WO2022100062A1 - 风险管控方法、装置、电子设备和存储介质 - Google Patents

风险管控方法、装置、电子设备和存储介质 Download PDF

Info

Publication number
WO2022100062A1
WO2022100062A1 PCT/CN2021/096646 CN2021096646W WO2022100062A1 WO 2022100062 A1 WO2022100062 A1 WO 2022100062A1 CN 2021096646 W CN2021096646 W CN 2021096646W WO 2022100062 A1 WO2022100062 A1 WO 2022100062A1
Authority
WO
WIPO (PCT)
Prior art keywords
individual
tested
risk
area
value
Prior art date
Application number
PCT/CN2021/096646
Other languages
English (en)
French (fr)
Inventor
赵婷婷
孙行智
赵惟
廖希洋
徐卓扬
刘卓
Original Assignee
平安科技（深圳）有限公司
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 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2022100062A1 publication Critical patent/WO2022100062A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu

Definitions

  • the present application relates to the technical field of intelligent decision-making, and in particular, to a risk management and control method, device, electronic device and storage medium.
  • Infectious disease outbreak is a public health event that seriously threatens public health and life safety. It has the characteristics of rapid spread and wide impact. Infectious diseases can spread between individuals through contact or breathing and become widespread. The prevention and control of infectious disease risks in specific areas, such as public places, is a very important and arduous task.
  • this method is not real-time enough, the control efficiency is low, and the control effect is poor.
  • the embodiments of the present application provide a risk management and control method, apparatus, electronic device, and storage medium, which can improve the real-time performance of infectious disease prevention and control to a certain extent, improve management and control efficiency, and improve management and control effects.
  • the embodiment of the present application provides a risk management and control method, the method includes:
  • the individual data includes at least one of the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the feature information of the target area where the individual to be tested is located.
  • the risk prediction model uses the individual data as the input data of the risk prediction model to perform risk prediction according to the individual data to obtain the risk value of the individual to be tested;
  • a corresponding security precaution strategy is executed according to the risk value of the individual to be tested.
  • an embodiment of the present application provides a risk management and control device, and the device includes:
  • the acquisition module is used to acquire the individual data of the individual to be tested for risk prediction, the individual data includes the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the information of the target area where the individual to be tested is located. at least one of the characteristic information;
  • a prediction module configured to use the individual data as input data of a risk prediction model, and the risk prediction model performs risk prediction according to the individual data to obtain the risk value of the individual to be tested;
  • the execution module is configured to execute the corresponding security prevention strategy according to the risk value of the individual to be tested.
  • an embodiment of the present application provides an electronic device, including a processor, an input interface, an output interface, and a memory, wherein the processor, the input interface, the output interface, and the memory are connected to each other, wherein the memory is used to store a computer A program, the computer program includes program instructions, and the processor is configured to invoke the program instructions to execute a risk management and control method, including:
  • the individual data includes at least one of the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the feature information of the target area where the individual to be tested is located.
  • the risk prediction model uses the individual data as the input data of the risk prediction model to perform risk prediction according to the individual data to obtain the risk value of the individual to be tested;
  • a corresponding security precaution strategy is executed according to the risk value of the individual to be tested.
  • an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and the computer program includes program instructions that, when executed by a processor, cause the The processor implements risk management methods, including:
  • the individual data includes at least one of the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the feature information of the target area where the individual to be tested is located.
  • the risk prediction model uses the individual data as the input data of the risk prediction model to perform risk prediction according to the individual data to obtain the risk value of the individual to be tested;
  • a corresponding security precaution strategy is executed according to the risk value of the individual to be tested.
  • the embodiment of the present application is that after the infected person is determined, the moving path of the infected person is published to remind others to pay attention to travel safety, and risk management and control cannot be performed in real time, and the control effect is poor.
  • the present application implements the above method, It is possible to understand the risk situation in real time and carry out risk management and control in a timely manner, which improves the management and control effect and efficiency to a certain extent.
  • FIG. 1 is a schematic diagram of a network architecture of a risk management and control system according to an embodiment of the present application
  • FIG. 2 is a schematic flowchart of a risk management and control method provided by an embodiment of the present application
  • FIG. 3 is a schematic diagram of a risk prediction model provided by an embodiment of the present application.
  • FIG. 4a is a schematic diagram of an early warning scenario provided by an embodiment of the present application.
  • 4b is a schematic diagram of another early warning scenario provided by an embodiment of the present application.
  • FIG. 5 is a schematic flowchart of another risk management and control method provided by an embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of a risk management and control device according to an embodiment of the present application.
  • FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • the technical solution of the present application relates to the field of artificial intelligence technology, and can be applied to scenarios such as risk prediction to realize risk management and control, thereby promoting the construction of smart cities.
  • the data involved in this application such as individual data and/or security protection policies, may be stored in a database, or may be stored in a blockchain, such as distributed storage through a blockchain, which is not limited in this application.
  • a risk management and control method proposed by the embodiment of the present application can be applied to an electronic device, and the electronic device can be a terminal device or a server.
  • the terminal device here can be a monitoring device or can be an intelligent terminal, such as an electronic device such as a smart phone, a notebook computer, and a desktop computer.
  • the network architecture of a risk management and control system proposed by an embodiment of the present application may include: a smartphone 101 and a server 102 .
  • the smartphone 101 communicates with the server 102 through a network.
  • user A needs to perform risk prediction
  • user A or a staff member in the target area can send a risk prediction request to server 102 through smartphone 101 to trigger server 102 to acquire user A's individual data for risk prediction for subsequent follow-up risk forecasting process.
  • the server 102 may acquire the risk data of the user A for use in the subsequent risk prediction process when the corresponding device detects that the user A enters the target area.
  • part or all of the data of the individual data may be acquired from the smartphone 101, or may be acquired in other ways, which are not limited herein.
  • the above process can improve the efficiency of risk prediction and the effectiveness of risk prediction by automatically collecting and processing individual data for risk prediction, and implementing subsequent risk management and control related processes.
  • FIG. 2 is a schematic flowchart of a risk management and control method proposed by an embodiment of the present application. The method is applied to the aforementioned electronic device. As shown in FIG. 2 , the risk management and control method in this embodiment may include the following steps:
  • the individual to be tested may refer to an individual entering the target area, an individual who has close contact with an infectious disease patient, etc.
  • the individual data may refer to the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the target of the individual to be tested. At least one of the feature information of the area.
  • the identity information may refer to the gender and age of the individual to be tested
  • the risk factor information may refer to the body temperature of the individual to be tested, information indicating whether to pass through a high-risk area, and information indicating whether there are suspected symptoms, such as : Cough, dizziness, fever, etc.
  • the characteristic information of the target area may refer to the area, traffic flow, and geographic location of the target area.
  • the target area can be an indoor or a public area of a place, such as a shopping mall or a bar.
  • the electronic device may acquire individual data of the individual to be tested for risk prediction when a request for risk prediction of the individual to be tested is received.
  • the electronic device may acquire individual data of the individual to be tested for risk prediction when the relevant device detects that the individual to be tested enters the target area.
  • the relevant device may be an image acquisition device such as a camera device, or a device such as a positioning device of a terminal device of an individual to be tested, which will not be listed here.
  • the image capture device such as the camera device can be set in the target area.
  • the electronic device can collect the individual data of the individual to be measured through various information collection methods, wherein the identity information of the individual to be measured can be obtained through information association, for example, the identity information of the individual to be measured can be obtained by correlating ID card information Information on risk factors can also be collected through field measurements and electronic questionnaires.
  • the electronic device pre-trains the risk prediction model.
  • the electronic device inputs the individual data into the risk prediction model, and the risk prediction model outputs the risk value of the individual to be tested.
  • the risk value can represent the probability that the tested individual will be infected.
  • the risk prediction model may refer to a pre-trained gradient boosting tree model
  • the process of obtaining the risk value of the individual to be tested through the risk prediction model may be that the electronic device uses the individual data for risk prediction as a gradient boosting tree
  • the input data of the model is divided into individual data by each decision tree included in the pre-trained gradient boosting tree model to determine the leaf node where the individual data is located in each decision tree, and the individual to be tested is determined according to the value corresponding to each leaf node. value at risk.
  • the gradient boosted tree model may be an extreme gradient model XGboost.
  • the pre-trained gradient boosting tree model includes K decision trees that are established, and the K decision trees included in the pre-training gradient boosting tree model divide the individual data into features to determine the location of the individual data in the K decision trees where each decision tree is located.
  • the leaf node, and the risk value of the individual to be tested is determined according to the value of the individual data corresponding to the leaf node where each decision tree is located.
  • the method for the electronic device to determine the risk value of the individual to be tested according to the value corresponding to each leaf node may be as follows: the electronic device adds the values corresponding to each leaf node, and outputs a total value, which is the total value to be tested.
  • the pre-trained gradient boosting tree model contains two decision trees tree1 and tree2.
  • tree1 according to the feature division, the individual data of individual 1 to be tested is divided into a1 node, and the corresponding The value is A1, the individual data of the individual to be tested 2 is divided into the a2 node, and the value corresponding to the a2 node is A2; in tree2, the individual data of the individual to be tested 1 and the individual data of the individual to be tested 2 are divided into the b1 node,
  • the value corresponding to node b1 is B1, that is, the risk value of individual 1 to be tested and the risk value of individual 2 to be tested
  • the specific manner of training the gradient boosting tree model may be: obtaining a training sample set, where the training sample set may include multiple training samples, and the training samples may be individual data of the individual to be tested.
  • the training sample set is ⁇ (x 1 , y 1 ), (x 2 , y 2 )...(x n , y n ) ⁇ , where x n represents the training sample, and y n represents the true value corresponding to the training sample.
  • the loss function is
  • the regularization term is ⁇ (f k ).
  • the optimization goal of the gradient boosted tree model is to minimize the objective function:
  • ⁇ k ⁇ (f k ) represents the complexity of K trees
  • i represents the ith sample
  • k represents the kth tree
  • the electronic device can use the training sample set to construct K decision trees of the gradient boosting model, and generate a gradient boosting model including the K decision trees as a pre-trained gradient boosting model.
