Claim settlement data risk prompting method and device, equipment, medium and product
By using logistic regression models to generate risk prediction models during insurance claims, the problems of manual inspection errors and inefficiency are solved, and more efficient and accurate claims data processing and risk prediction are achieved.
Patent Information
- Application Number
- CN202411756192.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-02
- Publication Date
- 2025-03-21
AI Technical Summary
In the prior art, there are problems of manual investigation errors and low efficiency in the insurance claims process, especially in the confirmation of compensation areas of engineering insurance cases, and there is a lack of system-level anti-leaking technology.
A preset logistic regression model is used to generate a risk prediction model based on historical area compensation data, and a risk prediction result is generated by inputting target claims data, and prompting when the result is risky.
It improves the accuracy and efficiency of claims data inspection, effectively predicts the probability of claims risk in engineering insurance coverage areas, thereby reducing claims risk and improving the quality of claims.
Smart Images
Figure CN119671751A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of financial technology, and in particular to a method, device, equipment, medium and product for risk warning of claims data. Background Art
[0002] In the financial field, such as the insurance industry, abnormal claims behavior refers to situations in the insurance claims process that are inconsistent with normal and reasonable claims procedures or regulations, or involve fraud, falsehood and other improper behaviors.
[0003] At present, whether engineering insurance cases will be compensated is based on the policy coverage area. However, there is no systematic anti-leakage technology between the actual accident area and the engineering insurance policy coverage area. Case handling relies on manual inspection by the claims adjuster, which is prone to inspection errors and low inspection efficiency.
[0004] Therefore, the current case handling is done through manual investigation, which is prone to errors and low efficiency. Summary of the invention
[0005] The present application provides a method, device, equipment, medium and product for risk warning of claims data, which are used to solve the problems of manual investigation in current case handling, which may cause errors in investigation and low investigation efficiency.
[0006] The first aspect of the present application provides a claim data risk warning method, comprising:
[0007] A preset logistic regression model is used to generate a corresponding risk prediction model based on a preset data set; the preset data set includes historical regional compensation related data; the historical regional compensation related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information;
[0008] Obtain target claims data for the target area to be processed;
[0009] Inputting the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data;
[0010] If the risk prediction result is that there is a risk, the risk prediction result is prompted.
[0011] Furthermore, in the above method, the step of using a preset logistic regression model to generate a corresponding risk prediction model based on a preset data set includes:
[0012] Inputting the natural disaster information and the historical claims information into a preset logistic regression model to train the preset logistic regression model;
[0013] Use the preset loss function to determine whether the preset logistic regression model meets the convergence condition;
[0014] If it is determined that the preset logistic regression model meets the convergence condition, the preset logistic regression model is determined as the risk prediction model.
[0015] Furthermore, in the above method, the step of obtaining target claims data of the target area to be processed includes:
[0016] Sending a claim data acquisition request to an electronic device related to claim settlement; the claim data acquisition request includes a target area identifier and a claim settlement time;
[0017] Receiving corresponding initial claim data sent by the electronic device related to claim settlement; the initial claim data is obtained by querying the electronic device related to claim settlement from a preset database based on the target area identifier and the claim settlement time;
[0018] The initial claims data is preprocessed to generate target claims data.
[0019] Furthermore, in the method described above, the target claims data includes a target insurance type and a target cumulative claim amount;
[0020] The step of inputting the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data includes:
[0021] Inputting the target insurance type and the target cumulative compensation amount into the risk prediction model, so as to generate the corresponding compensation probability and compensation amount range through the risk prediction model, and determine whether the target cumulative compensation amount falls within the compensation amount range; the compensation amount range corresponds to the target insurance type;
[0022] If it is determined that the target cumulative compensation amount does not fall within the compensation amount range and / or the compensation probability is greater than a preset probability threshold, a risk prediction result with risks is generated; the risk prediction result includes the target cumulative compensation amount and the compensation probability.
[0023] Furthermore, in the above method, the step of providing the risk prediction result comprises:
[0024] The risk prediction results are displayed on a preset screen, and the target cumulative compensation amount and the compensation probability in the risk prediction results are highlighted.
