Decision engine-based data source scheduling decision method and comprehensive decision platform

Abstract

The invention discloses a data source scheduling decision-making method based on a decision-making engine and a comprehensive decision-making platform, and belongs to the technical field of data scheduling, and the method comprises the following steps: firstly, receiving a decision-making request, extracting customer basic information and a decision-making type, and then calling the decision-making engine according to the decision-making type to carry out data source importance analysis. Based on the analysis, a data scheduling strategy and an evaluation model are generated, and then data are retrieved and decided to obtain a decision result. In the process, the availability, timeliness and accuracy of the data are concerned, the relevance, complementarity and potential data conflicts among the data sources are comprehensively considered, and efficient, flexible and low-cost calling of the data sources is achieved. The data scheduling scheme is generated according to needs, the result is evaluated, and the selection and combination of the data sources are dynamically adjusted, so that the decision accuracy and response speed are remarkably improved, powerful support is provided for various complex decisions, and the decision cost is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data scheduling, and more specifically to a data source scheduling decision method and a comprehensive decision platform based on a decision engine. Background Art

[0002] With the rapid development of financial markets and the increasing complexity of credit business, financial institutions are facing unprecedented challenges in credit risk management. Credit risk involves not only the credit status and repayment ability of borrowers, but also multiple factors such as market fluctuations and policy changes. In order to more effectively manage credit risk and improve the accuracy and efficiency of credit decisions, financial institutions have built risk decision-making platforms and rely on advanced data scheduling methods to support the operation of the platform. As a core component of the risk decision-making platform, the effectiveness of data scheduling directly affects the accuracy of risk assessment and the timeliness of decision-making.

[0003] At present, banks and other credit financial institutions generally choose to purchase the required data services from data source vendors during their operations, and usually adopt a billing method based on the number of calls. When processing each credit application, in order to comprehensively evaluate the credit status of the applicant, it is often necessary to call multiple data source interfaces to obtain relevant information. The traditional data scheduling method often needs to wait for all preset data source results to be returned before the next decision analysis can be carried out. This processing method will significantly reduce the efficiency of credit decision-making when the data source response time is long or the network latency is high. In addition, in the actual risk control strategy deployment process, it is sometimes necessary to adjust the call of other data sources based on the score of key data sources or the credit risk situation of the applicant. However, traditional data scheduling methods are often difficult to flexibly respond to such changes in demand.

[0004] Therefore, how to provide an efficient and flexible data source scheduling decision method and a comprehensive decision platform is a problem that technical personnel in this field need to solve urgently. Summary of the invention

[0005] In view of this, the present invention provides a data source scheduling decision method and a comprehensive decision platform based on a decision engine, which analyzes the relevant data sources of the decision request and performs multiple data scheduling and evaluation based on the analysis results until a decision can be made, thereby realizing efficient and flexible calling of the data source.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] In one aspect, the present invention provides a data source scheduling decision method based on a decision engine, comprising:

[0008] Receiving a decision request, and extracting basic customer information and decision type of the decision request;

[0009] Based on the decision type, the decision engine is called to perform data source importance analysis to obtain the data source importance;

[0010] Generate a data scheduling strategy and an evaluation model according to the importance of the data source;

[0011] Based on the basic information of the customer and the data scheduling strategy, data is retrieved, and a decision is made according to the evaluation model to obtain a decision result.

[0012] Preferably, calling the decision engine to perform data source importance analysis based on the decision type includes:

[0013] Determine a data source associated with the decision type according to the decision type, and construct a relationship network between the decision type and the data source;

[0014] Determining a first importance index of the data source according to the relationship network;

[0015] Identifying abnormal events of the data source according to the relationship network, and determining the impact value of the abnormal events on the decision result;

[0016] Determining a second importance index of the data source based on the impact value;

[0017] The importance of the data source is determined based on the first importance indicator and the second importance indicator.