  • S203 Execute a corresponding security prevention strategy according to the risk value of the individual to be tested.
  • executing the corresponding security prevention strategy may be that the electronic device determines the risk group category to which the individual to be tested belongs according to the risk value of the individual to be tested.
  • the risk group category may be a low-risk group, a medium-risk group , high-risk groups.
  • the electronic device will give an early warning to the target person.
  • the target person can be the individual to be tested, and it can also be a staff member in the target area where the individual to be tested is located.
  • the early warning prompt can be sent to the terminal device corresponding to the target person, and the early warning prompt can be used to prompt the individual to be tested as a designated risk Crowd category, the individual to be tested is not allowed to enter the target area, etc.
  • the electronic device may determine the risk group category to which the individual to be tested belongs according to the risk value of the individual to be tested.
  • the electronic device may determine the risk level of the individual to be tested according to the corresponding relationship between the risk value and the risk group category.
  • the corresponding risk group type, and the risk level category corresponding to the risk level of the individual to be tested is determined as the risk group category to which the individual to be tested belongs.
  • executing the corresponding security range policy may also be that the electronic device determines the safety distance of the individual to be tested according to the risk value of the individual to be tested, and sends safety prompt information to the terminal device corresponding to the individual to be tested, for example, by The short message is sent to the terminal device, and the safety prompt information includes at least the safety distance of the individual to be tested, and the safety prompt information is used to remind the individual to be tested and other individuals at least the safety distance that needs to be maintained.
  • the specific method of determining the safety distance according to the risk value may be to obtain some discrete samples (Q1, R1), (Q2, R2), . . . (Qn, Rn), the discrete samples representing the risk value
  • the individual of Q should have a safety distance R, take the risk value Q as the independent variable and R as the dependent variable, and use the function fitting method to calculate an expression representing the mapping relationship between the risk value and the safety distance. Enter the value into this expression, and calculate the safety distance of the individual to be tested.
  • executing the corresponding safety range policy may also be that after the electronic device determines the safety distance of the individual to be measured, the electronic device determines the safety area of the individual to be measured according to the safety distance of the individual to be measured.
  • the electronic device will give an early warning prompt to the individual to be tested, wherein the safe area can be a circle with a safe distance as the radius, and the method of early warning can be to send prompt information to the corresponding individual to be measured.
  • the prompt information can be used to prompt that there are other individuals in the safe area of the individual to be tested, as shown in Figure 4a.
  • executing the corresponding security range policy may also be that after the electronic device determines the security area of the individual to be tested, the electronic device determines the area of the security area of each individual in the target area, and the electronic device determines the area of the security area of each individual in the target area, and the electronic device determines the area of the security area of each individual in the target area according to When the area of the total safe area is greater than or equal to the preset area, the electronic device outputs the warning prompt information, and the warning prompt information is used to prompt the number of individuals in the target area to reach The upper limit, the preset area is determined according to the area of the target area.
  • the area of the total safety area can be the sum of the areas of the safety areas of each individual, namely
  • S the total area of the target area
  • ⁇ an adjustable parameter, which reflects the target area’s impact on risk
  • the warning prompt information can be sent to the terminal equipment corresponding to the staff in the target area, or sent to the terminal equipment corresponding to each individual in the target area.
  • the staff of the company can notify relevant departments to take risk control measures, such as restricting the flow of people, prohibiting individuals with large safety areas from staying for too long, etc.
  • the electronic device obtains the individual data of the individual to be tested for risk prediction, and uses the individual data as the input data of the risk prediction model, and the risk prediction model performs risk prediction according to the individual data to obtain the risk value of the individual to be tested. , the electronic device executes the corresponding security strategy according to the risk value of the individual to be tested.
  • FIG. 5 is a schematic flowchart of a risk management and control method proposed by an embodiment of the present application. The method is applied to the aforementioned electronic device. As shown in FIG. 5 , the risk management and control method in this embodiment may include the following steps:
  • step S501 Obtain individual data of the individual to be tested for risk prediction.
  • step S501 reference may be made to the relevant description of step S201 in the foregoing embodiment, which will not be repeated here.
  • the risk prediction model performs risk prediction according to the individual data, and obtains the risk value of the individual to be tested.
  • the electronic device inputs the individual data used for risk prediction into the risk prediction model, and obtains the risk value of the individual to be tested, where the risk value can be used to represent the probability of the individual to be tested being infected. The higher the probability of being infected, that is, the more dangerous the individual to be tested is, and the lower the risk value, the lower the probability of being infected, that is, the safer the individual to be tested.
  • the risk prediction model may be the aforementioned pre-trained gradient boosted tree model.
  • the risk prediction model may refer to a random forest model
  • the method of obtaining the risk value of the individual to be tested through the risk prediction model may be as follows: the electronic device uses the individual data used for risk prediction as the input of the random forest model
  • Each decision tree included in the random forest model divides the individual data into features to determine the leaf node where the individual data is located in each decision tree, and determines the risk value of the individual to be tested according to the value of each leaf node.
  • the establishment of a random forest model includes N decision trees, and each leaf node where each decision tree in the N decision trees is located corresponds to a value, and the risk value of the individual to be tested is determined according to the value corresponding to each leaf node.
  • the method for determining the risk value of the individual to be tested according to the value corresponding to each leaf node may be: taking the average value of the individual to be tested divided into the corresponding values of each leaf node, and the average value is the value of the individual to be tested. value at risk.
  • the random forest contains three decision trees Tree1, Tree2 and Tree3.
  • Tree1 according to the characteristics, the individual data of the individual to be tested is divided into the c1 node, and the value corresponding to the c1 node is C1; in Tree2, according to the characteristics Division, the individual data of the individual to be tested is divided into the d1 node, and the value corresponding to the d1 node is D1; in Tree3, according to the feature division, the individual data of the individual to be tested is divided into the e1 node, and the value corresponding to the e1 node is E1, That is, the risk value of the individual to be tested is (C1+D1+E1)/3.
  • the specific manner in which the electronic device trains the random forest model may be as follows: the electronic device obtains a training sample set, the training sample set may include multiple training samples, and the training samples may be individual data of the individual to be tested, The electronic device constructs a sample vector corresponding to each training sample in the training sample set, and generates a sample vector set including the sample vector corresponding to each training sample; the electronic device extracts N sample vector subsets from the sample vector set, where N is smaller than the sample vector set The total number of sample vectors in the medium. Specifically, the extraction method may be to randomly extract a plurality of sample vectors from the sample vector set to construct a sample vector subset.
  • a random sampling method may be adopted, including random sampling with replacement, and N repeating the sample vector set.
  • Round extraction take the result of each round of extraction as a sample vector subset, and then obtain N sample vector subsets, in which the N sample vector subsets are independent of each other, and there may be duplicate sample vectors between the sample vector subsets.
  • the sample vectors that have not been extracted can be used to test the trained random forest model to verify the correctness of the random forest model.
  • the number of sample vectors extracted in each round can be obtained according to experience, or determined according to different target areas, which is not limited here; the electronic device uses the random forest algorithm to construct the decision tree according to the subset of sample vectors.
  • a decision tree is constructed by using a subset of sample vectors to obtain N decision trees, each node of the decision tree corresponds to a value, and a random forest is constructed by using N decision trees to obtain a risk prediction model.
  • the electronic device may store the risk value of the individual whose risk prediction has been performed in a database.
  • the electronic device may first query whether the risk value of the individual to be tested is stored in the database. : If it does not exist, perform risk prediction for the individual; if it exists, check whether the risk value satisfies the update condition: if it does not meet the update condition, call the risk value directly; Individuals make risk predictions and update the risk values in the database.
  • the update condition may be the time when the individual to be tested performed risk prediction last time, and if the time exceeds a preset time range, the update condition is satisfied.
  • the specific method for obtaining the risk value of each individual except the individual to be tested in the target area may be the same as the method for obtaining the risk value of the individual to be tested. Refer to step S202 or S502, which will not be repeated here. .
  • S504 Calculate the total risk value according to the risk value of the individual to be tested and the risk value of the individuals other than the individual to be tested in the target area.
  • the total risk value can be obtained by adding the risk value of the individual to be tested and the risk value of each individual except the individual to be tested in the target area.
  • the risk value of the individual to be tested and the risk value of each individual other than the individual to be tested in the target area may be weighted first, and then the weighted total risk value may be obtained by summing the weighted risk values.
  • an individual with a risk value between 0.7 and 1 is given a weight of 1.2
  • an individual with a risk value between 0.6 and 0.4 is given a weight of 1
  • an individual with a risk value between 0 and 0.3 is given a weight of 0.8
  • the preset value may be used to represent the early warning risk value of the target area, and when the total risk value is greater than or equal to the early warning risk value, it indicates that the overall risk of the target area is high, wherein the preset value of The setting method can be obtained according to factors such as the flow of people in the target area, the area, and the interval of the risk value corresponding to each individual in the target area.
  • the method for setting the preset value may further include: determining the total risk value of the target area for each preset time period within the preset time range, and determining the total risk value of each preset time period within the preset time range according to the total risk value of the target area in each preset time period The value determines the risk threshold, the preset value.
  • the total risk value of each preset time period within the preset time range may be the total risk value of each preset time period within the historical time range, for example, the total risk value of each week in the past month.
  • the total risk value for each preset period can be the sum of the risk values of each individual in the target area during this period, or it can also be the sum of the weighted risk values of each individual in the target area for this period.
  • the total risk value can intuitively reflect the change of risk.
  • the specific method of determining the risk threshold can be: discrete the total risk value of each preset time period into multiple sample points, and fit the multiple sample points into a representation of the time to The function of the mapping relationship of the total risk value, and the risk threshold of the target area is determined according to the function.
  • the warning prompt information may be sent to the terminal equipment corresponding to the individual to be tested and the individuals other than the individual to be tested in the target area, or may be sent to the terminal equipment corresponding to the staff in the target area , after receiving the warning information, the staff can notify the relevant departments to take risk control measures, such as prohibiting individuals with higher risk values in the current target area from continuing to stay in the target area.
