Commercial vehicle insurance anti-fraud method and system based on big data risk control model

Abstract

The invention relates to a commercial vehicle insurance anti-fraud method and system based on a big data risk control model. The method comprises the following steps: synchronizing accident data in fraud risk detection with accident data in a collision accident; the synchronized data are imported through a fraud list page for data verification, and verified complete information is imported for processing; matching the vehicle frame number and the license plate number in a memory after importing, fusing the successfully matched data, and carrying out a new accident addition processing flow on the unsuccessfully matched data; generating a fraud report, and automatically carrying out fraud factor calculation on the imported data; obtaining operator factor configuration, querying factor levels under an accident according to calculation results of the factors, obtaining operator fraud configuration, and calculating an overall fraud level according to the levels of the factors and the operator fraud configuration; the whole process does not need human intervention, the data security is high, the flow design is reasonable, the automation degree is high, and the accident fraud level can be fed back efficiently and accurately.

Description

Technical Field

[0001] The present invention relates to the field of commercial vehicle insurance anti-fraud technology, and more specifically, to a commercial vehicle insurance anti-fraud method and system based on a big data risk control model. Background Art

[0002] Automobile insurance fraud is a form of insurance fraud. In minor cases, it is a commercial fraud lawsuit. In serious cases, criminal lawsuits can be filed according to relevant laws, and the relevant criminal liability will be borne in accordance with the provisions of relevant laws. At present, the identification method of insurance fraud is mostly to customize the scoring rules and manually input the scoring factors into the engine for automatic scoring. The operation process requires too much manual intervention and has a high security risk. At the same time, the degree of automation is low and the response speed is slow. There is a need for a highly automated commercial vehicle insurance anti-fraud method and system based on a big data risk control model. Summary of the invention

[0003] The technical problem to be solved by the present invention is that, in view of the above-mentioned defects of the prior art, a commercial vehicle insurance anti-fraud method based on a big data risk control model is provided, and a commercial vehicle insurance anti-fraud system is also provided.

[0004] The technical solution adopted by the present invention to solve its technical problem is:

[0005] Construct a commercial vehicle insurance anti-fraud method based on big data risk control model, which includes the following methods:

[0006] Synchronize incident data from fraud risk detection with incident data from collision incidents;

[0007] Import the synchronized data through the fraud list page to perform data verification. If the verification is successful, the successfully verified data will be returned. If the verification fails, an error message will be fed back.

[0008] Import the complete information after verification into the processing;

[0009] After importing, the frame number and license plate number are matched in the memory, and the successfully matched data are merged. For the unmatched data, a new accident adding process is carried out;

[0010] Generate fraud reports and automatically calculate fraud factors for imported processed data;

[0011] Obtain the operator's factor configuration, query the factor level under the accident based on the calculation results of each factor, obtain the operator's fraud configuration, and calculate the overall fraud level based on the level of each factor and the operator's fraud configuration;

[0012] The fraud level is written back into the accident information table. If the conditions for reaching any fraud level are not met, the fraud level is empty and the front-end display page does not show the result.

[0013] The commercial vehicle insurance anti-fraud method based on the big data risk control model of the present invention, wherein the data verification includes the following method:

[0014] The incident processor receives the fraud list page from the Web end and starts data verification based on the fraud list page;

[0015] Obtain vehicle information from the vehicle end;

[0016] Obtain insurance company information from the management end;

[0017] Query the longitude and latitude information from the server by address;

[0018] Verify the vehicle frame number;

[0019] Verification time format;

[0020] Verify the driver;

[0021] Verify the reporter;

[0022] Return verification information to the Web client.

[0023] The commercial vehicle insurance anti-fraud method based on the big data risk control model of the present invention adopts the following method when matching the imported data:

[0024] The incident processor receives the complete information imported from the Web terminal;

[0025] The accident processor uses the VIN to search for the accident in the memory and matches the accident with the VIN;

[0026] The accident processor uses the vehicle license plate to search for the accident in the memory and matches the accident with the vehicle license plate;

[0027] The remaining data components of the accident processor create new accident information;

[0028] Determine whether the information has been modified or not, and if there is any modification, calculate the system fraud factor;

[0029] Return success information to the Web client.

