径在连续路径B-D-F和B-D-G中均不具有编号小于该待算路径的子路径，因此，将待算路径的真实传导系数设为0.2并将其加载于待算路径B-D上；
若待算路径为D-F，其风险传导系数为0.4时，则在连续路径B-D-F中具有编号小于该待算路径的子路径B-D，因此提取该子路径B-D中的风险传导系数并形成系数集(0.2)，将系数集中的风险传导系数0.2与待算路径的风险传导系数0.4相乘，获得风险传导率 0.08，并将其加载于待算路径D-F上；
若待算路径为D-G，其风险传导系数为0.5时，则在连续路径B-D-G中具有编号小于该待算路径的子路径B-D，因此提取该子路径B-D中的风险传导系数并形成系数集(0.2)，将系数集中的风险传导系数0.2与待算路径的风险传导系数0.5相乘，获得风险传导率0.1，并将其加载于待算路径D-G上；
经上述方法所获得的风险感染模型如图6所示。
综上，根据感染节点，并通过风险传导系数和平均传导率，计算无标度模型中各路径的风险传导率，用于表达风险从感染节点传导至其他节点的风险传导率，因该风险传导率是基于平均传导率和风险传导系数所获得，因此能够最真实的反应出相关的两个节点之间风险传导的概率。
同时，采用黑名单制度识别感染节点，并将风险传导率加载于其对应的所述无标度模型的路径上，以获得风险感染模型，以便于快速识别风险感染模型中任意节点的传入风险和传出风险。
在一个优选的实施例中，所述根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点包括：
S301：接收用户端发送的具有节点名称的节点请求。
本步骤中的节点名称可为企业名称，也可为企业编号。
S302：将所述风险感染模型中所有节点的节点名称与所述节点请求进行对比，并提取与所述节点请求匹配的节点，并将所述节点作为目标节点。
例如，基于上述举例，用户端发送的节点请求具有节点名称D，则将其与无标度模型中的节点A-G依次比对，获得节点D为与节点名称D相匹配的节点。
在一个优选的实施例中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率包括：
S311：将与所述节点请求匹配的节点设为目标节点，在所述无标度模型中提取与所述目标节点连接的路径；
S312：将指向所述目标节点的路径设为传入路径，将从所述目标节点指出的路径设为传出路径；
本步骤中的路径具有方向，将指向目标节点设为传入路径，用于描述风险从外界传入目标节点的情况；将从所述目标节点指出的路径设为传出路径，用于描述风险从目标节点传出的情况。
S313：通过加权调整公式计算所述目标节点的传入路径的风险传导率，以获得传入风险率，通过加权调整公式计算所述目标节点的传出路径的风险传导率，以获得传出风险率。
示例性地，加权调整公式为
Figure PCTCN2020093201-appb-000002
其中，T为传入风险率或传出风险率，其中T中的公式(1)用于计算传入风险率，公式(2)用于计算传出风险率；
x为传入路径的风险传导率，a为传入系数，m为传入调整值，传入系数和传入调整值可根据用户的需求自行调节；y为传出路径的风险传导率，b为传出系数，n为传出调整值，传出系数和传出调整值可根据用户的需求自行调节。
例如，基于上述举例，在无标度模型中提取与节点D连接的路径，其分别为B-D，D-F和D-G，且由风险感染模型的布局图可知：B-D为传入路径，D-F和D-G为传出路径；假设加权调整公式中传入系数和传出系数均为1，传入调整值和传出调整值均为0，则通过加权调整公式计算传入路径B-D的风险传导率，获得传入风险率0.2，通过加权调整公式计算传出路径D-F和D-G的风险传导率，获得传出风险率0.1+0.08＝0.18。
综上，通过全面的获知目标节点的传出风险率和传入风险率，以全面评价目标节点的风险特征，以便于向用户全面传递目标节点的风险环境，有利于用户根据该环节进行判断。
在一个优选的实施例中，将所述传入风险率和传出风险率录入预设的四象限模型获得风险点，识别所述风险点所在区域并将该区域的名称作为判断结果，包括：
S401：将所述传入风险率和传出风险率录入预设的四象限模型；
本步骤中，所述四象限模型是一种基于节点的传入风险率和传出风险率描述该节点风险特征的坐标系，该坐标系的原点为传入风险率和传出风险率分别为0的坐标点，所述传入风险率沿着上述坐标系纵轴的延伸方向增加，所述传出风险率沿着上述坐标系的横轴延伸方向增加，具体如图7所示。
示例性地，在四象限模型中，通过横坐标分界线和纵坐标分界线相互交叉将坐标系分为四个部分，分别为用于表达传入风险率和传出风险率均较低的“沉寂冰川”区域，用于表达传入风险率较高但传出风险率较低的“浩瀚海洋”区域，用于表达传入风险率较低但传出风险率较高的“活跃火山”区域，以及用于表达传入风险率和传出风险率均较高的“风暴中心”区域。进一步的，上述四个部分的横坐标分界线和纵坐标分界线可根据用户需求自行调节。
S402：在所述四象限模型中以所述传入风险率作为纵坐标，并以所述传出风险率作为横坐标获得风险点；
例如，基于上述举例，将所述传入风险率0.2和传出风险率0.18分别录入四象限模型的纵坐标和横坐标，以获得风险点。
S403：识别所述风险点所在的区域并将该区域的名称作为判断结果。
本步骤中，该风险点所在的区域，即为用于描述用户端发送的节点请求所对应节点的风险特征，将该区域的名称作为判断结果；例如，基于上述举例，横坐标分界线和纵坐标分界线的交点为(0.5,0.5)，识别出所述风险点位于“沉寂冰川”区域，因此将“沉寂冰川”作为判断结果。