  • the electronic device obtains the individual data of the individual to be tested for risk prediction, and uses the individual data as the input data of the risk prediction model, and the risk prediction model performs risk prediction according to the individual data to obtain the risk value of the individual to be tested.
  • the electronic device obtains the risk value of the individual except the individual to be tested in the target area, and calculates the total risk according to the risk value of the individual to be tested and the risk value of the individual other than the individual to be tested in the target area value, when the total risk value is greater than or equal to the preset value, the electronic device outputs a warning message.
  • This application involves blockchain technology.
  • individual data and risk value can be encrypted and written into the blockchain.
  • a risk management and control device provided by an embodiment of the present application will be described in detail below with reference to FIG. 6 . It should be noted that the risk management and control device shown in FIG. 6 is used to execute the methods of the embodiments shown in FIG. 2 and FIG. 5 of the present application. For the convenience of description, only the parts related to the embodiments of the present application are shown. The technical details are not disclosed, but the embodiments shown in FIG. 2 and FIG. 5 of the present application are referred to.
  • the apparatus 600 for constructing a knowledge graph may include: an acquisition module 601 , a prediction module 602 , and an execution module 603 .
  • the acquisition module 601 is used to acquire individual data of the individual to be tested for risk prediction, the individual data includes the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the target area where the individual to be tested is located at least one of the characteristic information of ;
  • a prediction module 602 configured to use the individual data as input data of a risk prediction model, and the risk prediction model performs risk prediction according to the individual data to obtain the risk value of the individual to be tested;
  • the execution module 603 is configured to execute the corresponding security prevention strategy according to the risk value of the individual to be tested.
  • the risk prediction model is a pre-trained gradient boosting tree model
  • the prediction module 602 is specifically configured to use the individual data as the input data of the risk prediction model, and the pre-trained gradient
  • Each decision tree included in the boosting tree model performs feature division on the individual data to determine the leaf node where the individual data is located in the decision tree, and determines the individual to be tested according to the value of each leaf node. value at risk.
  • the execution module 603 is specifically configured to determine, according to the risk value of the individual to be tested, the category of the risk group to which the individual to be tested belongs, where the category of the risk group to which the individual to be tested belongs is: When specifying the category of risk groups, the target personnel will be warned.
  • the execution module 603 is specifically configured to determine the safety distance of the individual to be tested according to the risk value of the individual to be tested, and send safety prompt information to the terminal device corresponding to the individual to be tested,
  • the safety prompt information includes the safety distance of the individual to be tested, and the safety prompt information is used to prompt the individual to be tested to maintain at least the safety distance from other individuals.
  • the method further includes: determining a safe area of the individual to be tested according to the safe distance of the individual to be tested, and when detecting that there are other individuals in the safe area of the individual to be tested, The individual to be tested gives an early warning prompt.
  • the method further includes: determining the area of the safety area of each individual in the target area, and calculating the total area of the safety area according to the area of the safety area of each individual in the target area, When the area of the total safe area is greater than or equal to the preset area, first warning prompt information is output, and the first early warning prompt information is used to prompt that the number of individuals in the target area reaches the upper limit, and the preset area is determined according to the area of the target area.
  • the execution module 603 is specifically configured to obtain the risk value of the individual other than the individual to be tested in the target area, according to the risk value of the individual to be tested and the risk value of the individual to be tested in the target area
  • the risk value of the individual other than the individual to be tested is calculated to obtain the total risk value, and when the total risk value is greater than or equal to the preset value, the second early warning prompt information is output, and the second early warning prompt information is used. To indicate that the target area is a high-risk area.
  • the acquisition module 601 acquires individual data of the individual to be tested for risk prediction, and the individual data includes the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the information of the individual to be tested. At least one of the characteristic information of the target area, the prediction module 602 uses the individual data as the input data of the risk prediction model, and the risk prediction model performs risk prediction according to the individual data to obtain the individual data to be tested. Risk value, the execution module 603 is configured to execute the corresponding security prevention policy according to the risk value of the individual to be tested.
  • the electronic device 700 includes: at least one processor 701 , an input device 702 , an output device 703 , a memory 704 , and at least one communication bus 705 .
  • the input device 702 may be a control panel or a microphone, and the output device 703 may be a display screen or the like.
  • Input device 702 may include an input interface, and output device 703 may include an output interface.
  • the memory 704 may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk memory.
  • the communication bus 705 is used to realize the connection and communication between these components.
  • the memory 704 can optionally also be at least one storage device located away from the aforementioned processor 701 .
  • the processor 701 can be combined with the apparatus described in FIG. 3, a set of program codes are stored in the memory 704, and the processor 701, the input device 702, and the output device 703 call the program codes stored in the memory 704 for performing the following operations:
  • the processor 701 is configured to obtain individual data of the individual to be tested for risk prediction, the individual data includes the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the target area where the individual to be tested is located at least one of the characteristic information of ;
  • the processor 701 is configured to use the individual data as input data of a risk prediction model, and the risk prediction model performs risk prediction according to the individual data to obtain the risk value of the individual to be tested;
  • the processor 701 executes a corresponding security protection policy according to the risk value of the individual to be tested.
  • the risk prediction model is a pre-trained gradient boosting tree model
  • the processor 701 is specifically configured to:
  • the individual data is used as the input data of the risk prediction model, and the individual data is characterized by each decision tree included in the pre-trained gradient boosting tree model to determine where the individual data is located in the decision trees. and determine the risk value of the individual to be tested according to the value of each leaf node.
  • the processor 701 is specifically configured to:
  • the processor 701 is specifically configured to:
  • the safety prompt information includes the safety distance of the individual to be tested, and the safety prompt information is used to prompt the individual to be tested and other individuals to keep at least the safe distance.
  • the processor 701 is specifically configured to:
  • an early warning prompt is given to the individual to be tested.
  • the processor 701 is specifically configured to:
  • first warning prompt information is output, and the first early warning prompt information is used to prompt that the number of individuals in the target area reaches the upper limit, and the preset area is determined according to the area of the target area.
  • the processor 701 is specifically configured to:
  • second early warning prompt information is output, and the second early warning prompt information is used to prompt that the target area is a high-risk area.
  • the processor 701 obtains individual data of the individual to be tested for risk prediction, and the individual data includes the identity information of the individual to be tested, the risk factor information of the individual to be tested, and the individual data of the individual to be tested. At least one of the characteristic information of the target area, the processor 701 uses the individual data as the input data of the risk prediction model, and the risk prediction model performs risk prediction according to the individual data, and obtains the individual data to be tested. Risk value, the processor 701 executes the corresponding security prevention policy according to the risk value of the individual to be tested.
  • a highly relevant target associated entity is obtained, and then a highly relevant knowledge graph is constructed, which reduces the workload of building a knowledge graph to a certain extent. , while obtaining more accurate information.
  • the modules described in the embodiments of the present application may be implemented by a general-purpose integrated circuit, such as a CPU (Central Processing Unit, central processing unit), or an ASIC (AppIication Specific Integrated Circuit, application-specific integrated circuit).
  • a general-purpose integrated circuit such as a CPU (Central Processing Unit, central processing unit), or an ASIC (AppIication Specific Integrated Circuit, application-specific integrated circuit).
  • the processor 701 may be a central processing module (Central Processing Unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), Application Specific Integrated Circuit (ASIC), off-the-shelf Programmable Gate Array (FieId-ProgrammabIe Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the communication bus 705 can be an industry standard architecture (Industry Standard Architecture, ISA) bus, a peripheral device interconnect (Periphera I Component, PCI) bus or an extended industry standard architecture (Etended Industry Standard Architecture, EISA) bus, etc., the communication bus 705 can be It is divided into an address bus, a data bus, a control bus, etc. For convenience of presentation, Fig. 7 is only represented by a thick line, but it does not mean that there is only one bus or one type of bus.
  • Industry Standard Architecture ISA
  • PCI peripheral device interconnect
  • EISA Extended Industry Standard Architecture
  • Embodiments of the present application also provide a computer storage medium (Memory), where the computer storage medium is a memory component in an electronic device and is used to store programs and data.
  • the computer storage medium may be a computer-readable storage medium.
  • the computer-readable storage medium stores a computer program, and the computer program includes program instructions that, when executed by a processor, cause the processor to perform some or all of the steps in the above method, which will not be repeated here.
  • the storage medium involved in this application such as a computer-readable storage medium, may be non-volatile or volatile.
  • the computer storage medium here may include both the built-in storage medium in the electronic device, and certainly also the extended storage medium supported by the electronic device.
  • Computer storage media provide storage space in which an electronic device's operating system is stored.
  • one or more instructions suitable for being loaded and executed by the processor 701 are also stored in the storage space, and these instructions may be one or more computer programs (including program codes).
  • the computer storage medium here can be a magnetic disk, an optical disk, a read-only memory (Read-OnIy Memory, ROM) or a random access memory (Random Access Memory, RAM), or a non-stable memory Memory (non-volatile memory), etc., such as at least one disk storage, at least one high-speed RAM storage, and optionally at least one computer storage medium located away from the aforementioned processor.
  • ROM Read-OnIy Memory
  • RAM Random Access Memory
  • non-volatile memory non-volatile memory
  • the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function, and the like; Use the created data, etc.
  • Blockchain essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information to verify its Validity of information (anti-counterfeiting) and generation of the next block.
  • the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Public Health (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Data Mining & Analysis (AREA)
  • Business, Economics & Management (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Epidemiology (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Biomedical Technology (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Development Economics (AREA)
  • Evolutionary Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Alarm Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种风险管控方法、装置、电子设备和存储介质，其中，方法包括，获取待测个体用于风险预测的个体数据（S201），将个体数据作为风险预测模型的输入数据，由风险预测模型根据个体数据进行风险预测，得到待测个体的风险值（S202），根据待测个体的风险值执行对应的安全防范策略（S203）。该方法可以提升传染病防控的实时性，提高管控效率，并提升管控效果。

