Insurance behavior monitoring method and device, equipment and storage medium

Abstract

The invention relates to the technical field of insurance buying monitoring intellectualization, can be used for artificial intelligence scenes, financial business scenes and medical health business scenes, and provides an insurance buying behavior monitoring method and device and a storage medium, and the method comprises the steps: obtaining social behavior information and insurance buying behavior information of an insurer; performing insurance buying verification processing on the insurance buying behavior information of the insurance applicant to obtain insurance buying verification information of the insurance applicant; performing feature extraction processing on at least one of the social behavior information, the insurance behavior information and the insurance verification information to obtain insurance feature information associated with the insurance behavior of the insurance applicant; based on a risk assessment model, according to the insurance feature information, determining a risk score that the insurance behavior belongs to an insurance black-grey product behavior; according to the risk score, the insurance buying behavior is processed, so that the monitoring convenience and monitoring accuracy of the insurance buying behavior are improved, the monitoring convenience and monitoring accuracy of the insurance black and grey product behavior are further improved, and sustainable development of financial services and medical health services is promoted.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of insurance monitoring, and can be applied to artificial intelligence business scenarios, financial business scenarios, and medical health business scenarios, and particularly relates to an insurance behavior monitoring method and device, equipment, and a storage medium. BACKGROUND

[0002] At present, with the development of insurance business, insurance black production behaviors frequently occur. The insurance black production behaviors include, for example, an insured person providing false identity information to evade underwriting review, concealing a major illness to cheat high insurance money, illegally obtaining funds through cancellation, loans, and the like. The insurance black production behaviors will have an adverse impact on the sustainable development of insurance business. Correspondingly, financial business, medical health business, and the like associated with the insurance business will also be adversely affected, thereby adversely affecting the sustainable development of the financial business and the medical health business.

[0003] However, in the related art, the insurance behavior of the insured person needs to be audited by relying on manual work, and then the insurance behavior belonging to the insurance black production behavior is identified, which easily leads to poor monitoring convenience and poor monitoring accuracy of the insurance behavior. SUMMARY

[0004] The main purpose of the present application is to provide an insurance behavior monitoring method, device, equipment, and storage medium, aiming to improve the monitoring convenience and accuracy of the insurance behavior, and thereby improve the monitoring convenience and accuracy of the insurance black production behavior.

[0005] In a first aspect, the present application provides an insurance behavior monitoring method, comprising:

[0006] obtaining social behavior information and insurance behavior information of an insured person;

[0007] performing insurance verification processing on the insurance behavior information of the insured person to obtain insurance verification information of the insured person;

[0008] performing feature extraction processing on at least one of the social behavior information, the insurance behavior information, and the insurance verification information to obtain insurance feature information associated with the insurance behavior of the insured person;

[0009] determining, based on a risk assessment model, a risk score of the insurance behavior belonging to the insurance black production behavior according to the insurance feature information;

[0010] processing the insurance behavior according to the risk score corresponding to the insurance behavior.

[0011] In a second aspect, the present application further provides an insurance behavior monitoring device, comprising:

[0012] a data acquisition module configured to acquire social behavior information and insurance application behavior information of an insurance applicant;

[0013] a data verification module configured to perform insurance application verification processing on the insurance application behavior information of the insurance applicant to obtain insurance application verification information of the insurance applicant;

[0014] a feature extraction module configured to perform feature extraction processing on at least one of the social behavior information, the insurance application behavior information, and the insurance application verification information to obtain insurance application feature information associated with the insurance application behavior of the insurance applicant;

[0015] a risk assessment module configured to determine, based on a risk assessment model, a risk score of the insurance application behavior belonging to an insurance black and gray production behavior according to the insurance application feature information;

[0016] an insurance application behavior processing module configured to process the insurance application behavior according to the risk score corresponding to the insurance application behavior.

[0017] In a third aspect, the present application further provides a computer device, which comprises a memory and a processor;

[0018] the memory is configured to store a computer program;

[0019] the processor is configured to execute the computer program and implement the steps of the insurance application behavior monitoring method as described above when executing the computer program.

[0020] In a fourth aspect, the present application further provides a computer readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the steps of the insurance application behavior monitoring method as described above.

[0021] The present application provides an insurance application behavior monitoring method, device, equipment and storage medium, the insurance application behavior monitoring method comprising: acquiring social behavior information and insurance application behavior information of an insurance applicant; performing insurance application verification processing on the insurance application behavior information of the insurance applicant to obtain insurance application verification information of the insurance applicant; performing feature extraction processing on at least one of the social behavior information, the insurance application behavior information, and the insurance application verification information to obtain insurance application feature information associated with the insurance application behavior of the insurance applicant; determining, based on a risk assessment model, a risk score of the insurance application behavior belonging to an insurance black and gray production behavior according to the insurance application feature information; and processing the insurance application behavior according to the risk score corresponding to the insurance application behavior.