综上，实现了通过采用四象限模型评价目标节点的风险特征，并将该特征以名称或图标的方式输出至用户端，以便于用户快速获知用于描述周围节点对目标节点影响，以及目标节点对周围节点的影响的风险环境的技术效果。
实施例二
请参阅图8，本实施例的一种基于大数据的数据分析装置1，包括：
创建服务器11，用于创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
风险服务器12，用于识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
计算服务器13，用于根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
可选的，所述数据分析装置1还包括：
判断服务器14，用于将所述传入风险率和传出风险率录入预设的四象限模型获得风险点，识别所述风险点所在区域并将该区域的名称作为判断结果，将所述判断结果输出至用户端。
实施例三
为实现上述目的，本申请还提供一种计算机系统，该计算机系统包括多个计算机设备5， 实施例二的数据分析装置1的组成部分可分散于不同的计算机设备中，计算机设备可以是执行程序的智能手机、平板电脑、笔记本电脑、台式计算机、机架式服务器、刀片式服务器、塔式服务器或机柜式服务器(包括独立的服务器，或者多个服务器所组成的服务器集群)等。本实施例的计算机设备至少包括但不限于：可通过系统总线相互通信连接的存储器21、处理器22，如图9所示。需要指出的是，图9仅示出了具有组件-的计算机设备，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。
本实施例中，存储器21(即可读存储介质)包括闪存、硬盘、多媒体卡、卡型存储器(例如，SD或DX存储器等)、随机访问存储器(RAM)、静态随机访问存储器(SRAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、可编程只读存储器(PROM)、磁性存储器、磁盘、光盘等。在一些实施例中，存储器21可以是计算机设备的内部存储单元，例如该计算机设备的硬盘或内存。在另一些实施例中，存储器21也可以是计算机设备的外部存储设备，例如该计算机设备上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。当然，存储器21还可以既包括计算机设备的内部存储单元也包括其外部存储设备。本实施例中，存储器21通常用于存储安装于计算机设备的操作系统和各类应用软件，例如实施例一的数据分析装置的程序代码等。此外，存储器21还可以用于暂时地存储已经输出或者将要输出的各类数据。
处理器22在一些实施例中可以是中央处理器(Central Processing Unit，CPU)、控制器、微控制器、微处理器、或其他数据处理芯片。该处理器22通常用于控制计算机设备的总体操作。本实施例中，处理器22用于运行存储器21中存储的程序代码或者处理数据，例如运行数据分析装置，以实现实施例一的数据分析方法。
实施例四
为实现上述目的，本申请还提供一种计算机可读存储系统，其包括多个存储介质，所述存储介质可以是非易失性，也可以是易失性，如闪存、硬盘、多媒体卡、卡型存储器(例如，SD或DX存储器等)、随机访问存储器(RAM)、静态随机访问存储器(SRAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、可编程只读存储器(PROM)、磁性存储器、磁盘、光盘、服务器、App应用商城等等，其上存储有计算机程序，程序被处理器22执行时实现相应功能。本实施例的计算机可读存储介质用于存储数据分析装置，被处理器22执行时实现实施例一的数据分析方法。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。
以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种基于大数据的数据分析方法，其中，包括：
    创建服务器创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
    所述风险服务器识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
    所述计算服务器根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
  2. 根据权利要求1所述的数据分析方法，其中，所述创建用于描述节点间关联关系及资产关系的有向图包括：
    从服务系统中获取节点数据以及各节点之间的关联关系，根据所述关联关系构建用于描述节点间关联关系的有向图；
    计算所述有向图中相互关联的节点之间的资产关联比值，并将其加载于所述相互关联节点之间的路径上以描述所述有向图中节点间资产关系。
  3. 根据权利要求1所述的数据分析方法，其中，所述通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型包括：
    通过传染病模型计算所述有向图的模型指标以获得平均传导概率；
    将所述平均传导概率分别与所述有向图的路径上的资产关联比值相乘以获得初级传导系数，通过随机森林模型识别所述有向图中各相互关联节点之间的相关度，并将其加载于所述相互关联的节点的路径上；
    将所述有向图的路径上的初级传导系数和相关度相乘以获得风险传导系数，将所述风险传导系数加载于所述有向图的路径上以获得无标度模型。