Description

风险管控方法、装置、电子设备和存储介质
本申请要求于2020年11月12日提交中国专利局、申请号为202011265240.6，发明名称为“一种风险管控方法、装置、电子设备和存储介质”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
本申请涉及智能决策技术领域，具体涉及一种风险管控方法、装置、电子设备和存储介质。
背景技术
传染病暴发是一种严重威胁公众健康和生命安全的公共卫生事件，其具有传播速度快、影响范围广的特点，传染病能够在个体与个体之间通过接触或呼吸相互传播并广泛流行，因此特定区域，如公共场所的传染病风险防控是一项非常重要且艰巨的任务。
发明人意识到，现有的传染病风险管控方法，主要在确定感染者后，通过发布感染者的移动路径以提示他人注意出行安全，从而在一定程度上实现对传染病的风险管控。然而，这种方式不够实时，管控效率较低，管控效果较差。
发明内容
本申请实施例提供了一种风险管控方法、装置、电子设备和存储介质，可以在一定程度上提升传染病防控的实时性，提高管控效率，并提升管控效果。
一方面，本申请实施例提供了一种风险管控方法，所述方法包括：
获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
根据所述待测个体的风险值执行对应的安全防范策略。
一方面，本申请实施例提供了一种风险管控装置，所述装置包括：
获取模块，用于获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
预测模块，用于将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
执行模块，用于根据所述待测个体的风险值执行对应的安全防范策略。
一方面，本申请实施例提供了一种电子设备，包括处理器、输入接口、输出接口和存储器，所述处理器、输入接口、输出接口和存储器相互连接，其中，所述存储器用于存储计算机程序，所述计算机程序包括程序指令，所述处理器被配置用于调用所述程序指令，执行风险管控方法，包括：
获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
根据所述待测个体的风险值执行对应的安全防范策略。
一方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序包括程序指令，所述程序指令当被处理器执行时使所述处理器执行风险管控方法，包括：
获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
根据所述待测个体的风险值执行对应的安全防范策略。
本申请实施例相较于现有技术，是在确定感染者后，通过发布感染者的移动路径以提示他人注意出行安全，无法实时进行风险管控，管控效果较差，本申请通过实施上述方式，可以实时了解风险情况并及时进行风险管控，在一定程度上提升了管控效果和管控效率。
附图说明
为了更清楚地说明本申请实施例技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的一种风险管控系统的网络架构示意图；
图2为本申请实施例提供的一种风险管控方法的流程示意图；
图3为本申请实施例提供的一种风险预测模型的示意图；
图4a为本申请实施例提供的一种预警场景的示意图；
图4b为本申请实施例提供的另一种预警场景的示意图；
图5为本申请实施例提供的另一种风险管控方法的流程示意图；
图6为本申请实施例提供的一种风险管控装置结构示意图；
图7为本申请实施例提供的一种电子设备的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述。
本申请的技术方案涉及人工智能技术领域，如可应用于风险预测等场景中，以实现风险管控，从而推动智慧城市的建设。可选的，本申请涉及的数据如个体数据和/或安全防范策略等可存储于数据库中，或者可以存储于区块链中，比如通过区块链分布式存储，本申请不做限定。
本申请实施例提出的一种风险管控方法，可运用在电子设备中，电子设备可以为终端设备或服务器。此处的终端设备可以为监控设备或可以为智能终端，如：智能手机、笔记本电脑、台式计算机等电子设备。
在一个应用场景中，本申请实施例提出的一种风险管控系统的网络架构，如图1所示，该系统可以包括：智能手机101与服务器102。
智能手机101与服务器102通过网络进行通信。当需要对用户A进行风险预测时，用户A或者目标区域的工作人员可以通过智能手机101向服务器102发送风险预测请求，以触发服务器102获取用户A用于风险预测的个体数据以用于后续的风险预测过程。或者，服务器102可以在通过相应设备监测到用户A进入目标区域时，获取用户A的风险数据以用于后续的风险预测过程。在一个应用场景中，该个体数据的部分数据或全部数据可以从智能手机101获取，也可以采用其他方式获取，在此不做限制。上述过程通过自动采集并处理用于风险预测的个体数据，并通过实施后续的风险管控相关过程，可以提高风险预测效率，以及能够提高风险预测的有效性。
图2为本申请实施例提出的一种风险管控方法的流程示意图，该方法应用于前述提及的电子设备，如图2所示，本实施例中的风险管控方法可以包括以下步骤：
S201：获取待测个体用于风险预测的个体数据。
其中，待测个体可以是指进入目标区域的个体、与传染病患者有密切接触的个体等，个体数据可以是指待测个体的身份信息、待测个体的危险因素信息、待测个体所在目标区域的特征信息中的至少一种。在一种实施例中，身份信息可以是指待测个体的性别、年龄，危险因素信息可以是指待测个体的体温、指示是否经过高风险地区的信息、指示是否有疑 似症状的信息，例如：咳嗽、头晕、发烧等，所在目标区域的特征信息可以是指目标区域的面积、人流量、地理位置等。目标区域可以为某个场所的室内或某个公共区域、该场所如可以为商场或酒吧等场所。
在一种实施例中，电子设备可以是当接收到对待测个体的风险预测请求时，获取待测个体用于风险预测的个体数据。
在一种实施例中，电子设备可以是当通过相关设备监测到待测个体进入目标区域时，获取待测个体用于风险预测的个体数据。例如，相关设备可以为摄像设备等图像采集设备，或为待测个体的终端设备的定位装置等装置等设备，在此不一一列举。该摄像设备等图像采集设备可以设置在目标区域。
在一种实施例中，电子设备可以通过多种信息采集方式收集待测个体的个体数据，其中，可以通过信息关联获取待测个体的身份信息，例如关联身份证信息以获取待测个体的身份信息，还可以通过现场测量、电子问卷方式收集危险因素信息等。
S202：将个体数据作为风险预测模型的输入数据，由风险预测模型根据个体数据进行风险预测，得到待测个体的风险值。
本申请实施例中，电子设备预先训练好风险预测模型，在接收到待测个体的个体数据时，电子设备将个体数据输入到风险预测模型中，由风险预测模型输出待测个体的风险值，风险值可以表示待测个体被传染的概率。
在一种实施例中，风险预测模型可以是指预训练的梯度提升树模型，通过风险预测模型得到待测个体的风险值的过程可以是电子设备将用于风险预测的个体数据作为梯度提升树模型的输入数据，由预训练的梯度提升树模型包含的各决策树对个体数据进行特征划分，以确定个体数据在各决策树所在的叶子节点，并根据各叶子节点对应的数值确定待测个体的风险值。
在一种实施例中，所述梯度提升树模型可以是极端梯度模型XGboost。预训练的梯度提升树模型包括建立的K个决策树，由预训练的梯度提升树模型包含的K决策树对个体数据进行特征划分，以确定个体数据在K决策树中每个决策树所在的叶子节点，并根据个体数据在每个决策树所在叶子节点对应的数值确定待测个体的风险值。具体的，电子设备根据各叶子节点对应的数值确定待测个体的风险值的方法可以为：电子设备将各叶子节点对应的数值相加，输出的总的数值，该总的数值即为该待测个体的风险值。再如，如图3所示，预训练的梯度提升树模型含有两个决策树tree1和tree2，在tree1中，根据特征划分，待测个体1的个体数据被划分到a1节点，a1节点对应的数值为A1，待测个体2的个体数据被划分到a2节点，a2节点对应的数值为A2；在tree2中，待测个体1的个体数据和待测个体2的个体数据被划分到b1节点，b1节点对应的数值为B1，即待测个体1的风险值
Figure PCTCN2021096646-appb-000001
Figure PCTCN2021096646-appb-000002
和待测个体2的风险值
Figure PCTCN2021096646-appb-000003
在一种实施例中，对梯度提升树模型进行训练的具体方式可以为：获取训练样本集，该训练样本集可以包括多个训练样本，训练样本可以为待测个体的个体数据。假设，训练样本集为{(x 1,y 1),(x 2,y 2)…(x n,y n)}，其中x n表示训练样本，y n表示训练样本对应的真实值。损失函数为
Figure PCTCN2021096646-appb-000004
正则化项为Ω(f k)。梯度提升树模型的优化目标是最小化目标函数：
Figure PCTCN2021096646-appb-000005
其中，∑ kΩ(f k)表示K棵树的复杂度，i表示第i个样本，k表示第k棵树，
Figure PCTCN2021096646-appb-000006
表示第i个样本x i的预测值,且
Figure PCTCN2021096646-appb-000007
电子设备可以利用训练样本集构建梯度提升模型的K颗决策树，并生成包括该K棵决策树的梯度提升模型作为预训练的梯度提升模型。
S203：根据待测个体的风险值执行对应的安全防范策略。
在一种实施例中，执行对应的安全防范策略可以为，电子设备根据待测个体的风险值，确定待测个体所属的风险人群类别，例如，风险人群类别可以为低风险人群、中风险人群、 高风险人群。在待测个体所属的风险人群类别为指定风险人群类别时，电子设备对目标人员进行预警提示。其中，目标人员可以是待测个体，还可以是待测个体所在目标区域的工作人员，进行预警提示可以是发送至目标人员对应的终端设备，预警提示可以用于提示该待测个体为指定风险人群类别，不允许该待测个体进行该目标区域等。