[0030] The commercial vehicle insurance anti-fraud method based on the big data risk control model of the present invention, wherein the fraud report is generated by:

[0031] The web client sends a report request;

[0032] The accident handler obtains multi-language information and multi-language templates from the nacos module;

[0033] The accident processor obtains the accident details from the storage;

[0034] The accident processor obtains vehicle brand information and model information from the vehicle end;

[0035] The accident handler obtains tenant log image information from the management end;

[0036] The accident processor queries the memory for all enabled factor sets of the accident;

[0037] Assembly information;

[0038] The accident processor queries all factor attachment information sets from the storage;

[0039] Generate URL QR code and assemble template entity;

[0040] Render and generate word and save it in a temporary folder;

[0041] Read the word document, convert it to a pdf document, and save it in a temporary folder;

[0042] Upload to the cos bucket and return a pdf link;

[0043] Sign the PDF link with a validity period and send it to the Web client.

[0044] The commercial vehicle insurance anti-fraud method based on the big data risk control model of the present invention, wherein the fraud factor calculation method adopts:

[0045] After an accident is created or modified, an asynchronous thread is used to execute one or more of the accident creation process, accident modification process, and factor modification process;

[0046] The calculation factor type is selected according to the set conditions, and different factors are selected to calculate the factor level according to different factors;

[0047] Store factor level data.

[0048] The commercial vehicle insurance anti-fraud method based on the big data risk control model described in the present invention, wherein the timing for calculating the fraud factor is after the task workbench creates an accident, after the page creates the accident, after the fraud list page is imported, after the accident is modified, and after the page factor is modified. One or more of the times.

[0049] The commercial vehicle insurance anti-fraud method based on the big data risk control model described in the present invention, wherein the fraud factor calculation performed after the accident is created on the task workbench is divided into system calculation factors and other calculation factors.

[0050] The commercial vehicle insurance anti-fraud method based on the big data risk control model of the present invention, wherein the factor configuration adopts the method:

[0051] The accident handling end obtains the factor configuration information from the storage, determines whether the factor configuration information exists, and proceeds to the next step if it does. Otherwise, obtains the factor initialization configuration from the storage, copies the factor initialization configuration and persists it, and feeds it back to the Web end.

[0052] The accident processing end receives the modified factor configuration from the Web end and stores it in the memory.

[0053] The commercial vehicle insurance anti-fraud method based on the big data risk control model of the present invention, wherein the fraud configuration adopts the method:

[0054] The accident handling end obtains the fraud configuration information from the storage, determines whether the fraud configuration information exists, and proceeds to the next step if it does. Otherwise, obtains the fraud initialization configuration from the storage, copies the fraud initialization configuration and persists it, and feeds it back to the Web end.

[0055] The accident handling end receives the modified fraud configuration from the Web end and stores it in the storage.

[0056] A commercial vehicle insurance anti-fraud system based on a big data risk control model is used to implement the commercial vehicle insurance anti-fraud method based on a big data risk control model as described above, wherein the system includes a Web end, an accident processor, a tenant end, a management end, a service end, and a storage device;

[0057] The accident data in the fraud risk detection is synchronized with the accident data in the collision accident and stored in the memory;

[0058] The web terminal is used to import the fraud list page into the accident processor;

[0059] The accident processor is used to obtain vehicle information from the tenant side, insurance company information from the management side, and longitude and latitude information from the server side, and perform data verification. If the verification is successful, the successfully verified data is returned to the Web side, and if the verification fails, an error message is fed back to the Web side;

[0060] The web terminal is also used to import the complete information after verification into the accident processor for processing;

[0061] The accident processor is also used to match the vehicle frame number and the license plate number in the memory after the imported complete information, merge the successfully matched data, and perform a new accident adding process for the unmatched data;

[0062] The accident processor is also used to generate a fraud report and automatically calculate the fraud factor of the imported processed data; obtain the operator factor configuration, query the factor level under the accident for the calculation results of each factor, obtain the operator fraud configuration, and calculate the overall fraud level based on the level of each factor and the operator fraud configuration; the fraud level is written back into the accident information table. If the conditions for reaching any fraud level are not met, the fraud level is empty and the front-end display page does not display the result.