[0022] Once the insurance platform obtains information about the policyholder's insurance application behavior, it can verify this information to obtain verification data. Accordingly, the platform can combine the policyholder's social behavior, application behavior, and verification data to determine the likelihood that their application behavior constitutes illegal or unethical activities related to insurance. For example, the platform can extract features from at least one of these three sources to obtain application feature information associated with the policyholder's behavior. Using a risk assessment model, it can determine a risk score based on these features to indicate whether the application behavior falls under illegal or unethical activities. A higher risk score indicates a greater likelihood of such activity, allowing the platform to process the application accordingly. This process eliminates the need for manual review, improving the ease of monitoring application behavior and, consequently, illegal or unethical activities related to insurance. Correspondingly, when insurance business platforms can integrate different types of data, such as policyholders' social behavior information, insurance application behavior information, and insurance verification information, to identify whether a policyholder's insurance application behavior constitutes insurance-related black and gray market activities, it is beneficial to improve the accuracy of monitoring insurance application behavior, and thus improve the accuracy of monitoring insurance-related black and gray market activities. This improved accuracy in monitoring insurance-related black and gray market activities is conducive to promoting the sustainable development of financial and healthcare businesses. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of an application environment for a method for monitoring insurance purchase behavior in one embodiment of this application;

[0025] Figure 2 This is a flowchart illustrating a method for monitoring insurance purchase behavior in one embodiment of this application;

[0026] Figure 3 This is a schematic diagram of the structure of a monitoring device for insurance purchase behavior in one embodiment of this application;

[0027] Figure 4 This is a schematic diagram of the structure of a computer device according to one embodiment of this application;

[0028] Figure 5 This is another structural schematic diagram of a computer device in one embodiment of this application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of the present application.

[0030] The monitoring method of the insurance application behavior provided by the embodiments of the present application can be applied to, for example Figure 1In the application environment, the client communicates with the server through the network. The server can obtain the insurance application behavior information and the insurance application behavior information of the insurance applicant through the client, and perform insurance application verification processing on the insurance application behavior information of the insurance applicant to obtain insurance application verification information of the insurance applicant; perform feature extraction processing on the social behavior information, the insurance application behavior information and the insurance application verification information to obtain insurance application feature information corresponding to the insurance applicant; based on a risk assessment model, determine a risk score of the insurance application behavior of the insurance applicant according to the insurance application feature information; and process the insurance application behavior according to the risk score corresponding to the insurance application behavior, so that the server can feed back the processing strategy for the insurance application behavior, such as the insurance gray production identification strategy, to the client, so that the client can process the insurance application behavior of the insurance applicant according to the processing strategy fed back by the server, such as the gray production identification strategy. In this application, the risk assessment model includes an artificial intelligence model to intelligently predict whether the insurance application behavior of the insurance applicant belongs to the insurance gray production behavior according to the insurance application feature information corresponding to the insurance applicant, and obtain the risk score corresponding to the insurance application behavior. The higher the risk score, the higher the possibility that the insurance application behavior of the insurance applicant belongs to the insurance gray production behavior. Of course, it is not limited to this, and the acquisition of the social behavior information and the insurance application behavior information, the verification of the insurance application behavior information, the feature extraction of the social behavior information, the insurance application behavior information and the insurance application verification information, and the determination of the processing strategy of the insurance application behavior involved in the monitoring method of the insurance application behavior can all be determined by using the corresponding artificial intelligence model. For example, the monitoring method of the insurance application behavior can be applied to an insurance business platform to obtain the social behavior information and the insurance application behavior information of the insurance applicant, and then determine the processing strategy for the insurance application behavior, such as the insurance gray production identification strategy, according to the social behavior information and the insurance application behavior information, and then process the insurance application behavior according to the corresponding processing strategy. For example, the insurance business platform can continuously monitor the insurance application behavior of the insurance applicant before or after the insurance applicant applies for insurance to identify the risk score of the insurance application behavior of the insurance applicant belonging to the insurance gray production behavior, and then process the insurance application behavior of the insurance applicant according to the risk score. This is not limited. Since the risk score of the insurance application behavior of the insurance applicant belonging to the insurance gray production behavior is determined by comprehensively considering the social behavior information, the insurance application behavior information and the insurance application verification information of the insurance applicant, it is beneficial to improve the monitoring convenience and accuracy of the insurance application behavior, and thus improve the monitoring convenience and accuracy of the insurance gray production behavior. Under the condition that the monitoring convenience and accuracy of the insurance gray production behavior are improved, it is beneficial to reduce the adverse effects of the insurance gray production behavior on the financial business and the medical and health business, and thus promote the sustainable development of the financial business and the medical and health business. The client can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers. The application will be described in detail through specific embodiments.

[0031] Referring to Figure 2 as shown, Figure 2 A flowchart of a monitoring method of an insurance application behavior provided by an embodiment of the present application is shown, including the following steps:

[0032] S101: Obtain social behavior information and insurance application behavior information of an insurance applicant.