[0025] Furthermore, the method as described above, after prompting the risk prediction result, further comprises:
[0026] Acquire prediction error information from a preset database; the prediction error information is claim data where the risk prediction result indicates that there is no risk but there is actually a risk, or claim data where the risk prediction result indicates that there is a risk but there is actually no risk;
[0027] The risk prediction model is retrained based on the prediction error information to generate an optimized risk prediction model.
[0028] The second aspect of the present application provides a claim data risk warning device, comprising:
[0029] A first generating module is used to generate a corresponding risk prediction model based on a preset data set using a preset logistic regression model; the preset data set includes historical regional compensation related data; the historical regional compensation related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information;
[0030] An acquisition module, used for acquiring target claims data of a target area to be processed;
[0031] A second generating module, used for inputting the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data;
[0032] A prompt module is used to prompt the risk prediction result if the risk prediction result is risky.
[0033] Further, in the above device, the first generating module is specifically used for:
[0034] The natural disaster information and the historical claims information are input into a preset logistic regression model to train the preset logistic regression model; a preset loss function is used to determine whether the preset logistic regression model meets the convergence condition; if it is determined that the preset logistic regression model meets the convergence condition, the preset logistic regression model is determined as a risk prediction model.
[0035] Furthermore, in the above device, the acquisition module is specifically used for:
[0036] Send a claim data acquisition request to an electronic device related to a claim; the claim data acquisition request includes a target area identifier and a claim time; receive corresponding initial claim data sent by the electronic device related to the claim; the initial claim data is obtained by querying a preset database by the electronic device related to the claim based on the target area identifier and the claim time; perform data preprocessing on the initial claim data to generate target claim data.
[0037] Further, in the above-mentioned device, the target claim settlement data includes a target insurance type and a target cumulative claim payment amount;
[0038] The second generating module is specifically used for:
[0039] The target insurance type and the target cumulative compensation amount are input into the risk prediction model to generate the corresponding compensation probability and compensation amount range through the risk prediction model, and determine whether the target cumulative compensation amount falls within the compensation amount range; the compensation amount range corresponds to the target insurance type; if it is determined that the target cumulative compensation amount does not fall within the compensation amount range and / or the compensation probability is greater than a preset probability threshold, a risk prediction result with risk is generated; the risk prediction result includes the target cumulative compensation amount and the compensation probability.
[0040] Furthermore, in the above-mentioned device, the prompt module is specifically used to:
[0041] The risk prediction results are displayed on a preset screen, and the target cumulative compensation amount and the compensation probability in the risk prediction results are highlighted.
[0042] Furthermore, the device as described above further comprises:
[0043] An optimization module is used to obtain prediction error information from a preset database; the prediction error information is claim data where the risk prediction result is that there is no risk but there is actually a risk, or the risk prediction result is that there is a risk but there is actually no risk; based on the prediction error information, the risk prediction model is re-trained to generate an optimized risk prediction model.
[0044] A third aspect of the present application provides an electronic device, including: a memory and a processor;
[0045] The memory stores computer-executable instructions;
[0046] The processor executes the computer-executable instructions stored in the memory to implement the claims data risk warning method as described in any one of the first aspects.
[0047] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the claims data risk warning method described in any one of the first aspects.
[0048] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the claims data risk warning method described in any one of the first aspects.
[0049] The present application provides a method, device, equipment, medium and product for risk warning of claims data, the method comprising: using a preset logistic regression model to generate a corresponding risk prediction model based on a preset data set; the preset data set includes historical regional claims-related data; the historical regional claims-related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information; obtaining the target claims data of the target area to be processed; inputting the target claims data into the risk prediction model to generate a risk prediction result corresponding to the target claims data; if the risk prediction result is risky, prompting the risk prediction result. The risk warning method for claims data of the present application uses a preset logistic regression model to generate a corresponding risk prediction model. When it is necessary to investigate the target claims data, the target claims data can be input into the risk prediction model to generate a risk prediction result corresponding to the target claims data. If the risk prediction result is risky, prompting the risk prediction result, thereby improving the accuracy of the investigation and improving the efficiency of the investigation by means of a risk prediction model. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0051] Figure 1 A scenario diagram for implementing the claim data risk warning method of the embodiment of the present application;
[0052] Figure 2 A flowchart of a method for risk warning of claims data provided in one embodiment of the present application;
[0053] Figure 3 A flowchart of a method for risk warning of claims data provided by another embodiment of the present application;
[0054] Figure 4 A schematic diagram of the structure of the claim data risk warning device provided in this application;
[0055] Figure 5 A schematic diagram of the structure of the electronic device provided in this application.