[0063] The beneficial effects of the present invention are: by applying the method of the present application, the accident data is directly synchronized to the storage device, and after the accident data is verified, matched, and automatically merged according to the set process, and the processing process steps for the new accident are added, the operator's factor configuration and fraud configuration are automatically pulled to calculate the fraud level. The entire process does not require human intervention, the data security is high, and the process design is reasonable and the degree of automation can efficiently and accurately feedback the accident fraud level. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. The drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work:

[0065] Figure 1 It is a flow chart of a commercial vehicle insurance anti-fraud method based on a big data risk control model according to a preferred embodiment of the present invention;

[0066] Figure 2 It is a fraud introduction flow chart of a commercial vehicle insurance anti-fraud method based on a big data risk control model in a preferred embodiment of the present invention;

[0067] Figure 3 It is a flow chart of fraud report generation of a commercial vehicle insurance anti-fraud method based on a big data risk control model in a preferred embodiment of the present invention;

[0068] Figure 4 This is a time sequence diagram of fraud factor calculation of a commercial vehicle insurance anti-fraud method based on a big data risk control model in a preferred embodiment of the present invention;

[0069] Figure 5 This is a flow chart of fraud factor calculation of a commercial vehicle insurance anti-fraud method based on a big data risk control model in a preferred embodiment of the present invention;

[0070] Figure 6 This is a flow chart of accident fraud level calculation of a commercial vehicle insurance anti-fraud method based on a big data risk control model in a preferred embodiment of the present invention;

[0071] Figure 7It is a factor configuration flow chart of a commercial vehicle insurance anti-fraud method based on a big data risk control model in a preferred embodiment of the present invention;

[0072] Figure 8 It is a fraud configuration flow chart of a commercial vehicle insurance anti-fraud method based on a big data risk control model in a preferred embodiment of the present invention;

[0073] Fig. 9 It is a principle block diagram of a commercial vehicle insurance anti-fraud system based on a big data risk control model in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the following will be described clearly and completely in combination with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are partial embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0075] The commercial vehicle insurance anti-fraud method and system based on the big data risk control model of the preferred embodiment of the present invention is as follows: Figure 1 See also Figure 2-Figure 8 , including the following methods:

[0076] S01: Synchronize the accident data in fraud risk detection with the accident data in collision accidents;

[0077] S02: Data verification is performed on the synchronized data imported through the fraud list page. If the verification is successful, the successfully verified data is returned; if the verification fails, an error message is fed back;

[0078] S03: Importing the complete information after verification into processing;

[0079] S04: after importing, the frame number and the license plate number are matched in the memory, the successfully matched data are merged, and the unmatched data are processed by the new accident adding process;

[0080] S05: Generate a fraud report and automatically calculate the fraud factor for the imported and processed data;

[0081] S06: Obtain the operator factor configuration, query the factor level under the accident for the calculation results of each factor, obtain the operator fraud configuration, and calculate the overall fraud level according to the level of each factor and the operator fraud configuration;

[0082] S07: The fraud level is written back into the accident information table. If the conditions for reaching any fraud level are not met, the fraud level is empty and the front-end display page does not display the result.

[0083] By applying the method of the present application, accident data is directly synchronized to the storage device. After the accident data is verified, matched, and automatically merged according to the set process, and the processing steps for new accidents are added, the operator's factor configuration and fraud configuration are automatically pulled to calculate the fraud level. The entire process does not require human intervention, has high data security, and has a reasonable process design and a high degree of automation, which can efficiently and accurately feedback the accident fraud level.

[0084] Specific:

[0085] like Figure 2 As shown, the data in Excel is imported through the fraud list page. First, the data is verified, then the successfully verified data is given, and finally the import process is performed. In the subsequent import process, the frame number and license plate number will be matched. The matching data will be merged. If there is no matching data at all, the new accident addition process will be carried out. Finally, the system factor calculation level process (see Figure 4 The calculation of the accident fraud level is a subsequent process of factor calculation. The calculation results of each factor are the basis for calculating the accident fraud level. Each operator will calculate the final fraud level of an accident based on its own unique fraud level configuration. If the conditions are not met, the fraud level will be empty and the front-end display page will not be displayed).

[0086] Data verification includes methods:

[0087] The incident processor receives the fraud list page from the Web end and starts data verification based on the fraud list page;

[0088] Obtain vehicle information from the vehicle end;

[0089] Obtain insurance company information from the management end;

[0090] Query the longitude and latitude information from the server by address;

[0091] Verify the vehicle frame number;

[0092] Verification time format;

[0093] Verify the driver;

[0094] Verify the reporter;

[0095] Return verification information to the Web client.