  4. 根据权利要求1所述的数据分析方法，其中，所述识别所述无标度模型中的感染节点包括：
    判断所述无标度模型的节点是否属于预设的黑名单；
    将属于所述黑名单的节点设为感染节点。
  5. 根据权利要求1所述的数据分析方法，其中，所述根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率包括：
    在所述无标度模型中识别与所述感染节点连续关联的连续路径，并以所述感染节点为起始依次对所述连续路径的子路径进行编号；其中，所述连续路径是指在所述无标度模型中以感染节点为输出源，依次与其串联的关联节点的整条路径，所述子路径为所述连续路径中相邻的两个节点之间的路径；
    将所述连续路径上的任一子路径设为待算路径，识别编号小于所述待算路径的子路径并提取其中的风险传导系数后汇总形成系数集，将所述系数集中的风险传导系数与所述待算路径的风险传导系数相乘获得真实传导系数，并将其加载于所述待算路径上。
  6. 根据权利要求1所述的数据分析方法，其中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率包括：
    将与所述节点请求匹配的节点设为目标节点，在所述无标度模型中提取与所述目标节点连接的路径；
    将指向所述目标节点的路径设为传入路径，将从所述目标节点指出的路径设为传 出路径；
    通过加权调整公式计算所述目标节点的传入路径的风险传导率，以获得传入风险率，通过加权调整公式计算所述目标节点的传出路径的风险传导率，以获得传出风险率。
  7. 根据权利要求1所述的数据分析方法，其中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率之后包括：
    将所述传入风险率和传出风险率录入预设的四象限模型获得风险点，识别所述风险点所在区域并将该区域的名称作为判断结果，将所述判断结果输出至用户端。
  8. 一种基于大数据的数据分析装置，其中，包括：
    创建服务器，用于创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
    风险服务器，用于识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
    计算服务器，用于根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
  9. 一种计算机系统，其包括多个计算机设备，各计算机设备包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序，其中，所述多个计算机设备的处理器执行所述计算机程序时共同实现以下步骤：
    创建服务器创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
    所述风险服务器识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
    所述计算服务器根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
  10. 根据权利要求9所述的计算机系统，其中，所述创建用于描述节点间关联关系及资产关系的有向图包括：
    从服务系统中获取节点数据以及各节点之间的关联关系，根据所述关联关系构建用于描述节点间关联关系的有向图；
    计算所述有向图中相互关联的节点之间的资产关联比值，并将其加载于所述相互关联节点之间的路径上以描述所述有向图中节点间资产关系。
  11. 根据权利要求9所述的计算机系统，其中，所述通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型包括：
    通过传染病模型计算所述有向图的模型指标以获得平均传导概率；
    将所述平均传导概率分别与所述有向图的路径上的资产关联比值相乘以获得初级传导系数，通过随机森林模型识别所述有向图中各相互关联节点之间的相关度，并将其加载于所述相互关联的节点的路径上；
    将所述有向图的路径上的初级传导系数和相关度相乘以获得风险传导系数，将所述风险传导系数加载于所述有向图的路径上以获得无标度模型。
  12. 根据权利要求9所述的计算机系统，其中，所述识别所述无标度模型中的感染节点包括：
    判断所述无标度模型的节点是否属于预设的黑名单；
    将属于所述黑名单的节点设为感染节点；
    所述根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率包括：
    在所述无标度模型中识别与所述感染节点连续关联的连续路径，并以所述感染节点为起始依次对所述连续路径的子路径进行编号；其中，所述连续路径是指在所述无标度模型中以感染节点为输出源，依次与其串联的关联节点的整条路径，所述子路径为所述连续路径中相邻的两个节点之间的路径；
    将所述连续路径上的任一子路径设为待算路径，识别编号小于所述待算路径的子路径并提取其中的风险传导系数后汇总形成系数集，将所述系数集中的风险传导系数与所述待算路径的风险传导系数相乘获得真实传导系数，并将其加载于所述待算路径上。
  13. 根据权利要求9所述的计算机系统，其中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率包括：