[0018] Preferably, generating a data scheduling strategy according to the importance of the data source includes:

[0019] Inputting the importance of the data source into a preset first scheduling model to obtain a first data source scheduling plan;

[0020] Inputting the result of the first data source scheduling scheme into the evaluation model to obtain a first decision;

[0021] Determine whether to introduce secondary data scheduling;

[0022] If not, a termination command is returned;

[0023] If so, the importance of the remaining data sources is input into the preset second scheduling model to schedule the remaining data sources to obtain a second data source scheduling plan, the result of the second data source scheduling plan and the result of the first data source scheduling plan are input into the evaluation model to obtain a second decision result, and continue to determine whether to introduce secondary data scheduling until the determination result is no.

[0024] Preferably, the process of constructing the first scheduling model and the second scheduling model is:

[0025] A multi-objective mathematical optimization model is established with the minimization of the linear weighted sum of various costs as the objective function. The multi-objective mathematical optimization model is the first scheduling model and the second scheduling model. The various costs include: cost cost, time cost, and computing cost. The constraints of the multi-objective mathematical optimization model include: cost constraint, time constraint, and importance constraint of data source.

[0026] Preferably, retrieving data based on the basic information of the customer and the data scheduling strategy, and making a decision according to the evaluation model to obtain a decision result includes:

[0027] Extract risk events based on the data retrieved by the data scheduling strategy and obtain risk events for each type of data;

[0028] Weighting is performed based on the threat level of risk events for each type of data;

[0029] Obtain the number of risk events and each risk event in each data source;

[0030] Calculate the risk level of each data source based on each risk event and the number of risk events;

[0031] The risk level of the customer is calculated based on the risk level and the importance of each data source, and the risk level is the decision result of the customer.

[0032] On the other hand, the present invention provides a comprehensive decision-making platform based on a decision engine, comprising: a comprehensive decision-making platform and a decision engine, wherein the comprehensive decision-making platform interacts with the decision engine multiple times to obtain a decision result;

[0033] The comprehensive decision platform includes a processing unit for receiving a decision request, extracting basic customer information and a decision type of the decision request, and a decision unit for retrieving data based on the basic customer information and a data scheduling strategy generated by a decision engine, and outputting a decision according to an evaluation result of an evaluation model of the decision engine;

[0034] The decision engine includes an analysis unit for calling the decision engine to perform data source importance analysis based on the decision type and a strategy unit for generating a data scheduling strategy and an evaluation model according to the data source importance.

[0035] Preferably, the analysis unit comprises:

[0036] A relationship network unit, used to determine a data source associated with a decision type according to the decision type, and to construct a relationship network between the decision type and the data source;

[0037] A first indicator unit, used to determine a first importance indicator of the data source according to the relationship network;

[0038] An abnormal impact unit, used to identify abnormal events of the data source according to the relationship network, and determine the impact value of the abnormal event on the decision result;

[0039] A second indicator unit, configured to determine a second importance indicator of the data source based on the impact value;

[0040] A fusion unit is used to determine the importance of the data source based on the first importance indicator and the second importance indicator.

[0041] Preferably, the strategy unit includes:

[0042] A first scheduling unit, configured to input the importance of the data source into a preset first scheduling model to obtain a first data source scheduling scheme;

[0043] An evaluation unit, configured to input the result of the first data source scheduling scheme into the evaluation model to obtain a first decision;

[0044] A determination unit, used for determining whether to introduce secondary data scheduling;

[0045] The second scheduling unit is used to input the importance of the remaining data sources into a preset second scheduling model to schedule the remaining data sources, obtain a second data source scheduling scheme, input the result of the second data source scheduling scheme and the result of the first data source scheduling scheme into the evaluation model to obtain a second decision result, and continue to determine whether to introduce secondary data scheduling until the determination result is no.

[0046] Preferably, the evaluation unit comprises:

[0047] An event extraction unit is used to extract risk events based on the data retrieved by the data scheduling strategy, and obtain risk events for each data source;

[0048] A weight division unit, used to divide the weights based on the threat level of risk events of each data source;

[0049] The statistical unit is used to obtain the number of risk events and each type of risk event in each data source;

[0050] A risk calculation unit, used to calculate the risk level of each data source based on each risk event and the number of risk events;

[0051] The result output unit is used to calculate the risk level of the customer based on the risk level and the importance of each data source, and the risk level is the decision result of the customer.