[0033] For example, the insurance business platform can monitor the insurance application behavior of the insurance applicant to identify whether the insurance application behavior of the insurance applicant belongs to insurance black and gray production behavior. In the case of determining that the insurance application behavior of the insurance applicant belongs to insurance black and gray production behavior or the insurance application behavior does not belong to insurance black and gray production behavior, the insurance business platform can take different processing strategies to process the insurance application behavior of the insurance applicant.

[0034] In the process of monitoring the insurance application behavior of the insurance applicant, the insurance business platform can obtain the social behavior information and the insurance application behavior information of the insurance applicant.

[0035] For example, the social behavior information can be used to indicate at least one of the life habits, social relationships, and consumption habits of the insurance applicant. The insurance application behavior information can be used to indicate at least one of the insurance application track, identity information, and health information of the insurance applicant. The insurance application track can include at least one of the insurance application information, claim information, and cancellation information of the insurance applicant. Of course, the social behavior information and the insurance application behavior information are not limited to this, and are not limited herein.

[0036] For example, the insurance business platform can obtain the social behavior information and the insurance application behavior information of the insurance applicant according to the identity information of the insurance applicant. The identity information of the insurance applicant can include the name of the insurance applicant, the ID number of the insurance applicant, and other information that can be used to determine the identity of the user, which is not limited herein.

[0037] For example, the insurance business platform can obtain the social behavior information and the insurance application behavior information of the insurance applicant by using at least one of an application programming interface (API) related to obtaining the social behavior information and the insurance application behavior information, an automated program related to obtaining the social behavior information and the insurance application behavior information, and other business platforms related to the insurance business platform that can be used to obtain the social behavior information and the insurance application behavior information. Other business platforms that can be used to obtain the social behavior information and the insurance application behavior information include, for example, a bank associated with the insurance business platform, different types of insurance business platforms associated with the insurance business platform, a tax platform, an anti-money laundering system, and the like, which are not limited herein.

[0038] In the case that the social behavior information and the insurance behavior information of the applicant are acquired, the insurance business platform can utilize the multi-source data involved in the social behavior information and the insurance behavior information of the applicant to construct a comprehensive user portrait, so as to serve as a basis for evaluating whether the insurance behavior of the applicant belongs to the insurance black and gray production behavior, and then the risk score of determining that the insurance behavior of the applicant belongs to the insurance black and gray production behavior and the processing strategy of the insurance behavior can be provided, which is beneficial to subsequently improving the monitoring convenience and accuracy of the insurance behavior, and then improving the monitoring convenience and accuracy of the insurance black and gray production behavior.

[0039] S102: performing insurance verification processing on the insurance behavior information of the applicant to obtain insurance verification information of the applicant.

[0040] In some embodiments, in the case that the insurance behavior information of the applicant is acquired, based on the consideration that the applicant may provide false insurance behavior information, such as providing false identity information, informing false health information, etc., the insurance business platform can perform insurance verification processing on the insurance behavior information of the applicant to obtain insurance verification information of the applicant.

[0041] For example, in the case that the insurance behavior information includes the identity information provided by the applicant, the insurance business platform can verify the identity information of the applicant. For example, the insurance business platform can determine the identity verification information of the applicant through the public security network verification, the optical character recognition (OCR) model on the identity card provided by the applicant, the voiceprint recognition on the applicant, etc. The identity verification information can be used to indicate whether the identity information of the applicant is real identity information or false identity information.

[0042] For example, in the case that the insurance behavior information includes the health information informed by the applicant, the insurance business platform can verify the health information of the applicant. For example, the insurance business platform can utilize the natural language processing (NLP) model to analyze the health information of the applicant, and obtain the health information of the applicant. Accordingly, the insurance business platform can verify the health information of the applicant in combination with the preset medical information of the applicant stored in the medical database, such as verifying whether the health information of the applicant is real, to obtain the health verification information of the applicant. The health verification information can be used to indicate whether the health information of the applicant is real health information or false health information.

[0043] In a case where at least one of the identity verification information and the health verification information of the applicant is determined, the insurance business platform can determine the application verification information of the applicant according to at least one of the identity verification information and the health verification information of the applicant. Of course, the application verification information of the applicant is not limited to this, and is not limited herein.

[0044] In a case where the application verification information of the applicant is determined, the insurance business platform can evaluate whether the application behavior of the applicant belongs to the insurance black and gray production behavior by comprehensively considering the social behavior information, the application behavior information and the multi-source data involved in the application verification information of the applicant, and then provide a risk score for determining that the application behavior of the applicant belongs to the insurance black and gray production behavior, and determine a processing strategy for the application behavior, which is beneficial to subsequently improving the monitoring convenience and accuracy of the application behavior, and thus improving the monitoring convenience and accuracy of the insurance black and gray production behavior.