[0096] When matching imported data, use:

[0097] The incident processor receives the complete information imported from the Web terminal;

[0098] The accident processor uses the VIN to search for the accident in the memory and matches the accident with the VIN;

[0099] The accident processor uses the vehicle license plate to search for the accident in the memory and matches the accident with the vehicle license plate;

[0100] The remaining data components of the accident processor create new accident information;

[0101] Determine whether the information has been modified or not, and if there is any modification, calculate the system fraud factor;

[0102] Return success information to the Web client.

[0103] Possible abnormalities include:

[0104] 1. The license plate number and frame number verification failed, and the corresponding vehicle information could not be found;

[0105] 2. Format verification of insurance start time, accident time, reporting time, etc.;

[0106] 3. Accident address error exception. This exception is mainly due to the address reverse query latitude and longitude, calling the map service component interface, which may not get a response (cross-domain request, service may not be stable), and may not get the longitude and latitude. The most important key information for accident preservation is the accident address, so if it is not here, it will not be saved.

[0107] like Figure 3 As shown, the generation of fraud reports is basically the same as that of accident reports. The accident details of the specified accident are queried, and then the data is filled according to the entity (map) specified in the accident template. After filling, the poi-tl component interface will be called to render the word document and save it in a temporary folder. Then it will be read through the stream, and Aspose (paid, cracked version) will be called to convert the word document into a pdf file, which will be uploaded to COS, a link will be generated, and returned to the web end. The specific method is:

[0108] The web client sends a report request;

[0109] The accident handler obtains multi-language information and multi-language templates from the nacos module;

[0110] The accident processor obtains the accident details from the storage;

[0111] The accident processor obtains vehicle brand information and model information from the vehicle end;

[0112] The accident handler obtains tenant log image information from the management end;

[0113] The accident processor queries the memory for all enabled factor sets of the accident;

[0114] Assembly information;

[0115] The accident processor queries all factor attachment information sets from the storage;

[0116] Generate URL QR code and assemble template entity;

[0117] Render and generate word and save it in a temporary folder;

[0118] Read the word document, convert it to a pdf document, and save it in a temporary folder;

[0119] Upload to the cos bucket and return a pdf link;

[0120] Sign the PDF link with a validity period and send it to the Web client;

[0121] Possible abnormalities include:

[0122] 1. Obtain tenant logo image information, which is a long process. During the acquisition process of mta-admin, it will go through proxy services and access the middle platform interface. During this process, there will be various errors such as RPC call errors. The template has done default processing;

[0123] 2. The acquisition of vehicle brand information and model information may also go wrong. Exception capture is also done here. If an error occurs, the data in the template will be displayed as empty. However, the probability of an exception is very small;

[0124] 3. During the factor attachment processing, it is very likely that due to uncontrollable factors, the image cannot be matched, or one factor has multiple attachment information, etc. However, this has been handled in the code.

[0125] Fraud Factor Calculation:

[0126] There are several possibilities for factor calculation.

[0127] 1. After the task workbench creates an incident, it calls the asynchronous process to handle the system factor calculation.

[0128] 2. After the page creation accident is completed.

[0129] 3. After fraudulent excel import.

[0130] 4. After modifying the accident (accident time modification).

[0131] 5. After the page factor is modified;

[0132] like Figure 5 and Figure 6 As shown in the figure, a factor is used as an example to describe the entire calculation process. In the actual implementation process, each factor is different, and of course, there are different logics and RPC call relationships;

[0133] like Figure 7 and Figure 8 As shown, each operator can configure its own knowledge base (factor level configuration and restriction conditions, etc.). When entering the knowledge base configuration page for the first time, the backend will actively initialize a set of configurations for the operator. Of course, this configuration is copied from the pre-made configuration. The operator administrator can change the level configuration that suits his fleet or region on the configuration page. After completion, the accident data information within three years will be recalculated for the system factor, and then the fraud level of the entire accident will be calculated; after each accident creation (task workbench, web page, excel import) is completed, the system factor calculation will be performed. After the factor level of the accident is calculated, the entire accident fraud level will be calculated. At this time, the operator's configuration needs to be obtained. Similarly, for this configuration, the operator defaults to the global initialization configuration, and here you can achieve personalized configuration by changing the page. When the personalized configuration is changed, it is necessary to recalculate the fraud level of all accidents that meet the conditions under the operator's account according to the new configuration.