[0052] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a data source scheduling decision method and a comprehensive decision platform based on a decision engine. By deeply analyzing the relevant data sources involved in the decision request, it not only pays attention to the availability, timeliness and accuracy of the data, but also comprehensively considers the correlation, complementarity and potential data conflicts between data sources. Based on the analysis results, a data scheduling plan is generated on demand, and then the results are evaluated. The selection and combination of the next data source are dynamically adjusted according to the evaluation results to optimize the quality and efficiency of decision support and reduce decision costs. The present invention realizes efficient, flexible and low-cost calling of data sources, significantly improves the accuracy and response speed of decisions, and provides strong support for various complex decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0054] Figure 1 A schematic diagram of the process provided by the present invention;

[0055] Figure 2 The overall system framework diagram provided by the present invention;

[0056] Figure 3 It is a schematic diagram of the specific structure of the evaluation unit. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The embodiment of the present invention discloses a data source scheduling decision method based on a decision engine, such as Figure 1 As shown, including:

[0059] Step 1: Receive a decision request and extract the basic customer information and decision type of the decision request.

[0060] Receive decision requests from clients through preset interfaces (such as API, web forms or message queues). Basic customer information includes customer ID, name, age, gender, occupation, etc. This information is crucial for subsequent data retrieval, data analysis and decision making. The decision type clarifies the specific problem that the customer wants to solve, such as credit risk assessment, product recommendation, market trend forecast, etc.

[0061] Step 2: Based on the decision type, call the decision engine to analyze the first scheduling model and the second scheduling model for the importance of the data source.

[0062] By analyzing the relevance and reliability of various data sources for the current decision-making problem, we can determine which data is necessary, which data is auxiliary, and their priority. This helps optimize the data calling process, reduce unnecessary computing burden, and improve decision-making efficiency.

[0063] Step 3: Generate data scheduling strategy and evaluation model based on the importance of data source.

[0064] Step 4: Retrieve data based on basic customer information and data scheduling strategies, make decisions based on the evaluation model, and obtain decision results.

[0065] Specifically, step 2: based on the decision type, calling the decision engine to perform data source importance analysis includes:

[0066] Step 2.1: Determine the data sources associated with the decision type based on the decision type, and construct a relationship network between the decision type and the data source; these data sources include customer transaction records, credit history, industry reports, market trend data, social media sentiment analysis, etc. In the constructed relationship network, the nodes represent the data sources, and the edges represent the dependencies, complementary relationships, or conflicting relationships between the data sources. By analyzing these relationships, the system can understand how the data sources work together to support the decision process and identify key data paths and potential bottlenecks. When constructing the relationship network, factors such as the timeliness, accuracy, and completeness of the data source are also considered to ensure that the selected data source can provide valuable input for the decision process.

[0067] Step 2.2: Determine the first importance index of the data source based on the relationship network; the first importance index takes into account factors such as the position of the data source in the relationship network (such as graph theory indicators such as centrality and closeness centrality), the historical performance of the data source (such as data quality and update frequency), and the direct contribution of the data source to the decision result. By combining these factors, the system can preliminarily evaluate the importance of each data source for the current decision type.

[0068] Step 2.3: Identify abnormal events of data sources based on the relationship network and determine the impact of abnormal events on decision-making results; abnormal events include: credit rating decline, default or default, debt evasion, major changes in business management, changes in equity structure, changes in senior management, changes in balance sheet structure, decline in profitability, abnormal changes in funds, involvement in major litigation, violation of laws and regulations, industry policy adjustments, changes in the economic environment, etc. For the identified abnormal events, use sensitivity analysis, causal inference or risk assessment models to quantify the impact of abnormal events.

[0069] Step 2.4: Determine the second importance indicator of the data source based on the impact value; quantify the impact of abnormal events on credit decision results by building a risk assessment model or using expert judgment.

[0070] Step 2.5: Determine the importance of the data source based on the first importance index and the second importance index. The importance of the first importance index and the second importance index not only reflects the importance of the data source under normal circumstances, but also considers the stability and reliability of the data source under abnormal circumstances. Based on this comprehensive importance, optimize the data scheduling strategy, adjust the evaluation model parameters, or take other measures to ensure the accuracy and efficiency of the decision-making process.