[0045] S103: performing feature extraction processing on at least one of the social behavior information, the application behavior information and the application verification information to obtain application feature information associated with the application behavior of the applicant.

[0046] In some embodiments, the insurance business platform can perform feature extraction processing on at least one of the social behavior information, the application behavior information and the application verification information of the applicant respectively to obtain corresponding application feature information. Each application feature information can be used by the insurance business platform to determine a risk score for determining that the application behavior of the applicant belongs to the insurance black and gray production behavior.

[0047] Taking the case that the application behavior of the applicant includes purchasing a critical illness insurance as an example. The insurance business platform can determine that the application feature information associated with the application behavior of the applicant can cause one of a beneficial effect and an adverse effect on the physical health status of the applicant. In a case where it is determined according to the social behavior information of the applicant that the applicant exercises in a gym three times a week, since gym exercise can have a beneficial effect on the physical health status of the applicant, the insurance business platform can extract the keyword “gym” as the application feature information. In a case where it is determined according to the social behavior information of the applicant that the applicant often works overtime until the early morning, since overtime work can have an adverse effect on the physical health status of the applicant, the insurance business platform can extract the keyword “overtime work” as the application feature information.

[0048] The insurance business platform can determine the insurance application feature information associated with the insurance application behavior of the insurance applicant. The insurance application feature information can provide protection to the insurance applicant or cannot provide protection to the insurance applicant. In a case where it is determined according to the insurance application behavior information of the insurance applicant that the insurance applicant has purchased critical illness insurance of multiple insurance companies in the past three years and has no record of claim, since the behavior of the insurance applicant purchasing critical illness insurance can provide protection related to major diseases to the insurance applicant, the insurance business platform can extract the keyword “multiple critical illness insurance application” as the insurance application feature information. In a case where it is determined according to the insurance application behavior information of the insurance applicant that the insurance applicant has cancelled multiple short-term medical insurance in the past two months, since the behavior of the insurance applicant cancelling medical insurance causes the insurance business platform to be unable to provide protection related to medical treatment to the insurance applicant, the insurance business platform can extract the keyword “short-term cancellation” as the insurance application feature information.

[0049] The insurance business platform can also determine the insurance application feature information associated with the insurance application behavior of the insurance applicant, which can reflect at least one of whether the insurance applicant is real or whether the health status of the insurance applicant is real. In a case where it is determined according to the insurance application verification information of the insurance applicant that the face comparison result of the insurance applicant indicates that the preset face image of the insurance applicant does not match the real-time face image of the insurance applicant, it can be determined that the insurance applicant may have the behavior of providing false identity information, that is, there is a situation that the insurance applicant is not real, and the insurance business platform can extract the keyword “face mismatch” as the insurance application feature information. In a case where it is determined according to the insurance application verification information of the insurance applicant that the voiceprint comparison result of the insurance applicant indicates that the preset voiceprint of the insurance applicant does not match the real-time voice of the insurance applicant, it can be determined that the insurance applicant may have the behavior of providing false identity information, that is, there is a situation that the insurance applicant is not real, and the insurance business platform can extract the keyword “voiceprint mismatch” as the insurance application feature information. In a case where it is determined according to the insurance application verification information of the insurance applicant that the preset medical information of the insurance applicant does not match the health information self-reported by the insurance applicant, it can be determined that the insurance applicant may have concealed health information, that is, there is a situation that the health status of the insurance applicant is not real, and the insurance business platform can extract the keyword “health information mismatch” as the insurance application feature information.

[0050] Of course, the insurance application behavior of the insurance applicant is not limited to purchasing critical illness insurance, and the insurance application feature information associated with the insurance application behavior of the insurance applicant is also not limited to this, which is not limited herein.

[0051] In a case where the insurance business platform determines the insurance feature information associated with the insurance behavior of the applicant, the insurance business platform can comprehensively determine the insurance feature information associated with the insurance behavior of the applicant according to the social behavior information, the insurance behavior information, and the insurance verification information of the applicant, evaluate whether the insurance behavior of the applicant belongs to the insurance black and gray production behavior, and further provide a risk score for determining whether the insurance behavior of the applicant belongs to the insurance black and gray production behavior and a processing strategy for the insurance behavior, thereby facilitating subsequent improvement of the monitoring convenience and accuracy of the insurance behavior and further improvement of the monitoring convenience and accuracy of the insurance black and gray production behavior.

[0052] In a case where the insurance business platform determines the insurance feature information associated with the insurance behavior of the applicant, the insurance business platform can comprehensively determine the insurance feature information associated with the insurance behavior of the applicant according to the social behavior information, the insurance behavior information, and the insurance verification information of the applicant, evaluate whether the insurance behavior of the applicant belongs to the insurance black and gray production behavior, and further provide a risk score for determining whether the insurance behavior of the applicant belongs to the insurance black and gray production behavior and a processing strategy for the insurance behavior, thereby facilitating subsequent improvement of the monitoring convenience and accuracy of the insurance behavior and further improvement of the monitoring convenience and accuracy of the insurance black and gray production behavior.