[0056] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION
[0057] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0058] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of financial data or user data and other information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals. It should be noted that in the embodiments of this application, some existing solutions in the industry such as certain software, components, models, etc. may be mentioned, and they should be considered as exemplary. Their purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.
[0059] It should be noted that the disclosed claims data risk warning method, device, equipment, medium and product can be used in the field of financial technology. It can also be used in any field other than the field of financial technology. The application field of the disclosed claims data risk warning method, device, equipment, medium and product is not limited.
[0060] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.
[0061] In order to clearly understand the technical solution of this application, the invention concept of this application is first introduced in detail. At present, in the financial field, such as the insurance industry, whether engineering insurance cases are compensated needs to be based on the policy subject protection area, but there is no system-level anti-leakage technology between the actual accident area and the engineering insurance policy protection area, and case handling relies on manual inspection at the claims post. Since manual inspection has a certain risk of error (intentional or accidental), this method has the problem of inspection errors and low inspection efficiency.
[0062] Therefore, the current case handling is done through manual investigation, which is prone to errors and low efficiency.
[0063] Therefore, in view of the problems of manual investigation and error in case handling in the prior art and low investigation efficiency, the inventors found in their research that, in order to solve this problem, a corresponding risk prediction model can be generated based on a preset data set through a logistic regression model. Therefore, when processing target claim data, a risk prediction result corresponding to the target claim data can be generated based on the risk prediction model, thereby improving the accuracy of the investigation and improving the investigation efficiency.
[0064] Specifically, a preset logistic regression model is used to generate a corresponding risk prediction model based on a preset data set. The preset data set includes historical regional compensation related data. The historical regional compensation related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information. Obtain the target claims data of the target area to be processed. Input the target claims data into the risk prediction model to generate a risk prediction result corresponding to the target claims data. If the risk prediction result is risky, the risk prediction result is prompted.
[0065] The claim data risk warning method of the present application uses a preset logistic regression model to generate a corresponding risk prediction model. When it is necessary to investigate the target claim data, the target claim data can be input into the risk prediction model to generate a risk prediction result corresponding to the target claim data. If the risk prediction result is risky, the risk prediction result is prompted, thereby improving the accuracy of the investigation and improving the efficiency of the investigation through the risk prediction model. At the same time, the risk prediction model can be used to effectively predict the probability of claim risks occurring in the engineering insurance coverage area, thereby achieving the purpose of reducing compensation and losses and improving the quality of case claims settlement and payment.
[0066] Based on the above creative findings, the inventor proposed the technical solution of the present application.
[0067] The following is an introduction to the application scenarios of the claim data risk warning method provided in the embodiment of the present application. Figure 1 As shown, 1 is a first electronic device, and 2 is a second electronic device. The first electronic device 1 can be a user terminal, an electronic device related to a claims settlement system, etc. The second electronic device 2 can be an Internet platform device, a database device, etc.
[0068] Exemplarily, when the first electronic device 1 performs risk prediction on the claim settlement data, it is specifically as follows:
[0069] ① Use the preset logistic regression model to generate the corresponding risk prediction model based on the preset data set. The preset data set includes historical regional compensation related data. The historical regional compensation related data includes: natural disaster information in the target area and historical claims information corresponding to the natural disaster information.
[0070] ② The second electronic device 2 sends the target claim data to the first electronic device 1. The target claim data corresponds to the target area.
[0071] The first electronic device 1 performs the following process:
[0072] ③ Input the target claims data into the risk prediction model to generate risk prediction results corresponding to the target claims data.