[0071] Furthermore, a data scheduling strategy is generated according to the importance of the data source, including:

[0072] Inputting the importance of the data source into a preset first scheduling model to obtain a first data source scheduling plan;

[0073] Inputting the result of the first data source scheduling scheme into the evaluation model to solve and obtain a first decision;

[0074] Determine whether to introduce secondary data scheduling;

[0075] If not, a termination command is returned;

[0076] If so, the importance of the remaining data sources is input into the preset second scheduling model to schedule the remaining data sources to obtain a second data source scheduling plan, and the results of the second data source scheduling plan and the results of the first data source scheduling plan are input into the evaluation model to obtain a second decision result, and continue to determine whether to introduce secondary data scheduling until the determination result is no.

[0077] Specifically, the process of constructing the first scheduling model and the second scheduling model is:

[0078] Taking the minimum linear weighted sum of each cost as the objective function, a multi-objective mathematical optimization model is established. The multi-objective mathematical optimization model is the first scheduling model and the second scheduling model. Each cost includes: cost cost, time cost, and calculation cost. The constraints of the objective mathematical optimization model include: cost constraint, time constraint, and importance constraint of data source. The specific steps are:

[0079] (1) Define variables:

[0080] Let x i is the i-th decision variable, which represents a certain allocation or choice in the scheduling process; let C f is the cost, C t is the time cost, C c To calculate the cost.

[0081] (2) Define the objective function:

[0082] The goal is to minimize the total cost by weighting the sum of the costs. It can be expressed as:

[0083] minZ=w f C f +w t C t +w c C c

[0084] Among them, w f , w t , w c is the weight coefficient of each cost, satisfying w f +w t +w c =1.

[0085] (3) Constraints:

[0086] Cost constraints:

[0087] If you need to control the total cost, you can set an upper limit B, then there are constraints:

[0088] C f ≤B.

[0089] Time constraints:

[0090] Control the total time to complete the task and set a maximum allowed time T:

[0091] C t ≤T.

[0092] Importance constraints of data sources:

[0093] If there are multiple data sources, each of which has its own importance, you can set the constraints as follows:

[0094] ∑ j α j x j ≥I;

[0095] Among them, α j is the importance of the jth data source, x j is the jth decision variable, and I is the lower limit of the importance required to be achieved.

[0096] (4) Comprehensive model:

[0097] Combining the above contents, the multi-objective mathematical optimization model is obtained as follows:

[0098] minZ=w f C f +w t C t +w c C c

[0099] St

[0100] C f ≤B;

[0101] C t ≤T;

[0102] ∑ j α j x j ≥I;

[0103]

[0104] Furthermore, for the multi-objective mathematical optimization model, linear programming (LP) or multi-objective programming related algorithms can be used to solve it, such as weighted method, goal programming method, etc., and the appropriate solution method is selected according to the needs of the specific problem.

[0105] In another embodiment, data is retrieved based on the basic information of the customer and the data scheduling strategy, and a decision is made according to the evaluation model to obtain a decision result, including:

[0106] Extract risk events based on the data retrieved by the data scheduling strategy and obtain risk events for each type of data;

[0107] Weighting is performed based on the threat level of risk events for each type of data;

[0108] Assume that the set of risk events E = {e1, e2, ..., e n}; Each event e i There is a threat level T(e i ),but:

[0109]

[0110] Where: W(e i ) is the risk event category e i The weight of .

[0111] Obtain the number of risk events and each risk event in each data source;

[0112] For each data source D i , record the categories and quantities of risk events, and use a mapping S(D i ) indicates that S(D i )(e i ) represents the risk event category e i In data source D i The number on.

[0113] Calculate the risk level of each data source based on each risk event and the number of risk events;

[0114]

[0115] Where: H(D i ) indicates data source D i degree of threat.

[0116] Calculate the risk level of the customer based on the risk level and the importance of each data source. The risk level is the decision result of the customer.