[0053] In some embodiments, in a case where the insurance business platform determines the insurance feature information associated with the insurance behavior of the applicant, the insurance business platform can input the insurance feature information into the risk assessment model to enable the risk assessment model to determine the risk score of the insurance behavior belonging to the insurance black and gray production behavior according to the insurance feature information.

[0054] For example, the risk assessment model can include a machine learning model. The risk assessment model can analyze whether the insurance behavior of the applicant has identity information fraud, false health information, capital cashing, and beneficiary relationship forgery according to the insurance feature information to determine the risk score of the insurance behavior belonging to the insurance black and gray production behavior. In a case where the risk score of the insurance behavior belonging to the insurance black and gray production behavior is higher, the possibility of the insurance behavior belonging to the insurance black and gray production behavior is higher. In a case where the risk score of the insurance behavior belonging to the insurance black and gray production behavior is lower, the possibility of the insurance behavior belonging to the insurance black and gray production behavior is lower.

[0055] For example, the insurance behavior of the applicant includes the purchase of critical illness insurance. In a case where the insurance feature information associated with the insurance behavior of the applicant includes at least one of “gym”, “multiple critical illness insurance”, “face matching”, and “health information matching”, the risk assessment model determines a lower risk score of the insurance behavior belonging to the insurance black and gray production behavior. In a case where the insurance feature information associated with the insurance behavior of the applicant includes at least one of “overtime work”, “short-term insurance refund”, “voiceprint mismatch”, and “health information mismatch”, the risk assessment model determines a higher risk score of the insurance behavior belonging to the insurance black and gray production behavior. Of course, the insurance behavior of the applicant and the insurance feature information are not limited to this and are not limited herein.

[0056] In the case of determining the risk score of the insurance application behavior according to the risk assessment model and the insurance application feature information, the risk score of the insurance application behavior can be used to determine the possibility of the insurance application behavior belonging to the insurance black and gray production behavior. Then, the insurance business platform can process the insurance application behavior according to the risk score of the insurance application behavior, which is beneficial to improve the monitoring convenience and accuracy of the insurance application behavior, and further improve the monitoring convenience and accuracy of the insurance black and gray production behavior.

[0057] S105: processing the insurance application behavior according to the risk score of the insurance application behavior.

[0058] In some embodiments, the insurance business platform can determine the risk level of the insurance application behavior according to the risk score of the insurance application behavior, and process the insurance application behavior according to the risk level of the insurance application behavior and the processing strategy corresponding to the risk level.

[0059] For example, the risk level of the insurance application behavior can include low, medium and high. When the risk score of the insurance application behavior is less than or equal to a first score threshold, the risk level of the insurance application behavior can be determined as low. When the risk score of the insurance application behavior is greater than the first score threshold and less than or equal to a second score threshold, the risk level of the insurance application behavior can be determined as medium. The second score threshold is greater than the first score threshold. When the risk score of the insurance application behavior is greater than the second score threshold, the risk level of the insurance application behavior can be determined as high. Of course, the risk level of the insurance application behavior is not limited to this, and the risk level can be pre-set or set by the staff. Correspondingly, the risk score range corresponding to the risk level can be pre-set or set by the staff, which is not limited herein.

[0060] For different insurance behaviors of different risk levels, the insurance business platform can adopt different processing strategies to process the corresponding insurance behaviors. Different processing strategies can correspond to different risk levels one by one. For example, the processing strategy can include an insurance black and gray production identification strategy. In the case of a low risk level of the insurance behavior, the insurance business platform can determine that the possibility of the insurance behavior belonging to the insurance black and gray production behavior is low, and the insurance black and gray production identification strategy for the insurance behavior can be maintained as automatic monitoring. In the case of a medium risk level of the insurance behavior, the insurance business platform can determine that the possibility of the insurance behavior belonging to the insurance black and gray production behavior is moderate, and the insurance black and gray production identification strategy for the insurance behavior can be maintained as automatic monitoring while limiting the transaction behavior related to the insurance behavior. For example, when the policyholder insures, the policyholder's insurance amount is limited. For another example, when the policyholder claims, the policyholder's claim amount is manually audited, and the like, which is not limited herein. In the case of a high risk level of the insurance behavior, the insurance business platform can determine that the possibility of the insurance behavior belonging to the insurance black and gray production behavior is high, and the insurance black and gray production identification strategy for the insurance behavior can be switched to manual audit, so that relevant personnel can focus on auditing the insurance behavior with a high risk level, so as to avoid the adverse effects of insurance black and gray production behavior on the sustainable development of insurance business as much as possible, and further avoid the adverse effects of insurance black and gray production behavior on the sustainable development of financial business and medical and health business as much as possible.