    将与所述节点请求匹配的节点设为目标节点，在所述无标度模型中提取与所述目标节点连接的路径；
    将指向所述目标节点的路径设为传入路径，将从所述目标节点指出的路径设为传出路径；
    通过加权调整公式计算所述目标节点的传入路径的风险传导率，以获得传入风险率，通过加权调整公式计算所述目标节点的传出路径的风险传导率，以获得传出风险率。
  14. 根据权利要求9所述的计算机系统，其中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率之后包括：
    将所述传入风险率和传出风险率录入预设的四象限模型获得风险点，识别所述风险点所在区域并将该区域的名称作为判断结果，将所述判断结果输出至用户端。
  15. 一种计算机可读存储介质，其包括多个存储介质，各存储介质上存储有计算机程序，其中，所述多个存储介质存储的所述计算机程序被处理器执行时共同实现以下步骤：
    创建服务器创建用于描述节点间关联关系及资产关系的有向图，并通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型，并将其发送风险服务器；其中，所述节点是指信息所有者，所述关联关系用于反应各信息所有者之间的牵连和影响，所述资产关系用于反应各信息所有者之间的资产关联比值；
    所述风险服务器识别所述无标度模型中的感染节点，根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率，以获得风险感染模型并将其发送计算服务器；
    所述计算服务器根据用户端发送的节点请求提取所述风险感染模型中的节点并将其作为目标节点，计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率；其中，所述节点请求包括有与风险感染模型中节点对应的节点名称，其用于提取所述风险感染模型中的节点。
  16. 根据权利要求15所述的计算机可读存储介质，其中，所述创建用于描述节点间关联关系及资产关系的有向图包括：
    从服务系统中获取节点数据以及各节点之间的关联关系，根据所述关联关系构建 用于描述节点间关联关系的有向图；
    计算所述有向图中相互关联的节点之间的资产关联比值，并将其加载于所述相互关联节点之间的路径上以描述所述有向图中节点间资产关系。
  17. 根据权利要求15所述的计算机可读存储介质，其中，所述通过传染病模型计算所述有向图中各路径的风险传导系数以获得无标度模型包括：
    通过传染病模型计算所述有向图的模型指标以获得平均传导概率；
    将所述平均传导概率分别与所述有向图的路径上的资产关联比值相乘以获得初级传导系数，通过随机森林模型识别所述有向图中各相互关联节点之间的相关度，并将其加载于所述相互关联的节点的路径上；
    将所述有向图的路径上的初级传导系数和相关度相乘以获得风险传导系数，将所述风险传导系数加载于所述有向图的路径上以获得无标度模型。
  18. 根据权利要求15所述的计算机可读存储介质，其中，所述识别所述无标度模型中的感染节点包括：
    判断所述无标度模型的节点是否属于预设的黑名单；
    将属于所述黑名单的节点设为感染节点；
    所述根据感染节点并结合所述风险传导系数计算所述无标度模型中各路径的风险传导率包括：
    在所述无标度模型中识别与所述感染节点连续关联的连续路径，并以所述感染节点为起始依次对所述连续路径的子路径进行编号；其中，所述连续路径是指在所述无标度模型中以感染节点为输出源，依次与其串联的关联节点的整条路径，所述子路径为所述连续路径中相邻的两个节点之间的路径；
    将所述连续路径上的任一子路径设为待算路径，识别编号小于所述待算路径的子路径并提取其中的风险传导系数后汇总形成系数集，将所述系数集中的风险传导系数与所述待算路径的风险传导系数相乘获得真实传导系数，并将其加载于所述待算路径上。
  19. 根据权利要求15所述的计算机可读存储介质，其中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率包括：
    将与所述节点请求匹配的节点设为目标节点，在所述无标度模型中提取与所述目标节点连接的路径；
    将指向所述目标节点的路径设为传入路径，将从所述目标节点指出的路径设为传出路径；
    通过加权调整公式计算所述目标节点的传入路径的风险传导率，以获得传入风险率，通过加权调整公式计算所述目标节点的传出路径的风险传导率，以获得传出风险率。
  20. 根据权利要求15所述的计算机可读存储介质，其中，所述计算所述目标节点传入方向的风险传导率和传出方向的风险传导率以获得传入风险率和传出风险率之后包括：
    将所述传入风险率和传出风险率录入预设的四象限模型获得风险点，识别所述风险点所在区域并将该区域的名称作为判断结果，将所述判断结果输出至用户端。
PCT/CN2020/093201 2020-03-05 2020-05-29 一种数据分析方法、装置、计算机系统及可读存储介质 Ceased WO2021174693A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010146003.1A CN111401700B (zh) 2020-03-05 2020-03-05 一种数据分析方法、装置、计算机系统及可读存储介质
CN202010146003.1 2020-03-05