[0117] DG(D i )=H(D i )+I'(D i );

[0118] Where: DG(D i ) is the risk level score of the data source, I'(D i ) is the data source D i The importance of.

[0119] On the other hand, the present invention provides a comprehensive decision-making platform based on a decision engine, referring to Figure 2 , including: a comprehensive decision-making platform and a decision-making engine. The comprehensive decision-making platform interacts with the decision-making engine multiple times to obtain decision results;

[0120] The comprehensive decision platform includes a processing unit for receiving a decision request, extracting basic customer information and decision type of the decision request, and a decision unit for retrieving data based on basic customer information and a data scheduling strategy generated by a decision engine, and outputting a decision based on an evaluation result of an evaluation model of the decision engine;

[0121] The decision engine includes an analysis unit for calling the decision engine to perform data source importance analysis based on the decision type and a strategy unit for generating a data scheduling strategy and an evaluation model according to the data source importance.

[0122] Specifically, the analysis units include:

[0123] A relationship network unit, used to determine a data source associated with a decision type according to the decision type, and to construct a relationship network between the decision type and the data source;

[0124] A first indicator unit, used to determine a first importance indicator of a data source according to the relationship network;

[0125] The abnormal impact unit is used to identify abnormal events of the data source according to the relationship network and determine the impact value of the abnormal events on the decision results;

[0126] A second indicator unit, used to determine a second importance indicator of the data source based on the influence value;

[0127] The fusion unit is used to determine the importance of the data source based on the first importance index and the second importance index.

[0128] Furthermore, the strategy unit includes:

[0129] A first scheduling unit, used to input the importance of the data source into a preset first scheduling model to obtain a first data source scheduling plan;

[0130] An evaluation unit, configured to input the result of the first data source scheduling scheme into the evaluation model to obtain a first decision;

[0131] A determination unit, used for determining whether to introduce secondary data scheduling;

[0132] The second scheduling unit is used to input the importance of the remaining data sources into a preset second scheduling model to schedule the remaining data sources, obtain a second data source scheduling plan, input the result of the second data source scheduling plan and the result of the first data source scheduling plan into the evaluation model to obtain a second decision result, and continue to determine whether to introduce secondary data scheduling until the determination result is no.

[0133] Furthermore, Figure 3 As shown, the evaluation unit consists of:

[0134] An event extraction unit is used to extract risk events based on the data retrieved by the data scheduling strategy, and obtain risk events for each data source;

[0135] A weight division unit, used to divide the weights based on the threat level of risk events of each data source;

[0136] The statistical unit is used to obtain the number of risk events and each type of risk event in each data source;

[0137] A risk calculation unit, used to calculate the risk level of each data source based on each risk event and the number of risk events;

[0138] The result output unit is used to calculate the risk level of the customer based on the risk level and the importance of each data source, and the risk level is the decision result of the customer.

[0139] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0140] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data source scheduling decision method based on a decision engine, characterized in that: include: Receiving a decision request, and extracting basic customer information and decision type of the decision request; Based on the decision type, the decision engine is called to perform data source importance analysis to obtain the data source importance; Generate a data scheduling strategy and an evaluation model according to the importance of the data source; Based on the basic information of the customer and the data scheduling strategy, data is retrieved, and a decision is made according to the evaluation model to obtain a decision result.

2. A data source scheduling decision method based on a decision engine according to claim 1, characterized in that: Calling the decision engine to perform data source importance analysis based on the decision type includes: Determine a data source associated with the decision type according to the decision type, and construct a relationship network between the decision type and the data source; Determining a first importance index of the data source according to the relationship network; Identifying abnormal events of the data source according to the relationship network, and determining the impact value of the abnormal events on the decision result; Determining a second importance index of the data source based on the impact value; The importance of the data source is determined based on the first importance indicator and the second importance indicator.