[0061] In the case of processing the insurance behavior according to the risk score corresponding to the insurance behavior, the processing strategy for the insurance behavior can be adapted to the risk score corresponding to the insurance behavior, which is beneficial to improve the processing flexibility of the insurance behavior, and further beneficial to reduce the adverse effects of the insurance behavior belonging to the insurance black and gray production behavior on the insurance business, thereby promoting the sustainable development of the insurance business. Therefore, it is beneficial to promote the sustainable development of financial business and medical and health business.

[0062] In some embodiments, according to the insurance behavior information, the social media information associated with the policyholder is obtained; and according to the social media information, the social behavior information of the policyholder is determined.

[0063] For example, the social media information can include public information such as social media and news reports. The social media includes different types of application programs, television information, and the like, which are not limited herein.

[0064] The insurance business platform can obtain the social media information associated with the policyholder according to the insurance behavior information of the policyholder.

[0065] For example, the insurance business platform can determine that the social media information is associated with the applicant in the case that the social media information includes the identity information of the applicant. For example, the identity information of the applicant appearing in the social media information can include one or more of the name of the applicant, the ID number of the applicant, the mobile phone number of the applicant, the social media account of the applicant, and the like, without limitation. In an example embodiment, the applicant provides the identity document information, such as an ID card picture, an ID card scan, and the like, to the insurance business platform, without limitation. The insurance business platform can identify the identity information of the applicant, such as at least one of the name of the applicant and the ID number of the applicant, by using an OCR model to recognize the identity document information. Accordingly, the insurance business platform can obtain the mobile phone number of the applicant. The insurance business platform can use a search engine or a social media API to search for the corresponding social media information according to the identity information of the applicant, such as the name of the applicant, the ID number of the applicant, and the mobile phone number of the applicant. In the case that the corresponding social media information is searched, the insurance business platform can analyze the text content in the social media information by using an NLP model to further determine whether the social media information is associated with the applicant. For example, in the case that at least one of the event, the person, and the location mentioned in the text content in the social media information matches the identity information of the applicant, it can be determined that the social media information is associated with the applicant.

[0066] In the case that the corresponding social media information is searched, the insurance business platform can also screen the social media information to obtain the social media information that can be used to identify whether the insurance application behavior of the applicant belongs to the insurance black production behavior.

[0067] For example, the insurance application behavior information can include an insurance application trajectory of the insurance applicant. For example, in a case where the activity involved in the social media information matches the type of insurance indicated by the insurance application trajectory of the insurance applicant, it can be determined that the social media information is associated with the insurance applicant. For example, in a case where the activity involved in the social media information includes travel and the type of insurance indicated by the insurance application trajectory of the insurance applicant includes travel insurance, it can be determined that the activity matches the type of insurance, and thus the social media information is associated with the insurance applicant. For another example, in a case where the activity involved in the social media information includes sports and the type of insurance indicated by the insurance application trajectory of the insurance applicant includes health insurance, it can be determined that the activity matches the type of insurance, and thus the social media information is associated with the insurance applicant. Of course, the present disclosure is not limited thereto, and no limitation is made herein. For example, in a case where the geographic location involved in the social media information matches the insurance application location indicated by the insurance application trajectory of the insurance applicant, it can be determined that the social media information is associated with the insurance applicant. For example, the insurance business platform can extract the geographic location in the social media information by using a geographic location API. Of course, the present disclosure is not limited thereto, and no limitation is made herein.

[0068] For example, after obtaining the social media information associated with the insurance applicant, the insurance business platform can further perform secondary screening on the social media information. For example, the insurance business platform can perform secondary screening on the social media information according to at least one of the source reliability, timeliness, and completeness of the social media information.

[0069] For example, in a case where the information publishing platform to which the obtained social media information belongs does not belong to the social media account of the insurance applicant himself, official news media, or an officially certified network platform, it can be determined that the source reliability of the social media information is low, and thus the social media information can be excluded. Correspondingly, in a case where the information publishing platform to which the obtained social media information belongs belongs to the social media account of the insurance applicant himself, official news media, or an officially certified network platform, it can be determined that the source reliability of the social media information is high, and thus the social media information can be retained.

[0070] For example, in a case where a time interval between the information publishing time of the obtained social media information and the current time is greater than or equal to a preset time threshold, it can be determined that the timeliness of the social media information is low, and the social media information can be discarded. The preset time threshold can include one month, two months, three months, half a year, etc. The preset time threshold can be pre-set or set by the staff, which is not limited herein. In an exemplary embodiment, the preset time threshold corresponding to different types of social media information can be different or the same. For example, in a case where the social media information includes public information of social media, the insurance business platform can retain public information of social media with a time interval between the information publishing time and the current time less than one month as social media information associated with the applicant. For another example, in a case where the social media information includes news reports, the insurance business platform can retain news reports with a time interval between the information publishing time and the current time less than three months as social media information associated with the applicant. Of course, it is not limited to this, which is not limited herein. Accordingly, based on the consideration of improving the timeliness of the applicant's insurance application behavior information, the insurance business platform can obtain the social media information associated with the applicant according to the applicant's insurance application behavior information in the preset time period. The preset time period can include the last six months, the last year, the last three months, etc. The preset time period can be pre-set or set by the staff, which is not limited herein.