在一个实施例中，电子设备根据待测个体的风险值，确定待测个体所属的风险人群类别的过程可以为电子设备可以根据风险值与风险人群类别的对应关系，确定待测个体的风险等级对应的风险人群类型，并将待测个体的风险等级对应的风险等级类别确定为待测个体所属的风险人群类别。
在一种实施例中，执行对应的安全范围策略还可以为，电子设备根据待测个体的风险值确定待测个体的安全距离，并发送安全提示信息至待测个体对应的终端设备，例如通过短信发送至终端设备，安全提示信息至少包括待测个体的安全距离，该安全提示信息用于提示待测个体与其他个体至少需要保持的安全距离。
在一种实施例中，根据风险值确定安全距离的具体方式可以为，获取一些离散的样本(Q1,R1),(Q2,R2),…(Qn,Rn)，该离散样本表示风险值为Q的个体应有安全距离R，将风险值Q作为自变量，R作为因变量，使用函数拟合方法，计算出一个表示风险值到安全距离的映射关系的表达式，将待测个体的风险值输入该表达式，计算得到待测个体的安全距离。
在一种实施例中，执行对应的安全范围策略还可以为，电子设备在确定出待测个体的安全距离之后，根据待测个体的安全距离确定待测个体的安全区域，当监测到待测个体的安全区域内存在其他个体时，电子设备对待测个体进行预警提示，其中，安全区域可以为以安全距离为半径的圆，进行预警提示的方式可以是发送提示信息至该待测个体的对应的终端设备，该提示信息可以用于提示该待测个体安全区域内存在其他个体，如图4a所示。
在一种实施例中，执行对应的安全范围策略还可以为，电子设备在确定出待测个体的安全区域之后，确定目标区域内各个体的安全区域的面积，电子设备根据目标区域内各个体的安全区域的面积，计算得到总的安全区域的面积，当总的安全区域的面积大于或等于预设面积时，电子设备输出预警提示信息，预警提示信息用于提示该目标区域内个体数达到上限，该预设面积是根据该目标区域的面积确定出的。其中，总的安全区域的面积可以为各个体的安全区域的面积之和，即
Figure PCTCN2021096646-appb-000008
预设面积可以表示达到目标区域预警条件时的面积，即S alert＝S×α，如图4b所示，其中，S为目标区域的总面积，α为一个可调节参数，反映目标区域对风险的敏感程度，α越大，目标区域的敏感程度越低，代表目标区域能承受的总人数越多；α越小，目标区域的敏感程度越高，代表目标区域能承受的总人数越少，通过α能够灵活调整目标区域达到预警条件的面积，预警提示信息可以是发送到该目标区域的工作人员对应的终端设备，还可以是发送到该目标区域中各个体对应的终端设备，该目标区域的工作人员收到预警提示信息之后，可以通知相关部门采取风险控制措施，例如，限制人流、禁止安全区域面积较大的个体停留时间过长等。
本申请实施例中，电子设备获取待测个体用于风险预测的个体数据，并将个体数据作为风险预测模型的输入数据，由风险预测模型根据个体数据进行风险预测，得到待测个体的风险值，电子设备根据待测个体的风险值执行对应的安全防范策略。通过实施上述方法，可以有效提升风险防控的实时性，并在一定程度上提升管控效率和管控效果。
图5为本申请实施例提出的一种风险管控方法的流程示意图，该方法应用于前述提及的电子设备，如图5所示，本实施例中的风险管控方法可以包括以下步骤：
S501：获取待测个体用于风险预测的个体数据。其中，步骤S501的具体实施方式可以参见上述实施例中步骤S201的相关描述，此处不再赘述。
S502：将个体数据作为风险预测模型的输入数据，由风险预测模型根据个体数据进行 风险预测，得到待测个体的风险值。
本申请实施例中，电子设备将用于风险预测的个体数据输入风险预测模型，得到待测个体的风险值，其中，风险值可以用于表示待测个体被传染的概率，风险值越高表明被传染的概率越高，即待测个体越危险，风险值越低表明被传染的概率越低，即待测个体越安全。
在一个实施例中，该风险预测模型可以为前述提及的预训练的梯度提升树模型。
在一种实施例中，风险预测模型可以是指随机森林模型，通过该风险预测模型得到待测个体的风险值的方式可以如下：电子设备将用于风险预测的个体数据作为随机森林模型的输入数据，由随机森林模型包含的各决策树对个体数据进行特征划分，以确定个体数据在各决策树所在的叶子节点，并根据各叶子节点的数值确定待测个体的风险值。例如，随机森林模型建立包括N个决策树，并且N个决策树中每个决策树所在的每个叶子节点对应一个数值，根据各叶子节点对应的数值确定待测个体的风险值。具体的，根据各叶子节点对应的数值确定待测个体的风险值的方法可以为：将待测个体所划分到各叶子节点对应的数值取平均值，该平均值值即为该待测个体的风险值。再如，随机森林包含三个决策树Tree1、Tree2和Tree3，在Tree1中，根据特征划分，待测个体的个体数据被划分到c1节点，c1节点对应的数值为C1；在Tree2中，根据特征划分，待测个体的个体数据被划分到d1节点，d1节点对应的数值为D1；在Tree3中，根据特征划分，待测个体的个体数据被划分到e1节点，e1节点对应的数值为E1，即待测个体的风险值为(C1+D1+E1)/3。
在一种实施例中，电子设备对随机森林模型进行训练的具体方式可以如下：电子设备获取训练样本集，该训练样本集可以包括多个训练样本，训练样本可以为待测个体的个体数据，电子设备构建训练样本集中各个训练样本对应的样本向量，并生成包括各个训练样本对应的样本向量的样本向量集合；电子设备从样本向量集合中抽取N个样本向量子集合，其中N小于样本向量集合中样本向量总数。具体的，抽取方式可以是从样本向量集合中随机抽取多个样本向量以构建样本向量子集合，更具体的，可以采取随机采样的方式，有放回的随机抽样，重复对样本向量集合进行N轮抽取，将每一轮抽取结果作为一个样本向量子集合，进而得到N个样本向量子集合，其中N个样本向量子集合之间相互独立，样本向量子集合之间可以存在重复的样本向量，其中，在样本向量集合中，未被抽取的样本向量可以用于对训练好的随机森林模型进行测试，以验证随机森林模型的正确性。需要说明的是，每轮抽取的样本向量的数量具体可以根据经验进行获取，或者根据不同的目标区域确定，此处不作限制；电子设备根据样本向量子集合使用随机森林算法进行决策树的构建，利用一个样本向量子集合构建一个决策树，从而得到N个决策树，每个决策树的节点都对应一个数值，利用N个决策树构建随机森林，得到风险预测模型。
在一个实施例中，电子设备可以将已进行风险预测的个体的风险值存储于数据库中，当电子设备对待测个体进行风险预测时，可以先查询数据库中是否储存有该待测个体的风险值：若不存在，则对该个体进行风险预测；若存在，则检测该风险值是否满足更新条件：若不满足更新条件，则直接调用该风险值；若满足更新条件，就重新对该待测个体进行风险预测，并将数据库中的风险值进行更新。其中，更新条件可以是检测该待测个体上次进行风险预测的时间，若时间超过预设的时间范围，则满足更新条件。或者，还可以为检测该待测个体的个体数据中重要的个体数据是否发生变化，若发生变化，则满足更新条件。例如，该待测个体上一次进行风险预测时未进入过高风险地区，此次进行风险预测时进入过高风险地区，则满足更新条件；或者，该待测个体上一次进行风险预测时未密切接触过传染病感染者，此次进行风险预测时密切接触过传染病感染者，则满足更新条件。
S503：获取目标区域内除待测个体之外的个体的风险值。
在一种实施例中，获取目标区域内除待测个体之外的各个体的风险值的具体方法可以 同获取待测个体的风险值的方式，可以参见步骤S202或者S502，此处不再赘述。
S504：根据待测个体的风险值以及所述目标区域内除待测个体之外的个体的风险值，计算得到总的风险值。
在一种实施例中，总的风险值可以通过将待测个体的风险值和目标区域内除待测个体之外的各个体的风险值相加得到。或者，还可以先对待测个体的风险值和目标区域内除待测个体之外的各个体的风险值进行加权，再将各加权后的风险值求和得到加权总风险值。例如，风险值在0.7～1之间的个体权重为1.2，风险值在0.6～0.4之间的个体权重为1，风险值为0～0.3之间的个体权重为0.8，因此风险值为0.1的待测个体1、风险值为0.5的待测个体2和风险值为0.7的待测个体3的加权总风险值为0.1*0.8+0.5*1+0.7*1.2＝1.42。
S505：当总的风险值大于或等于预设值时，输出预警提示信息。
在一种实施例中，预设值可以用于表示目标区域的预警风险值，当总的风险值大于或等于预警风险值时，表示该目标区域整体的风险偏高，其中，预设值的设置方法可以根据综合目标区域的人流量、面积、目标区域中各个体对应的风险值所处区间等因素得到。
在一个实施例中，预设值的设置方法还可以为，确定目标区域在预设时间范围内每一预设时段的风险总值，并根据预设时间范围内每一预设时段的风险总值确定风险阈值，即预设值。其中，预设时间范围内每一预设时段的风险总值可以是历史时间范围内每一预设时段的风险总值，例如过去一个月内每一个星期的风险总值。每一预设时段的风险总值可以是此期间目标区域内的各个体的风险值之和，或者，还可以为此期间目标区域内的各个体的加权风险值之和，每一预设时段的风险总值可以直观反映风险变化情况，确定风险阈值的具体方式可以为：将每个预设时段的风险总值离散成多个样本点，并基于多个样本点拟合成一个表示时间到风险总值的映射关系的函数，根据函数确定出目标区域的风险阈值。
在一种实施例中，预警提示信息可以是发送至该目标区域中待测个体和除待测个体之外的个体对应的终端设备，还可以是发送至该目标区域的工作人员对应的终端设备，工作人员收到预警提示信息之后，可以通知相关部门采取风险控制措施，例如，禁止目前目标区域中为较高风险值的个体继续停留在该目标区域。
本申请实施例中，电子设备获取待测个体用于风险预测的个体数据，并将个体数据作为风险预测模型的输入数据，由风险预测模型根据个体数据进行风险预测，得到待测个体的风险值，电子设备获取目标区域内除待测个体之外的个体的风险值，并根据待测个体的风险值以及所述目标区域内除待测个体之外的个体的风险值，计算得到总的风险值，当总的风险值大于或等于预设值时，电子设备输出预警提示信息。通过实施上述方式，可以实时了解目标区域的风险变化情况，并根据风险变化情况及时进行管控，提高了目标区域风险管控的效率和管控效果。