[0134] Factor configuration method:

[0135] The accident handling end obtains the factor configuration information from the storage, determines whether the factor configuration information exists, and proceeds to the next step if it does. Otherwise, obtains the factor initialization configuration from the storage, copies the factor initialization configuration and persists it, and feeds it back to the Web end.

[0136] The accident processing end receives the modified factor configuration from the Web end and stores it in the memory.

[0137] Fraudulent configuration uses:

[0138] The accident handling end obtains the fraud configuration information from the storage, determines whether the fraud configuration information exists, and proceeds to the next step if it does. Otherwise, obtains the fraud initialization configuration from the storage, copies the fraud initialization configuration and persists it, and feeds it back to the Web end.

[0139] The accident handling end receives the modified fraud configuration from the Web end and stores it in the storage.

[0140] A commercial vehicle insurance anti-fraud system based on a big data risk control model is used to implement the commercial vehicle insurance anti-fraud method based on a big data risk control model as described above, such as Fig. 9 As shown, the system includes a web terminal 100 (web), an accident processor 101 (mta-accident), a tenant terminal 102 (mta-truck), a management terminal 103 (mta-admin), a service terminal 104 (map-service) and a storage 105 (DB);

[0141] The accident data in the fraud risk detection is synchronized with the accident data in the collision accident and stored in the memory 105;

[0142] Web terminal 100, used to import the fraud list page into the incident processor;

[0143] The accident processor 101 is used to obtain vehicle information from the tenant terminal 102, insurance company information from the management terminal 103, and longitude and latitude information from the server terminal 104, and perform data verification. If the verification is successful, the successfully verified data is returned to the Web terminal 100, and if the verification fails, an error message is fed back to the Web terminal;

[0144] The web terminal 100 is also used to import the complete information after verification into the accident processor 101 for processing;

[0145] The accident processor 101 is also used to match the vehicle frame number and the license plate number in the memory after the complete information is imported, and the successfully matched data is merged, and the unmatched data is processed to add a new accident processing flow;

[0146] The accident processor 101 is also used to generate a fraud report and automatically calculate the fraud factor of the imported processed data; obtain the operator factor configuration, query the factor level under the accident for the calculation results of each factor, obtain the operator fraud configuration, and calculate the overall fraud level according to the level of each factor and the operator fraud configuration; the fraud level is written back into the accident information table. If the conditions for reaching any fraud level are not met, the fraud level is empty and the front-end display page does not display the result;

[0147] By using the system of the present application, accident data is directly synchronized to the storage device. After the accident data is verified, matched, and automatically merged according to the set process, and the processing steps for new accidents are added, the operator's factor configuration and fraud configuration are automatically pulled to calculate the fraud level. The entire process does not require human intervention, has high data security, and has a reasonable process design and a high degree of automation, which can efficiently and accurately feedback the accident fraud level.

[0148] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A commercial vehicle insurance anti-fraud method based on a big data risk control model, characterized in that: The following methods are included: Synchronize incident data from fraud risk detection with incident data from collision incidents; Import the synchronized data through the fraud list page to perform data verification. If the verification is successful, the successfully verified data will be returned. If the verification fails, an error message will be fed back. Import the complete information after verification into the processing; After importing, the frame number and license plate number are matched in the memory, and the successfully matched data are merged. For the unmatched data, a new accident adding process is carried out; Generate fraud reports and automatically calculate fraud factors for imported processed data; Obtain the operator's factor configuration, query the factor level under the accident based on the calculation results of each factor, obtain the operator's fraud configuration, and calculate the overall fraud level based on the level of each factor and the operator's fraud configuration; The fraud level is written back into the accident information table. If the conditions for reaching any fraud level are not met, the fraud level is empty and the front-end display page does not show the result.

2. The commercial vehicle insurance anti-fraud method based on big data risk control model according to claim 1 is characterized in that: Data verification includes methods: The incident processor receives the fraud list page from the Web end and starts data verification based on the fraud list page; Obtain vehicle information from the vehicle end; Obtain insurance company information from the management end; Query the longitude and latitude information from the server by address; Verify the vehicle frame number; Verification time format; Verify the driver; Verify the reporter; Return verification information to the Web client.