[0073] ④ If the risk prediction result is that there is a risk, the risk prediction result will be prompted.
[0074] Optionally, the method of prompting the risk prediction result may be to send the risk prediction result to the user terminal so that the user knows the risk prediction result, or to display the risk prediction result on the screen and highlight the important content therein.
[0075] The embodiments of the present application are introduced below in conjunction with the drawings in the specification.
[0076] Figure 2 A flowchart of a method for risk warning of claims data provided by an embodiment of the present application is shown in FIG. Figure 2 As shown, the execution subject in this embodiment is a claim data risk warning device, which can be integrated into an electronic device, such as an electronic device related to a claim system. The claim data risk warning method provided in this embodiment includes the following steps:
[0077] Step S101, using a preset logistic regression model to generate a corresponding risk prediction model based on a preset data set. The preset data set includes historical regional compensation related data. The historical regional compensation related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information.
[0078] In this embodiment, the preset logistic regression model is a statistical model and also a classification algorithm. The risk prediction model generated by the preset logistic regression model can determine the risk prediction results corresponding to the target claims data. The risk prediction results include at least whether there is a risk result (i.e., the classification result), and may also include the payment probability corresponding to the target claims data, the target cumulative payment amount, etc.
[0079] The preset data set can be pre-set and used to train and generate the risk prediction model, including historical regional compensation related data. The historical regional compensation related data includes: natural disaster information in the target area and historical claims information corresponding to the natural disaster information. The natural disaster information can further include the type of natural disaster, the time of occurrence, the scope of impact, etc. The historical claims information can further include: the amount of compensation, the cause of the accident, the time of reporting, the time of passing the claim verification, the time of closing, etc.
[0080] Step S102, obtaining target claims data of the target area to be processed.
[0081] In this embodiment, the acquisition method may be to directly acquire from other related electronic devices, or to acquire raw data from other related electronic devices and then generate the data after data preprocessing.
[0082] Step S103, inputting the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data.
[0083] In this embodiment, the target claim data includes the cumulative compensation amount of the target area, the target insurance type, etc. The cumulative compensation amount refers to the cumulative compensation amount of the target area within a preset time period. After the target claim data is input into the risk prediction model, the risk prediction model judges and determines the risk prediction result, that is, determines whether there is a risk.
[0084] Step S104: If the risk prediction result is that there is a risk, the risk prediction result is prompted.
[0085] In this embodiment, the risk prediction result of having risk may refer to the risk of inconsistency between the target claim data and the actual claim data, the risk of a high probability of payment, etc. The way of prompting the risk prediction result may be to send the risk prediction result to the user terminal so that the user knows the risk prediction result, or to display the risk prediction result on the screen and highlight the important content therein.
[0086] The claim data risk warning method of this embodiment uses a preset logistic regression model to generate a corresponding risk prediction model based on a preset data set. The preset data set includes historical regional claims related data. The historical regional claims related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information. Obtain the target claim data of the target area to be processed. Input the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data. If the risk prediction result is risky, the risk prediction result is prompted.
[0087] The claim data risk warning method of the present application uses a preset logistic regression model to generate a corresponding risk prediction model. When it is necessary to investigate the target claim data, the target claim data can be input into the risk prediction model to generate a risk prediction result corresponding to the target claim data. If the risk prediction result is risky, the risk prediction result is prompted, thereby improving the accuracy of the investigation and improving the efficiency of the investigation through the risk prediction model.
[0088] Figure 3 A flowchart of a method for risk warning of claims data provided by another embodiment of the present application is shown in FIG. Figure 3As shown, the claim data risk warning method provided in this embodiment is a further refinement of the claim data risk warning method provided in the previous embodiment of this application. The claim data risk warning method provided in this embodiment includes the following steps.
[0089] Step S201, inputting natural disaster information and historical claims information into a preset logistic regression model to train the preset logistic regression model.