Publications (1)

Publication Number Publication Date
WO2021174693A1 true WO2021174693A1 (zh) 2021-09-10

Family

ID=71430500

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/093201 Ceased WO2021174693A1 (zh) 2020-03-05 2020-05-29 一种数据分析方法、装置、计算机系统及可读存储介质

Country Status (2)

Country Link
CN (1) CN111401700B (zh)
WO (1) WO2021174693A1 (zh)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114048330A (zh) * 2021-11-29 2022-02-15 平安银行股份有限公司 风险传导概率知识图谱生成方法、装置、设备及存储介质
CN114999654A (zh) * 2022-06-28 2022-09-02 神州医疗科技股份有限公司 基于知识图谱的糖尿病风险预测方法、装置、设备及介质
CN115086013A (zh) * 2022-06-13 2022-09-20 北京奇艺世纪科技有限公司 风险识别方法、装置、电子设备、存储介质和计算机程序产品
CN115374940A (zh) * 2022-08-08 2022-11-22 蚂蚁区块链科技(上海)有限公司 基于知识图谱的风险标签确定方法以及装置
CN115795055A (zh) * 2022-12-19 2023-03-14 广州城市规划技术开发服务部有限公司 一种关于土地用途数据的知识图谱构建方法及装置
CN115964507A (zh) * 2022-11-28 2023-04-14 北京海致星图科技有限公司 一种基于知识平台的图谱管理系统及计算机可读存储介质
CN116010657A (zh) * 2022-11-29 2023-04-25 兴业银行股份有限公司 基于有向图的信息数据管控方法及系统
CN120894028A (zh) * 2025-07-31 2025-11-04 山东融汇物产集团有限公司 一种基于风控管理的大宗商品交易平台系统及方法
CN120995035A (zh) * 2025-10-24 2025-11-21 中国医学科学院北京协和医院 医疗机构安全生产风险推理方法、装置、设备及存储介质