3. The commercial vehicle insurance anti-fraud method based on the big data risk control model according to claim 1 or 2 is characterized in that: When matching imported data, use: The incident processor receives the complete information imported from the Web terminal; The accident processor uses the VIN to search for the accident in the memory and matches the accident with the VIN; The accident processor uses the vehicle license plate to search for the accident in the memory and matches the accident with the vehicle license plate; The remaining data components of the accident processor create new accident information; Determine whether the information has been modified or not, and if there is any modification, calculate the system fraud factor; Return success information to the Web client.

4. The commercial vehicle insurance anti-fraud method based on big data risk control model according to claim 1 is characterized in that: The fraud report is generated using the following method: The web client sends a report request; The accident handler obtains multi-language information and multi-language templates from the nacos module; The accident processor obtains the accident details from the storage; The accident processor obtains vehicle brand information and model information from the vehicle end; The accident handler obtains tenant log image information from the management end; The accident processor queries the memory for all enabled factor sets of the accident; Assembly information; The accident processor queries all factor attachment information sets from the storage; Generate URL QR code and assemble template entity; Render and generate word and save it in a temporary folder; Read the word document, convert it to a pdf document, and save it in a temporary folder; Upload to the cos bucket and return a pdf link; Sign the PDF link with a validity period and send it to the Web client.

5. The commercial vehicle insurance anti-fraud method based on big data risk control model according to claim 1 is characterized in that: The fraud factor is calculated using the following method: After an accident is created or modified, an asynchronous thread is used to execute one or more of the accident creation process, accident modification process, and factor modification process; The calculation factor type is selected according to the set conditions, and different factors are selected to calculate the factor level according to different factors; Store factor level data.

6. The commercial vehicle insurance anti-fraud method based on big data risk control model according to claim 5 is characterized in that: The fraud factor calculation is performed at one or more of the following times: after an incident is created on the task workbench, after an incident is created on the page, after the fraud list page is imported, after an incident is modified, or after the page factor is modified.

7. The commercial vehicle insurance anti-fraud method based on big data risk control model according to claim 5 is characterized in that: After an incident is created in the task workbench, the fraud factor calculation is divided into system calculation factors and other calculation factors.

8. The commercial vehicle insurance anti-fraud method based on big data risk control model according to claim 1 is characterized in that: The factor configuration adopts the method: The accident handling end obtains the factor configuration information from the storage, determines whether the factor configuration information exists, and proceeds to the next step if it does. Otherwise, obtains the factor initialization configuration from the storage, copies the factor initialization configuration and persists it, and feeds it back to the Web end. The accident processing end receives the modified factor configuration from the Web end and stores it in the memory.

9. The commercial vehicle insurance anti-fraud method based on big data risk control model according to claim 1 is characterized in that: The fraudulent configuration adopts the method: The accident handling end obtains the fraud configuration information from the storage, determines whether the fraud configuration information exists, and proceeds to the next step if it does. Otherwise, obtains the fraud initialization configuration from the storage, copies the fraud initialization configuration and persists it, and feeds it back to the Web end. The accident handling end receives the modified fraud configuration from the Web end and stores it in the storage.

10. A commercial vehicle insurance anti-fraud system based on a big data risk control model, used to implement the commercial vehicle insurance anti-fraud method based on a big data risk control model as described in any one of claims 1 to 9, characterized in that: The system includes a Web terminal, an accident processor, a tenant terminal, a management terminal, a server terminal and a storage device; The accident data in the fraud risk detection is synchronized with the accident data in the collision accident and stored in the memory; The web terminal is used to import the fraud list page into the accident processor; The accident processor is used to obtain vehicle information from the tenant side, insurance company information from the management side, and longitude and latitude information from the server side, and perform data verification. If the verification is successful, the successfully verified data is returned to the Web side, and if the verification fails, an error message is fed back to the Web side; The web terminal is also used to import the complete information after verification into the accident processor for processing; The accident processor is also used to match the vehicle frame number and the license plate number in the memory after the imported complete information, merge the successfully matched data, and perform a new accident adding process for the unmatched data; The accident processor is also used to generate a fraud report and automatically calculate the fraud factor of the imported processed data; obtain the operator factor configuration, query the factor level under the accident for the calculation results of each factor, obtain the operator fraud configuration, and calculate the overall fraud level according to the level of each factor and the operator fraud configuration; The fraud level is written back into the accident information table. If the conditions for reaching any fraud level are not met, the fraud level is empty and the front-end display page does not show the result.

Images