本申请涉及区块链技术，如可将个体数据与风险值一并加密后写入区块链。
下面将结合附图6对本申请实施例提供的一种风险管控装置进行详细介绍。需要说明的是，附图6所示的风险管控装置，用于执行本申请图2和图5所示实施例的方法，为了便于说明，仅示出了与本申请实施例相关的部分，具体技术细节未揭示，经参照本申请图2和图5所示的实施例。
请参见图6，为本申请提供的一种风险管控装置的结构示意图，该构建知识图谱的装置600可包括：获取模块601、预测模块602、执行模块603。
获取模块601，用于获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
预测模块602，用于将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
执行模块603，用于根据所述待测个体的风险值执行对应的安全防范策略。
在一种实施例中，所述风险预测模型为预训练的梯度提升树模型，所述预测模块602，具体用于将所述个体数据作为风险预测模型的输入数据，由所述预训练的梯度提升树模型包括的各决策树对所述个体数据进行特征划分，以确定所述个体数据在所述各决策树所在的叶子节点，并根据所述各叶子节点的数值确定所述待测个体的风险值。
在一种实施例中，所述执行模块603，具体用于根据所述待测个体的风险值，确定所述待测个体所属的风险人群类别，在所述待测个体所属的风险人群类别为指定风险人群类别时，对目标人员进行预警提示。
在一种实施例中，所述执行模块603，具体用于根据所述待测个体的风险值确定所述待测个体的安全距离，发送安全提示信息至所述待测个体对应的终端设备，所述安全提示信息包括所述待测个体的安全距离，所述安全提示信息用于提示所述待测个体与其他个体至少保持所述安全距离。
在一种实施例中，所述方法还包括，根据所述待测个体的安全距离确定所述待测个体的安全区域，当检测到所述待测个体的安全区域内存在其他个体时，对所述待测个体进行预警提示。
在一种实施例中，所述方法还包括，确定所述目标区域内各个体的安全区域的面积，根据所述目标区域内各个体的安全区域的面积，计算得到总的安全区域的面积，当所述总的安全区域的面积大于或等于预设面积时，输出第一预警提示信息，所述第一预警提示信息用于提示所述目标区域内的个体数达到上限，所述预设面积是根据所述目标区域的面积确定出的。
在一种实施例中，执行模块603，具体用于获取所述目标区域内除所述待测个体之外的个体的风险值，根据所述待测个体的风险值以及所述目标区域内除所述待测个体之外的个体的风险值，计算得到总的风险值，当所述总的风险值大于或等于预设值时，输出第二预警提示信息，所述第二预警提示信息用于提示所述目标区域为高风险区域。
本申请实施例中，获取模块601获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种，预测模块602将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值，执行模块603用于根据所述待测个体的风险值执行对应的安全防范策略。通过实施上述方法，可以在一定程度上提升风险管控的实时性，以及有效提高管控效率和管控效果。
请参见图7，为本申请实施例提供的一种电子设备的结构示意图。如图7所示，该电子设备700包括：至少一个处理器701、输入设备702、输出设备703、存储器704、至少一个通信总线705。其中，输入设备702可以是控制面板或者麦克风等，输出设备703可以是显示屏等。输入设备702可包括输入接口，输出设备703可包括输出接口。其中，存储器704可以是高速RAM存储器，也可以是非不稳定的存储器(non-voIatiIe memory)，例如至少一个磁盘存储器。其中，通信总线705用于实现这些组件之间的连接通信。存储器704可选的还可以是至少一个位于远离前述处理器701的存储装置。其中处理器701可以结合图3所描述的装置，存储器704中存储一组程序代码，且处理器701，输入设备702，输出设备703调用存储器704中存储的程序代码，用于执行以下操作：
处理器701，用于获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
处理器701，用于将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
处理器701，根据所述待测个体的风险值执行对应的安全防范策略。
在一种实施例中，所述风险预测模型为预训练的梯度提升树模型，所述处理器701，具体用于：
将所述个体数据作为风险预测模型的输入数据，由所述预训练的梯度提升树模型包括的各决策树对所述个体数据进行特征划分，以确定所述个体数据在所述各决策树所在的叶子节点，并根据所述各叶子节点的数值确定所述待测个体的风险值。
在一种实施例中，所述处理器701，具体用于：
根据所述待测个体的风险值，确定所述待测个体所属的风险人群类别；
在所述待测个体所属的风险人群类别为指定风险人群类别时，对目标人员进行预警提示。
在一种实施例中，所述处理器701，具体用于：
根据所述待测个体的风险值确定所述待测个体的安全距离；
发送安全提示信息至所述待测个体对应的终端设备，所述安全提示信息包括所述待测个体的安全距离，所述安全提示信息用于提示所述待测个体与其他个体至少保持所述安全距离。
在一种实施例中，所述处理器701，具体用于：
根据所述待测个体的安全距离确定所述待测个体的安全区域；
当检测到所述待测个体的安全区域内存在其他个体时，对所述待测个体进行预警提示。
在一种实施例中，所述处理器701，具体用于：
确定所述目标区域内各个体的安全区域的面积；
根据所述目标区域内各个体的安全区域的面积，计算得到总的安全区域的面积；
当所述总的安全区域的面积大于或等于预设面积时，输出第一预警提示信息，所述第一预警提示信息用于提示所述目标区域内的个体数达到上限，所述预设面积是根据所述目标区域的面积确定出的。
在一种实施例中，所述基于所述目标实体和各个所述目标关联实体构建目标知识图谱之后，所述处理器701，具体用于：
获取所述目标区域内除所述待测个体之外的个体的风险值；
根据所述待测个体的风险值以及所述目标区域内除所述待测个体之外的个体的风险值，计算得到总的风险值；
当所述总的风险值大于或等于预设值时，输出第二预警提示信息，所述第二预警提示信息用于提示所述目标区域为高风险区域。
本申请实施例中，处理器701获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种，处理器701将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值，处理器701根据所述待测个体的风险值执行对应的安全防范策略。通过实施上述方式，基于目标实体和目标关联实体之间的相关性，获取到高相关性的目标关联实体，进而构建出高相关性的知识图谱，在一定程度上减少了构建知识图谱的工作量，同时获得了更加准确的信息。
本申请实施例中所述模块，可以通过通用集成电路，例如CPU(CentraI Processing Unit，中央处理器)，或通过ASIC(AppIication Specific Integrated Circuit，专用集成电路)来实现。
应当理解，在本申请实施例中，所称处理器701可以是中央处理模块(CentraI Processing Unit，CPU)，该处理器还可以是其他通用处理器、数字信号处理器(DigitaISignaI Processor，DSP)、专用集成电路(AppIication Specific Integrated Circuit，ASIC)、现成可编程门阵列 (FieId-ProgrammabIe Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
通信总线705可以是工业标准体系结构(Industry Standard Architecture，ISA)总线、外部设备互联(PeripheraI Component，PCI)总线或扩展工业标准体系结构(EItended Industry Standard Architecture，EISA)总线等，该通信总线705可以分为地址总线、数据总线、控制总线等，为便于表示，图7仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
本申请实施例还提供了一种计算机存储介质(Memory)，所述计算机存储介质是电子设备中的记忆部件，用于存放程序和数据。该计算机存储介质可以是计算机可读存储介质。计算机可读存储介质存储有计算机程序，所述计算机程序包括程序指令，所述程序指令当被处理器执行时使所述处理器执行上述方法中的部分或全部步骤，此处不赘述。
可选的，本申请涉及的存储介质如计算机可读存储介质可以是非易失性的，也可以是易失性的。
可以理解的是，此处的计算机存储介质既可以包括电子设备中的内置存储介质，当然也可以包括电子设备所支持的扩展存储介质。计算机存储介质提供存储空间，该存储空间存储了电子设备的操作系统。并且，在该存储空间中还存放了适于被处理器701加载并执行的一条或多条的指令，这些指令可以是一个或一个以上的计算机程序(包括程序代码)。需要说明的是，此处的计算机存储介质可以是磁碟、光盘、只读存储记忆体(Read-OnIy Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)，也可以是非不稳定的存储器(non-volatile memory)等，例如至少一个磁盘存储器、至少一个高速RAM存储器，可选的还可以是至少一个位于远离前述处理器的计算机存储介质。
所述的计算机可读存储介质可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序等；存储数据区可存储根据区块链节点的使用所创建的数据等。