3. A data source scheduling decision method based on a decision engine according to claim 2, characterized in that: Generating a data scheduling strategy according to the importance of the data source includes: Inputting the importance of the data source into a preset first scheduling model to obtain a first data source scheduling plan; Inputting the result of the first data source scheduling scheme into the evaluation model to obtain a first decision; Determine whether to introduce secondary data scheduling; If not, a termination command is returned; If so, the importance of the remaining data sources is input into the preset second scheduling model to schedule the remaining data sources to obtain a second data source scheduling plan, the result of the second data source scheduling plan and the result of the first data source scheduling plan are input into the evaluation model to obtain a second decision result, and continue to determine whether to introduce secondary data scheduling until the determination result is no.

4. A data source scheduling decision method based on a decision engine according to claim 3, characterized in that: The process of constructing the first scheduling model and the second scheduling model is: A multi-objective mathematical optimization model is established with the minimization of the linear weighted sum of various costs as the objective function. The multi-objective mathematical optimization model is the first scheduling model and the second scheduling model. The various costs include: cost cost, time cost, and computing cost. The constraints of the multi-objective mathematical optimization model include: cost constraint, time constraint, and importance constraint of data source.

5. A data source scheduling decision method based on a decision engine according to claim 4, characterized in that: Retrieving data based on the basic customer information and the data scheduling strategy, and making decisions based on the evaluation model to obtain decision results, including: Extract risk events based on the data retrieved by the data scheduling strategy and obtain risk events for each type of data; Weighting is performed based on the threat level of risk events for each type of data; Obtain the number of risk events and each risk event in each data source; Calculate the risk level of each data source based on each risk event and the number of risk events; The risk level of the customer is calculated based on the risk level and the importance of each data source, and the risk level is the decision result of the customer.

6. A comprehensive decision-making platform based on a decision engine, characterized in that: include: A comprehensive decision-making platform and a decision-making engine, wherein the comprehensive decision-making platform interacts with the decision-making engine multiple times to obtain a decision result; The comprehensive decision platform includes a processing unit for receiving a decision request, extracting basic customer information and a decision type of the decision request, and a decision unit for retrieving data based on the basic customer information and a data scheduling strategy generated by a decision engine, and outputting a decision according to an evaluation result of an evaluation model of the decision engine; The decision engine includes an analysis unit for calling the decision engine to perform data source importance analysis based on the decision type and a strategy unit for generating a data scheduling strategy and an evaluation model according to the data source importance.

7. The comprehensive decision-making platform based on the decision engine according to claim 6, characterized in that: The analysis unit comprises: A relationship network unit, used to determine a data source associated with a decision type according to the decision type, and to construct a relationship network between the decision type and the data source; A first indicator unit, used to determine a first importance indicator of the data source according to the relationship network; An abnormal impact unit, used to identify abnormal events of the data source according to the relationship network, and determine the impact value of the abnormal event on the decision result; A second indicator unit, configured to determine a second importance indicator of the data source based on the impact value; A fusion unit is used to determine the importance of the data source based on the first importance indicator and the second importance indicator.

8. The comprehensive decision-making platform based on the decision engine according to claim 7, characterized in that: The strategy unit comprises: A first scheduling unit, configured to input the importance of the data source into a preset first scheduling model to obtain a first data source scheduling scheme; An evaluation unit, configured to input the result of the first data source scheduling scheme into the evaluation model to obtain a first decision; A determination unit, used for determining whether to introduce secondary data scheduling; The second scheduling unit is used to input the importance of the remaining data sources into a preset second scheduling model to schedule the remaining data sources, obtain a second data source scheduling scheme, input the result of the second data source scheduling scheme and the result of the first data source scheduling scheme into the evaluation model to obtain a second decision result, and continue to determine whether to introduce secondary data scheduling until the determination result is no.

9. The comprehensive decision-making platform based on the decision engine according to claim 8, characterized in that: The evaluation unit comprises: An event extraction unit is used to extract risk events based on the data retrieved by the data scheduling strategy, and obtain risk events for each data source; A weight division unit, used to divide the weights based on the threat level of risk events of each data source; The statistical unit is used to obtain the number of risk events and each type of risk event in each data source; A risk calculation unit, used to calculate the risk level of each data source based on each risk event and the number of risk events; The result output unit is used to calculate the risk level of the customer based on the risk level and the importance of each data source, and the risk level is the decision result of the customer.

Images