[0090] In this embodiment, the historical claims information includes the historical accumulated amount of claims and the historical claim ratio, wherein the historical claim ratio can reflect the actual probability of claim and can be used to determine whether the probability of claim predicted by the preset logistic regression model training is correct. The historical accumulated amount of claims can reflect the amount of claims in the region to determine the range of claims to be used subsequently. During the training process, the risk prediction results generated by the preset logistic regression model can be annotated to mark whether the risk prediction results are correct, so as to facilitate the adjustment of the preset logistic regression model during the training process.
[0091] Step S202: using a preset loss function to determine whether the preset logistic regression model meets the convergence condition.
[0092] In this embodiment, the preset loss function may use logarithmic loss (Log Loss) to measure the difference between the probability distribution predicted by the model and the actual one through the loss function. If the difference is small, it is determined that the preset logistic regression model meets the convergence condition.
[0093] Step S203: If it is determined that the preset logistic regression model meets the convergence condition, the preset logistic regression model is determined as the risk prediction model.
[0094] Optionally, in this embodiment, after the risk prediction model is determined, the risk prediction model may be further tested:
[0095] Get a test sample. The test sample includes test claims data.
[0096] The test sample is input into the risk prediction model to generate a test risk prediction result corresponding to the test sample.
[0097] Compare the test risk prediction results with the actual test risk results.
[0098] If the test risk prediction result is the same as the actual test risk result or the difference is small, the test result of the risk prediction model is determined to be a passed test.
[0099] In this embodiment, the test claims data may be simulation data or historical claims data. The historical claims data is different from the historical regional claims related data in the aforementioned embodiment and is used to test the accuracy of the risk prediction model.
[0100] Step S204, obtaining target claims data of the target area to be processed.
[0101] Optionally, in this embodiment, S204 may be specifically as follows:
[0102] Send a claim data acquisition request to the electronic device related to the claim. The claim data acquisition request includes the target area identifier and the claim time.
[0103] Receive the corresponding initial claim data sent by the electronic device related to the claim. The initial claim data is obtained by the electronic device related to the claim from a preset database based on the target area identifier and the claim time.
[0104] Perform data preprocessing on the initial claims data to generate target claims data.
[0105] In this embodiment, electronic equipment related to claims can be used as a preset database to implement a data storage function. At the same time, there is a mapping relationship between the target area identifier, the claim time, and the initial claim data in the preset database.
[0106] Step S205, input the target insurance type and the target cumulative compensation amount into the risk prediction model, so as to generate the corresponding compensation probability and compensation amount range through the risk prediction model, and determine whether the target cumulative compensation amount falls within the compensation amount range. The compensation amount range corresponds to the target insurance type.
[0107] In this embodiment, the target claim settlement data includes the target insurance type and the target cumulative claim amount. The claim probability is generated based on the target insurance type, and the claim probability is related to the target insurance type and the target area.
[0108] The target cumulative compensation amount is the current accumulated compensation amount. If the target cumulative compensation amount does not fall within the compensation amount range, for example, it exceeds the maximum value of the compensation amount range, it means there is a risk.
[0109] Step S206: If it is determined that the target cumulative compensation amount does not belong to the compensation amount range and / or the compensation probability is greater than the preset probability threshold, a risk prediction result with risk is generated. The risk prediction result includes the target cumulative compensation amount and the compensation probability.
[0110] In this embodiment, the preset probability threshold may be set according to actual application, and is related to the target area and the target insurance type.
[0111] Step S207: If the risk prediction result is that there is a risk, the risk prediction result is prompted.
[0112] Optionally, in this embodiment, the risk prediction result may be specifically as follows:
[0113] The risk prediction results are displayed on a preset screen, and the target cumulative payout amount and payout probability in the risk prediction results are highlighted.
[0114] In this embodiment, the preset screen may be a screen of other display devices or a screen of related electronic devices of the claims system. The highlight display may be performed by highlighting with a highlight color or by other marking methods, such as underlining, borders, etc.
[0115] Optionally, in this embodiment, after prompting the risk prediction result, the following is further included:
[0116] The prediction error information is obtained from the preset database. The prediction error information is the claim data where the risk prediction result is that there is no risk but there is actually a risk, or the claim data where the risk prediction result is that there is a risk but there is actually no risk.