[0071] For example, in a case where the occurrence frequency of the content involved in the obtained social media information is less than or equal to a preset frequency threshold, it can be determined that the completeness of the social media information is low, for example, the social media information cannot be used to determine the social behavior information of the applicant, and the social media information can be discarded. Accordingly, in a case where the occurrence frequency of the content involved in the obtained social media information is greater than the preset frequency threshold, it can be determined that the completeness of the social media information is high, for example, the social media information can be used to determine the social behavior information of the applicant, and the social media information can be retained. In an exemplary embodiment, the social media information retained by the insurance business platform includes, for example, information such as food, sports, travel, etc. shared frequently by the applicant on social media, the applicant appearing in the same social activity in news reports multiple times, etc., which is not limited herein.

[0072] In a case where the social media information associated with the applicant is determined, the insurance business platform can analyze the social media information associated with the applicant by using an NLP model, an OCR model, and the like, to obtain the social behavior information of the applicant. The social behavior information can include at least one of the life habit of the applicant and the social relationship of the applicant. Accordingly, the social behavior information of the applicant can be used by the insurance business platform to analyze whether the application behavior of the applicant involves the insurance black production behavior. For example, in a case where the economic status of the applicant is determined according to the social behavior information of the applicant, if the applicant frequently performs the policy cancellation or the loan, the insurance business platform can speculate that the application behavior of the applicant may involve the insurance black production behavior of illegally obtaining funds by means of policy cancellation, loan, and the like. Of course, the present disclosure is not limited thereto.

[0073] In a case where the social media information associated with the applicant is determined according to the application behavior information, the social media information can be used to determine the social behavior information of the applicant, which is beneficial to improve the convenience of determining the social behavior information of the applicant. Accordingly, the social behavior information of the applicant can be used to determine the risk score of the application behavior of the applicant belonging to the insurance black production behavior, which is beneficial to improve the convenience and accuracy of monitoring the application behavior, and further improve the convenience and accuracy of monitoring the insurance black production behavior.

[0074] In some embodiments, the social media information is analyzed to obtain at least one of the life habit, the social relationship, and the consumption habit of the applicant; and the social behavior information of the applicant is determined according to at least one of the life habit, the social relationship, and the consumption habit of the applicant.

[0075] For example, the social media information can include videos, pictures, text content, and the like. The insurance business platform can analyze the social media information to obtain at least one of the life habit, the social relationship, and the consumption habit of the applicant. In a case where the social media information includes videos or pictures, the insurance business platform can analyze the videos or pictures to convert the content of the videos or pictures into a text description or keywords. Accordingly, in a case where the social media information includes text content, the insurance business platform can analyze the text content to obtain a text description or keywords corresponding to the text content. The present disclosure is not limited thereto.

[0076] For example, in a case where the social media information is analyzed, the analyzed social media information can be used to indicate at least one of the life habit, the social relationship, and the consumption habit of the applicant.

[0077] For example, the social media information includes the public information of the applicant on the social media. In the case that the public information on the social media includes a video, and the video involves "three net celebrity restaurants explored this week", "dinner with friends on the weekend", it can be determined that the applicant posts videos related to dining out and exploring restaurants multiple times per week. In the case that the public information on the social media includes a picture, and the picture involves at least one of sports such as hiking, yoga, and bodybuilding, it can be determined that the picture shared by the applicant shows that he or she often participates in outdoor sports or indoor sports. In the case that the public information on the social media includes text content, and the text content involves "buying coffee or milk tea at a fixed time every week", it can be determined that the text content shared by the applicant shows that he or she often buys beverages. The insurance business platform can comprehensively determine the life habits of the applicant as being socially active, liking food, and exercising, based on the information analysis and processing results of the public information on the social media. Of course, the public information on the social media and the life habits of the applicant are not limited to this, and are not limited herein.

[0078] Correspondingly, in the case that the public information on the social media includes a picture, and the picture involves a group photo of the applicant with friends and colleagues, and mentions "dining with colleagues" and "traveling with friends", it can be determined that the applicant has close relationships with friends and colleagues. In the case that the public information on the social media includes text content, and the text content involves the applicant participating in family gatherings and weddings, it can be determined that the applicant has an active family social circle. The insurance business platform can comprehensively determine the social relationships of the applicant as being close to friends, colleagues, and family members, based on the information analysis and processing results of the public information on the social media. Of course, the public information on the social media and the social relationships of the applicant are not limited to this, and are not limited herein.