其中，本申请所指区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain)，本质上是一个去中心化的数据库，是一串使用密码学方法相关联产生的数据块，每一个数据块中包含了一批次网络交易的信息，用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层等。
上面结合附图对本申请的实施例进行了描述，但是以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。以上实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种风险管控方法，包括：
    获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
    将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
    根据所述待测个体的风险值执行对应的安全防范策略。
  2. 如权利要求1所述的方法，其中，所述风险预测模型为预训练的梯度提升树模型，所述将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值，包括：
    将所述个体数据作为风险预测模型的输入数据，由所述预训练的梯度提升树模型包括的各决策树对所述个体数据进行特征划分，以确定所述个体数据在所述各决策树所在的叶子节点，并根据所述各叶子节点的数值确定所述待测个体的风险值。
  3. 如权利要求1或2所述的方法，其中，所述根据所述待测个体的风险值执行对应的安全防范策略，包括：
    根据所述待测个体的风险值，确定所述待测个体所属的风险人群类别；
    在所述待测个体所属的风险人群类别为指定风险人群类别时，对目标人员进行预警提示。
  4. 如权利要求1或2所述的方法，其中，所述根据所述待测个体的风险值执行对应的安全防范策略，包括：
    根据所述待测个体的风险值确定所述待测个体的安全距离；
    发送安全提示信息至所述待测个体对应的终端设备，所述安全提示信息包括所述待测个体的安全距离，所述安全提示信息用于提示所述待测个体与其他个体至少保持所述安全距离。
  5. 如权利要求4所述的方法，其中，所述方法还包括：
    根据所述待测个体的安全距离确定所述待测个体的安全区域；
    当检测到所述待测个体的安全区域内存在其他个体时，对所述待测个体进行预警提示。
  6. 如权利要求5所述的方法，其中，所述方法还包括：
    确定所述目标区域内各个体的安全区域的面积；
    根据所述目标区域内各个体的安全区域的面积，计算得到总的安全区域的面积；
    当所述总的安全区域的面积大于或等于预设面积时，输出第一预警提示信息，所述第一预警提示信息用于提示所述目标区域内的个体数达到上限，所述预设面积是根据所述目标区域的面积确定出的。
  7. 如权利要求1所述的方法，其中，所述根据所述待测个体的风险值执行对应的安全防范策略，包括：
    获取所述目标区域内除所述待测个体之外的个体的风险值；
    根据所述待测个体的风险值以及所述目标区域内除所述待测个体之外的个体的风险值，计算得到总的风险值；
    当所述总的风险值大于或等于预设值时，输出第二预警提示信息，所述第二预警提示信息用于提示所述目标区域为高风险区域。
  8. 一种风险管控装置，包括：
    获取模块，用于获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
    预测模块，用于将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
    执行模块，用于根据所述待测个体的风险值执行对应的安全防范策略。
  9. 一种电子设备，包括处理器、输入接口、输出接口和存储器，所述处理器、输入接口、输出接口和存储器相互连接，其中，所述存储器用于存储计算机程序，所述计算机程序包括程序指令，所述处理器被配置用于调用所述程序指令，执行以下方法：
    获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
    将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
    根据所述待测个体的风险值执行对应的安全防范策略。
  10. 如权利要求9所述的电子设备，其中，所述风险预测模型为预训练的梯度提升树模型，执行所述将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值，包括：
    将所述个体数据作为风险预测模型的输入数据，由所述预训练的梯度提升树模型包括的各决策树对所述个体数据进行特征划分，以确定所述个体数据在所述各决策树所在的叶子节点，并根据所述各叶子节点的数值确定所述待测个体的风险值。
  11. 如权利要求9或10所述的电子设备，其中，执行所述根据所述待测个体的风险值执行对应的安全防范策略，包括：
    根据所述待测个体的风险值，确定所述待测个体所属的风险人群类别；在所述待测个体所属的风险人群类别为指定风险人群类别时，对目标人员进行预警提示；或，
    根据所述待测个体的风险值确定所述待测个体的安全距离；发送安全提示信息至所述待测个体对应的终端设备，所述安全提示信息包括所述待测个体的安全距离，所述安全提示信息用于提示所述待测个体与其他个体至少保持所述安全距离。
  12. 如权利要求11所述的电子设备，其中，所述处理器还用于执行：
    根据所述待测个体的安全距离确定所述待测个体的安全区域；
    当检测到所述待测个体的安全区域内存在其他个体时，对所述待测个体进行预警提示。
  13. 如权利要求12所述的电子设备，其中，所述处理器还用于执行：
    确定所述目标区域内各个体的安全区域的面积；
    根据所述目标区域内各个体的安全区域的面积，计算得到总的安全区域的面积；
    当所述总的安全区域的面积大于或等于预设面积时，输出第一预警提示信息，所述第一预警提示信息用于提示所述目标区域内的个体数达到上限，所述预设面积是根据所述目标区域的面积确定出的。
  14. 如权利要求9所述的电子设备，其中，执行所述根据所述待测个体的风险值执行对应的安全防范策略，包括：
    获取所述目标区域内除所述待测个体之外的个体的风险值；
    根据所述待测个体的风险值以及所述目标区域内除所述待测个体之外的个体的风险值，计算得到总的风险值；
    当所述总的风险值大于或等于预设值时，输出第二预警提示信息，所述第二预警提示信息用于提示所述目标区域为高风险区域。
  15. 一种计算机可读存储介质，其中，所述计算机可读存储介质存储有计算机程序，所述计算机程序包括程序指令，所述程序指令当被处理器执行时使所述处理器执行以下方法：
    获取待测个体用于风险预测的个体数据，所述个体数据包括所述待测个体的身份信息、 所述待测个体的危险因素信息、所述待测个体所在目标区域的特征信息中的至少一种；
    将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值；
    根据所述待测个体的风险值执行对应的安全防范策略。
  16. 如权利要求15所述的计算机可读存储介质，其中，所述风险预测模型为预训练的梯度提升树模型，执行所述将所述个体数据作为风险预测模型的输入数据，由所述风险预测模型根据所述个体数据进行风险预测，得到所述待测个体的风险值，包括：
    将所述个体数据作为风险预测模型的输入数据，由所述预训练的梯度提升树模型包括的各决策树对所述个体数据进行特征划分，以确定所述个体数据在所述各决策树所在的叶子节点，并根据所述各叶子节点的数值确定所述待测个体的风险值。
  17. 如权利要求15或16所述的计算机可读存储介质，其中，执行所述根据所述待测个体的风险值执行对应的安全防范策略，包括：
    根据所述待测个体的风险值，确定所述待测个体所属的风险人群类别；在所述待测个体所属的风险人群类别为指定风险人群类别时，对目标人员进行预警提示；或，
    根据所述待测个体的风险值确定所述待测个体的安全距离；发送安全提示信息至所述待测个体对应的终端设备，所述安全提示信息包括所述待测个体的安全距离，所述安全提示信息用于提示所述待测个体与其他个体至少保持所述安全距离。
  18. 如权利要求17所述的计算机可读存储介质，其中，所述程序指令当被处理器执行时还使所述处理器执行：
    根据所述待测个体的安全距离确定所述待测个体的安全区域；
    当检测到所述待测个体的安全区域内存在其他个体时，对所述待测个体进行预警提示。
  19. 如权利要求18所述的计算机可读存储介质，其中，所述程序指令当被处理器执行时还使所述处理器执行：
    确定所述目标区域内各个体的安全区域的面积；
    根据所述目标区域内各个体的安全区域的面积，计算得到总的安全区域的面积；
    当所述总的安全区域的面积大于或等于预设面积时，输出第一预警提示信息，所述第一预警提示信息用于提示所述目标区域内的个体数达到上限，所述预设面积是根据所述目标区域的面积确定出的。
  20. 如权利要求15所述的计算机可读存储介质，其中，执行所述根据所述待测个体的风险值执行对应的安全防范策略，包括：
    获取所述目标区域内除所述待测个体之外的个体的风险值；
    根据所述待测个体的风险值以及所述目标区域内除所述待测个体之外的个体的风险值，计算得到总的风险值；
    当所述总的风险值大于或等于预设值时，输出第二预警提示信息，所述第二预警提示信息用于提示所述目标区域为高风险区域。
PCT/CN2021/096646 2020-11-12 2021-05-28 风险管控方法、装置、电子设备和存储介质 WO2022100062A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011265240.6 2020-11-12
CN202011265240.6A CN112382407A (zh) 2020-11-12 2020-11-12 一种风险管控方法、装置、电子设备和存储介质