[0117] The risk prediction model is retrained based on the prediction error information to generate an optimized risk prediction model.
[0118] In this embodiment, by storing the prediction error information in a preset database, the prediction error information can be obtained from the preset database to retrain the risk prediction model and generate an optimized risk prediction model.
[0119] In order to further illustrate the claim data risk warning method of this embodiment in detail, it will be further explained in combination with specific applications below.
[0120] The claim data risk warning method of this embodiment may include the following steps:
[0121] Step 1: Conduct data exploration on historical production claims data, and calculate the probability of claims occurring in the target area under specific events (such as floods, fires, earthquakes, etc.) based on actual claims data in the past 10 years.
[0122] Step 2: Based on step 1, build a risk prediction model for the target area
[0123] The relationship between these events and the amount of compensation is analyzed using a logistic regression model, thereby predicting the possible amount of compensation or the probability of compensation if a given event occurs. The output is a probability value that represents the risk of compensation occurring under certain conditions.
[0124] Build the model
[0125] [\text{logit}(p) = \log\left(\frac{p}{1-p}\right) = \beta_0 + \beta_1x_1 + \beta_2x_2 + \cdots + \beta_nx_n]
[0126] Where p is the probability that the dependent variable is 1, logit(p) is the log odds, beta_0 is the intercept term, beta_1, beta_2,…, beta_n are the coefficients of the independent variables, and x_1, x_2,…, x_n are the independent variables.
[0127] Step 3: Based on step 2, guide the processing of claims cases based on the results of the model output and provide risk warning prompts on the page.
[0128] (1) Compensation amount range: Check whether the cumulative compensation amount of the target insurance type is within the compensation amount range output by the model: If the compensation amount does not fall within the compensation amount range, a page prompt will be displayed, and the adjuster can further carefully evaluate the compensation amount. At the same time, adjust the compensation amount rule configuration based on the compensation amount range to improve the upgrade review mechanism.
[0129] (2) Probability of compensation: A large screen will be developed based on the output of the model to dynamically display the probability of compensation occurring in each region at each time point. If the probability of compensation occurring in a certain region in the XX season is high and the amount is large, the large screen data can be used to monitor and intervene in case handling in real time.
[0130] Step 4: Based on the model evaluation results and business needs, further optimize and iterate the model to improve the performance and accuracy of the model.
[0131] The claims data risk warning method of this embodiment is based on the problem that current case handling generally relies on manual investigation by claims personnel. In order to prevent behavioral risks, a regional risk prediction model for engineering insurance has been developed. By exploring the historical production claims data, the probability of claims occurring in the region under specific events (such as floods, fires, earthquakes, etc.) is calculated based on the actual claims data in the past 10 years. The relationship between these events and the amount of compensation is analyzed using a logistic regression model to predict the possible amount of compensation or the probability of compensation when a given event occurs. The output is a probability value, which indicates the risk of compensation occurring under specific conditions. The results based on the model output guide the processing of claims cases and provide page risk warning prompts.
[0132] At the same time, the regional risk prediction model can combine historical compensation data to generate the probability of compensation and the compensation amount range of the case, and detect leakage risks in the case handling process through system technology. For the cumulative compensation amount of the case, a new compensation amount range check is added; for areas with high compensation rates, large-screen data monitoring is used to intervene in the case handling in a timely manner. This is of great significance for guiding the claim settlement and payment operations.
[0133] In the claims data risk warning method of this embodiment, the risk prediction model can combine historical claims data to generate the probability of claims and the range of claims, and detect leakage risks during case handling through system technology. A new range of claims is added for the cumulative claims amount of the case. For areas with high claims rates, large-screen data monitoring is used to promptly intervene in claims processing.
[0134] Figure 4 The structural diagram of the claim data risk warning device provided in this application is as follows: Figure 4 As shown, in this embodiment, the claim data risk prompting device 300 includes:
[0135] The first generation module 301 is used to generate a corresponding risk prediction model based on a preset data set using a preset logistic regression model. The preset data set includes historical regional compensation related data. The historical regional compensation related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information.