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112084343A (zh) * 2020-09-10 2020-12-15 杭州安恒信息安全技术有限公司 一种社会关系图谱的量化方法、装置和介质
CN112800242B (zh) * 2021-01-28 2023-07-28 平安科技（深圳）有限公司 谱系挖掘方法、装置、电子设备及计算机可读存储介质
CN112948381B (zh) * 2021-02-25 2022-10-28 平安科技（深圳）有限公司 数据处理方法、系统、计算机设备及可读存储介质
CN112883278A (zh) * 2021-03-23 2021-06-01 西安电子科技大学昆山创新研究院 基于智慧社区大数据知识图谱的不良舆论传播抑制方法
CN113077267B (zh) * 2021-03-31 2024-03-12 商运(江苏)科创发展有限公司 企业集群协调的供应链关系管理系统
CN115292424B (zh) * 2022-10-08 2022-12-20 凯美瑞德（苏州）信息科技股份有限公司 一种风险传导的分析方法、电子设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100198631A1 (en) * 2009-01-30 2010-08-05 Bank Of America Corporation Supplier stratification
CN106204264A (zh) * 2016-07-05 2016-12-07 天云融创数据科技（北京）有限公司 一种信贷担保网络风险传播模型构建方法
CN107563645A (zh) * 2017-09-04 2018-01-09 杭州云算信达数据技术有限公司 一种基于大数据的金融风险分析方法
CN108090709A (zh) * 2018-02-09 2018-05-29 重庆誉存大数据科技有限公司 一种基于风险传导模型的企业评估方法及系统
CN109472485A (zh) * 2018-11-01 2019-03-15 成都数联铭品科技有限公司 企业失信风险传播查询系统及方法
CN109949164A (zh) * 2019-03-28 2019-06-28 中山大学 一种基于投资关系网络的重要节点挖掘方法及装置

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170024531A1 (en) * 2015-07-22 2017-01-26 Radicalogic Technologies, Inc. Dba Rl Solutions Systems and methods for near-real or real-time contact tracing
US10747876B2 (en) * 2017-05-17 2020-08-18 Threatmodeler Software Inc. Systems and methods for assisted model generation
CN108335120A (zh) * 2018-03-07 2018-07-27 物数（上海）信息科技有限公司 基于区块链的资产溯源方法、装置、电子设备、存储介质
US20190311428A1 (en) * 2018-04-07 2019-10-10 Brighterion, Inc. Credit risk and default prediction by smart agents
CN110245165B (zh) * 2019-05-20 2023-04-11 平安科技（深圳）有限公司 风险传导关联图谱优化方法、装置和计算机设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100198631A1 (en) * 2009-01-30 2010-08-05 Bank Of America Corporation Supplier stratification
CN106204264A (zh) * 2016-07-05 2016-12-07 天云融创数据科技（北京）有限公司 一种信贷担保网络风险传播模型构建方法
CN107563645A (zh) * 2017-09-04 2018-01-09 杭州云算信达数据技术有限公司 一种基于大数据的金融风险分析方法
CN108090709A (zh) * 2018-02-09 2018-05-29 重庆誉存大数据科技有限公司 一种基于风险传导模型的企业评估方法及系统
CN109472485A (zh) * 2018-11-01 2019-03-15 成都数联铭品科技有限公司 企业失信风险传播查询系统及方法
CN109949164A (zh) * 2019-03-28 2019-06-28 中山大学 一种基于投资关系网络的重要节点挖掘方法及装置