Publications (1)

Publication Number Publication Date
WO2022100062A1 true WO2022100062A1 (zh) 2022-05-19

Family

ID=74583638

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/096646 WO2022100062A1 (zh) 2020-11-12 2021-05-28 风险管控方法、装置、电子设备和存储介质

Country Status (2)

Country Link
CN (1) CN112382407A (zh)
WO (1) WO2022100062A1 (zh)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112382407A (zh) * 2020-11-12 2021-02-19 平安科技（深圳）有限公司 一种风险管控方法、装置、电子设备和存储介质
CN113113154A (zh) * 2021-04-16 2021-07-13 南方科技大学 一种传染病防控方法、装置、计算机设备及存储介质
CN113688251B (zh) * 2021-07-27 2024-02-13 广东师大维智信息科技有限公司 一种室内体育赛事安保领域的知识图谱构建方法与系统
CN113707338B (zh) * 2021-10-28 2022-08-30 南方科技大学 景区疫情风险预测与限流方法、装置、设备和存储介质
CN114360737A (zh) * 2021-12-20 2022-04-15 深圳云天励飞技术股份有限公司 确定个体风险的方法及装置
CN114580979B (zh) * 2022-05-07 2022-08-02 中国科学院地理科学与资源研究所 高温防灾指数检测方法、装置、设备、存储介质及产品
CN115526434B (zh) * 2022-11-07 2023-07-28 广东中思拓大数据研究院有限公司 对象信息预测方法、装置、计算机设备和存储介质
CN117273467B (zh) * 2023-11-17 2024-01-26 江苏麦维智能科技有限公司 一种基于多因素耦合的工业安全风险管控方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107256327A (zh) * 2017-05-05 2017-10-17 中国科学院深圳先进技术研究院 一种传染病防控方法及系统
CN111403047A (zh) * 2020-03-17 2020-07-10 腾讯科技(成都)有限公司 疫情提示方法、装置、计算机设备和存储介质
CN111785380A (zh) * 2020-07-01 2020-10-16 医渡云（北京）技术有限公司 传染性疾病患病风险等级的预测方法及装置、介质、设备
CN111863280A (zh) * 2020-07-30 2020-10-30 深圳前海微众银行股份有限公司 健康检测方法、系统、终端设备及存储介质
CN112382407A (zh) * 2020-11-12 2021-02-19 平安科技（深圳）有限公司 一种风险管控方法、装置、电子设备和存储介质

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SG10201506012SA (en) * 2015-07-31 2017-02-27 Accenture Global Services Ltd Inventory, growth, and risk prediction using image processing
CN108172301B (zh) * 2018-01-31 2021-02-02 中国科学院软件研究所 一种基于梯度提升树的蚊媒传染病疫情预测方法及系统
CN110163481A (zh) * 2019-04-19 2019-08-23 深圳壹账通智能科技有限公司 电子装置、用户风控审核系统测试方法及存储介质
CN111354472A (zh) * 2020-02-20 2020-06-30 戴建荣 一种传染病传播监测预警系统和方法
CN111403048A (zh) * 2020-03-18 2020-07-10 唐宓 一种未知传染病预警及追溯方法
CN111524611B (zh) * 2020-04-24 2023-03-03 腾讯科技（深圳）有限公司 构建传染病趋势预测模型的方法、预测方法、装置及设备
CN111863271B (zh) * 2020-06-08 2024-03-12 浙江大学 一种新冠肺炎的重大传染病传播风险预警及防控分析系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107256327A (zh) * 2017-05-05 2017-10-17 中国科学院深圳先进技术研究院 一种传染病防控方法及系统
CN111403047A (zh) * 2020-03-17 2020-07-10 腾讯科技(成都)有限公司 疫情提示方法、装置、计算机设备和存储介质
CN111785380A (zh) * 2020-07-01 2020-10-16 医渡云（北京）技术有限公司 传染性疾病患病风险等级的预测方法及装置、介质、设备
CN111863280A (zh) * 2020-07-30 2020-10-30 深圳前海微众银行股份有限公司 健康检测方法、系统、终端设备及存储介质
CN112382407A (zh) * 2020-11-12 2021-02-19 平安科技（深圳）有限公司 一种风险管控方法、装置、电子设备和存储介质

Also Published As

Publication number Publication date
CN112382407A (zh) 2021-02-19

Similar Documents

Publication Publication Date Title
WO2022100062A1 (zh) 风险管控方法、装置、电子设备和存储介质
Mardani et al. A novel extended approach under hesitant fuzzy sets to design a framework for assessing the key challenges of digital health interventions adoption during the COVID-19 outbreak
Lin et al. Using machine learning to assist crime prevention
US10438308B2 (en) Systems and methods for identifying entities using geographical and social mapping
US9412141B2 (en) Systems and methods for identifying entities using geographical and social mapping
US20180096253A1 (en) Rare event forecasting system and method
WO2014150987A1 (en) Systems and methods for identifying entites using geographical and social mapping
CN113034145B (zh) 用户异常加密数字资产交易类别判断方法、装置
CN112216402A (zh) 基于人工智能的疫情预测方法、装置、计算机设备及介质
US10062034B2 (en) Method and system for obtaining and analyzing information from a plurality of sources
CN112712903A (zh) 一种基于人机物三元空间协同感知的传染病监测方法
KR102247188B1 (ko) 상황인식 및 퍼지이론 기반의 해수욕장 위험도 평가 시스템 및 방법과, 이를 위한 컴퓨터 프로그램
CN108092985A (zh) 网络安全态势分析方法、装置、设备及计算机存储介质
WO2022142903A1 (zh) 身份识别方法、装置、电子设备及相关产品
Grover et al. Prediction model for influenza epidemic based on Twitter data
CN107292174A (zh) 一种云计算系统安全性评估方法及装置
CN111754241A (zh) 一种用户行为感知方法、装置、设备及介质
CN111815177A (zh) 消防安全评估方法、服务器、系统及存储介质
Thomas et al. Real-time prediction of severe influenza epidemics using Extreme Value Statistics
Zheng et al. Predictive study of tuberculosis incidence by time series method and Elman neural network in Kashgar, China
CN114254867A (zh) 一种电信诈骗受害人风险评估系统及方法
CN110321582B (zh) 一种传销活动分析方法及装置
Patching et al. A supervised learning process to validate online disease reports for use in predictive models
Li et al. Dynamic risk assessment of emergency evacuation in large public buildings: A case study
Salamanis et al. A probabilistic framework for the reliability assessment of crowd sourcing urban traffic reports

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: 21890593

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: 21890593

Country of ref document: EP

Kind code of ref document: A1