[0136] The acquisition module 302 is used to acquire target claims data of the target area to be processed.
[0137] The second generating module 303 is used to input the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data.
[0138] The prompt module 304 is used to prompt the risk prediction result if the risk prediction result is risky.
[0139] The claim data risk warning device provided in this embodiment can execute Figure 2 The technical solution of the method embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The method embodiments shown are similar and will not be described in detail here.
[0140] The claim data risk prompting device provided in this application is based on the claim data risk prompting device provided in the previous embodiment, and further refines the claim data risk prompting device. The claim data risk prompting device 300 includes:
[0141] Optionally, in this embodiment, the first generating module 301 is specifically used for:
[0142] The natural disaster information and historical claims information are input into the preset logistic regression model to train the preset logistic regression model. The preset loss function is used to determine whether the preset logistic regression model meets the convergence condition. If it is determined that the preset logistic regression model meets the convergence condition, the preset logistic regression model is determined as the risk prediction model.
[0143] Optionally, in this embodiment, the acquisition module 302 is specifically used for:
[0144] Send a claim data acquisition request to the electronic device related to the claim. The claim data acquisition request includes the target area identifier and the claim time. Receive the corresponding initial claim data sent by the electronic device related to the claim. The initial claim data is obtained by the electronic device related to the claim from a preset database based on the target area identifier and the claim time. Perform data preprocessing on the initial claim data to generate target claim data.
[0145] Optionally, in this embodiment, the target claims data includes a target insurance type and a target cumulative compensation amount.
[0146] The second generating module 303 is specifically used for:
[0147] The target insurance type and the target cumulative compensation amount are input into the risk prediction model to generate the corresponding compensation probability and compensation amount range through the risk prediction model, and determine whether the target cumulative compensation amount falls within the compensation amount range. The compensation amount range corresponds to the target insurance type. If it is determined that the target cumulative compensation amount does not fall within the compensation amount range and / or the compensation probability is greater than the preset probability threshold, a risk prediction result with risk is generated. The risk prediction result includes the target cumulative compensation amount and the compensation probability.
[0148] Optionally, in this embodiment, the prompt module 304 is specifically used to:
[0149] The risk prediction results are displayed on a preset screen, and the target cumulative payout amount and payout probability in the risk prediction results are highlighted.
[0150] Optionally, in this embodiment, the claim data risk prompting device 300 further includes:
[0151] The optimization module is used to obtain prediction error information from a preset database. The prediction error information is the claim data where the risk prediction result is that there is no risk but there is actually a risk, or the claim data where the risk prediction result is that there is a risk but there is actually no risk. The risk prediction model is retrained based on the prediction error information to generate an optimized risk prediction model.
[0152] The claim data risk warning device provided in this embodiment can execute Figure 2-Figure 3The technical solution of the method embodiment shown in the figure has the same implementation principle and technical effect as Figure 2-Figure 3 The method embodiments shown are similar and will not be described in detail here.
[0153] According to an embodiment of the present application, the present application also provides an electronic device, a computer-readable storage medium and a computer program product.
[0154] like Figure 5 As shown, Figure 5 is a schematic diagram of the structure of the electronic device provided by the present application. The electronic device is intended for various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.
[0155] like Figure 5 As shown, the electronic device includes: a processor 401 and a memory 402. The various components are connected to each other using different buses and can be installed on a common mainboard or in other ways as needed. The processor can process instructions executed in the electronic device. The electronic device can also include a transceiver, which is used to send and receive data.
[0156] The memory 402 is a non-transitory computer-readable storage medium provided in the present application. The memory stores instructions executable by at least one processor to enable at least one processor to execute the claim data risk warning method provided in the present application. The non-transitory computer-readable storage medium of the present application stores computer instructions, which are used to enable a computer to execute the claim data risk warning method provided in the present application.
[0157] The memory 402 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the claim data risk warning method in the embodiment of the present application (for example, the attached Figure 4 The processor 401 executes various functional applications and data processing of the electronic device by running the non-transient software programs, instructions and modules stored in the memory 402, that is, the claim data risk prompt method in the above method embodiment is implemented.