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114048330A (zh) * 2021-11-29 2022-02-15 平安银行股份有限公司 风险传导概率知识图谱生成方法、装置、设备及存储介质
CN115086013A (zh) * 2022-06-13 2022-09-20 北京奇艺世纪科技有限公司 风险识别方法、装置、电子设备、存储介质和计算机程序产品
CN114999654A (zh) * 2022-06-28 2022-09-02 神州医疗科技股份有限公司 基于知识图谱的糖尿病风险预测方法、装置、设备及介质
CN115374940A (zh) * 2022-08-08 2022-11-22 蚂蚁区块链科技(上海)有限公司 基于知识图谱的风险标签确定方法以及装置
CN115964507A (zh) * 2022-11-28 2023-04-14 北京海致星图科技有限公司 一种基于知识平台的图谱管理系统及计算机可读存储介质
CN115964507B (zh) * 2022-11-28 2023-10-27 北京海致星图科技有限公司 一种基于知识平台的图谱管理系统及计算机可读存储介质
CN116010657A (zh) * 2022-11-29 2023-04-25 兴业银行股份有限公司 基于有向图的信息数据管控方法及系统
CN115795055A (zh) * 2022-12-19 2023-03-14 广州城市规划技术开发服务部有限公司 一种关于土地用途数据的知识图谱构建方法及装置
CN115795055B (zh) * 2022-12-19 2023-09-12 广州城市规划技术开发服务部有限公司 一种关于土地用途数据的知识图谱构建方法及装置
CN120894028A (zh) * 2025-07-31 2025-11-04 山东融汇物产集团有限公司 一种基于风控管理的大宗商品交易平台系统及方法
CN120995035A (zh) * 2025-10-24 2025-11-21 中国医学科学院北京协和医院 医疗机构安全生产风险推理方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN111401700B (zh) 2023-09-19
CN111401700A (zh) 2020-07-10

Similar Documents

Publication Publication Date Title
WO2021174693A1 (zh) 一种数据分析方法、装置、计算机系统及可读存储介质
CN110349009B (zh) 一种多头借贷违约预测方法、装置和电子设备
CN110795568A (zh) 基于用户信息知识图谱的风险评估方法、装置和电子设备
CN112925914B (zh) 数据安全分级方法、系统、设备及存储介质
CN110148053B (zh) 用户信贷额度评估方法、装置、电子设备和可读介质
Fronzetti Colladon et al. Forecasting financial markets with semantic network analysis in the COVID‐19 crisis
WO2023123933A1 (zh) 用户的类型信息的确定方法、设备及存储介质
CN111598360A (zh) 服务策略确定方法、装置及电子设备
CN112348659A (zh) 用户风险识别策略的分配方法、装置及电子设备
Kim Spatiotemporal scale dependency and other sensitivities in dynamic land-use change simulations
CN111815457A (zh) 目标对象的评估方法以及装置
CN111985578A (zh) 多源数据融合方法、装置、计算机设备及存储介质
CN116402625B (zh) 客户评估方法、装置、计算机设备及存储介质
Navdeep et al. Role of big data analytics in analyzing e-Governance projects
Gutmann et al. Banking crises and human rights
CN114638299B (zh) 僵尸许可证智能识别方法、装置、设备及存储介质
CN112348661A (zh) 基于用户行为轨迹的服务策略分配方法、装置及电子设备
Lee et al. Simulation experiment of disaster response organizational structures with alternative optimization techniques
CN105556514B (zh) 一种基于用户搜索行为进行数据挖掘的方法和装置
CN112712270A (zh) 信息处理方法、装置、设备及存储介质
CN115982654B (zh) 一种基于自监督图神经网络的节点分类方法及装置
CN118780793A (zh) 异常对象确定方法及相关设备
JP5156692B2 (ja) 擬似データ生成装置、擬似データ生成方法及びコンピュータプログラム
CN111583037B (zh) 风险关联对象的确定方法、装置和服务器
CN116756419A (zh) 基于人工智能的信用卡权益推荐方法、装置、设备及介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20923309

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20923309

Country of ref document: EP

Kind code of ref document: A1