[0158] This embodiment also provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute the claims data risk warning method of the above embodiment.
[0159] At the same time, this embodiment also provides a computer product. When the instructions in the computer product are executed by the processor of an electronic device, the electronic device can execute the claims data risk warning method of the above embodiment.
[0160] Those skilled in the art will readily come up with other implementations of the embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modifications, uses or adaptations of the embodiments of the present application, which follow the general principles of the embodiments of the present application and include common knowledge or customary technical means in the art that are not disclosed in the embodiments of the present application.
[0161] It should be understood that the embodiments of the present application are not limited to the precise structures described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the embodiments of the present application is limited only by the appended claims.
Claims
1. A method for risk warning of claims data, characterized in that: include: A preset logistic regression model is used to generate a corresponding risk prediction model based on a preset data set; The preset data set includes historical regional claims-related data; The historical regional compensation related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information; Obtain target claims data for the target area to be processed; Inputting the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data; If the risk prediction result is that there is a risk, the risk prediction result is prompted.
2. The method according to claim 1, characterized in that The method of using a preset logistic regression model to generate a corresponding risk prediction model based on a preset data set includes: Inputting the natural disaster information and the historical claims information into a preset logistic regression model to train the preset logistic regression model; Use the preset loss function to determine whether the preset logistic regression model meets the convergence condition; If it is determined that the preset logistic regression model meets the convergence condition, the preset logistic regression model is determined as the risk prediction model.
3. The method according to claim 1, characterized in that The step of obtaining target claims data of the target area to be processed includes: Sending a claim data acquisition request to an electronic device related to claim settlement; the claim data acquisition request includes a target area identifier and a claim settlement time; Receiving corresponding initial claim data sent by the electronic device related to claim settlement; the initial claim data is obtained by querying the electronic device related to claim settlement from a preset database based on the target area identifier and the claim settlement time; The initial claims data is preprocessed to generate target claims data.
4. The method according to claim 1, characterized in that The target claims data includes target insurance type and target cumulative compensation amount; The step of inputting the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data includes: Inputting the target insurance type and the target cumulative compensation amount into the risk prediction model, so as to generate the corresponding compensation probability and compensation amount range through the risk prediction model, and determine whether the target cumulative compensation amount falls within the compensation amount range; the compensation amount range corresponds to the target insurance type; If it is determined that the target cumulative compensation amount does not fall within the compensation amount range and / or the compensation probability is greater than a preset probability threshold, a risk prediction result with risks is generated; the risk prediction result includes the target cumulative compensation amount and the compensation probability.
5. The method according to claim 4, characterized in that The risk prediction result is indicated, including: The risk prediction results are displayed on a preset screen, and the target cumulative compensation amount and the compensation probability in the risk prediction results are highlighted.
6. The method according to claim 1, characterized in that After the risk prediction result is prompted, the method further includes: Acquire prediction error information from a preset database; the prediction error information is claim data where the risk prediction result indicates that there is no risk but there is actually a risk, or claim data where the risk prediction result indicates that there is a risk but there is actually no risk; The risk prediction model is retrained based on the prediction error information to generate an optimized risk prediction model.
7. A claim data risk warning device, characterized in that: include: A first generating module is used to generate a corresponding risk prediction model based on a preset data set using a preset logistic regression model; The preset data set includes historical regional claims-related data; The historical regional compensation related data includes: natural disaster information of the target area and historical claims information corresponding to the natural disaster information; An acquisition module, used for acquiring target claims data of a target area to be processed; A second generating module, used for inputting the target claim data into the risk prediction model to generate a risk prediction result corresponding to the target claim data; A prompt module is used to prompt the risk prediction result if the risk prediction result is risky.
8. An electronic device, characterized in that: include: Memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the claim data risk warning method as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the claim data risk warning method as described in any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for risk warning of claims data as described in any one of claims 1 to 6 is implemented.
Citation Information
Cited By
Financial insurance claim settlement risk management method and system based on multi-source space-time rainfall data
CN121724774A