[0079] Correspondingly, in the case that the public information on the social media includes a picture, such as the applicant often sharing purchase records of luxury goods on the social media, it can be determined that the applicant has high consumption ability and likes luxury goods. In the case that the public information on the social media includes text content, such as the applicant sharing high-end hotel and travel experiences multiple times, it can be determined that the applicant likes traveling. The insurance business platform can comprehensively determine the consumption habits of the applicant as having high consumption ability, liking luxury goods, and liking traveling, based on the information analysis and processing results of the public information on the social media. Of course, the public information on the social media and the consumption habits of the applicant are not limited to this, and are not limited herein.

[0080] Taking a news report as an example, the news report can involve the public activities of the applicant. In the case that the news report mentions the applicant as a participant in a public welfare activity, a participant in a marathon event, and a guest at a high-end social activity, the insurance business platform can determine, according to the information analysis processing result of the news report, that the applicant's living habit is actively participating in public welfare activities, loves sports, and that the applicant's social relationship is widely connected with social celebrities, entrepreneurs, and the like. Of course, the news report, the applicant's living habit, and the applicant's social relationship are not limited to this, and are not limited herein.

[0081] In the case that the information analysis processing is performed on the social media information to obtain at least one of the living habit, the social relationship, and the consumption habit of the applicant, the insurance business platform can determine the social behavior information of the applicant according to at least one of the living habit, the social relationship, and the consumption habit of the applicant, which is conducive to improving the convenience of determining the social behavior information of the applicant. The social behavior information of the applicant can be used by the insurance business platform to subsequently identify whether the application behavior of the applicant belongs to the insurance black and gray production behavior, which is conducive to subsequently improving the convenience of monitoring the application behavior, and further improving the convenience of monitoring the insurance black and gray production behavior.

[0082] In some embodiments, the application behavior information of the applicant at least includes at least one of the identity information and the health information of the applicant; the application verification processing is performed on the application behavior information of the applicant to obtain the application verification information of the applicant, including at least one of the following: according to the matching result between the preset identity information of the applicant and the identity information of the applicant, verifying the identity information of the applicant to obtain the identity verification information of the applicant; according to the matching result between the preset medical information of the applicant and the health information of the applicant, verifying the health information of the applicant to obtain the health verification information of the applicant.

[0083] For example, the identity information of the applicant can be provided by the applicant himself to the insurance business platform. The preset identity information of the applicant can be determined by the insurance business platform based on the identity information of the applicant through the public security network, and the preset identity information of the applicant can be used to verify whether the identity information of the applicant is true.

[0084] In a case where the identity information provided by the applicant includes an identity certificate, the insurance business platform can obtain the identity card number of the applicant through the identity certificate. Correspondingly, the insurance business platform can obtain the preset identity information of the applicant, such as the preset face image and the preset voiceprint of the applicant, through the public security network according to the identity card number of the applicant. The identity information of the applicant can include at least one of the real-time face image and the real-time voice of the applicant. In a case where the matching result between the preset identity information of the applicant and the identity information of the applicant is not matched, it can be determined that the identity verification information of the applicant is that the identity information of the applicant is not true. For example, in a case where the real-time voice of the applicant does not match the preset voiceprint of the applicant, such as a case where the real-time voice of the applicant involves background sound with the voice of another person, the insurance business platform can speculate that the other person is guiding the applicant to perform the insurance black and gray production behavior, and then the insurance business platform can determine that the identity verification information of the applicant is that the identity information is not true. For another example, in a case where the real-time face image of the applicant does not match the preset face image, the insurance business platform can speculate that the applicant provides false identity information for subsequent performance of the insurance black and gray production behavior, and then the insurance business platform can determine that the identity verification information of the applicant is that the identity information is not true. Correspondingly, in a case where the matching result between the preset identity information of the applicant and the identity information of the applicant is matched, such as a case where the real-time face image of the applicant matches the preset face image and / or the real-time voice of the applicant matches the preset voiceprint of the applicant, it can be determined that the identity verification information of the applicant is that the identity information of the applicant is true. Of course, the preset identity information of the applicant and the identity information of the applicant are not limited to this, which is not limited herein.

[0085] For example, the health information of the applicant can be informed by the applicant to the insurance business platform. In a case where the insurance business platform obtains the health information of the applicant, the insurance business platform can perform information analysis and processing on the health information to obtain key information corresponding to the health information, and the key information includes at least one of a disease name, a diagnosis time, and a treatment condition.

[0086] Taking the applicant Zhang San as an example. The health information informed by the applicant Zhang San to the insurance business platform includes “2020 was hospitalized for acute appendicitis surgery and has recovered”, and then the insurance business platform can analyze the health information of Zhang San by using the NLP model to obtain the corresponding key information. The key information includes the disease name of acute appendicitis, the time range of 2020, the treatment condition of hospitalization surgery, and the recovery.

[0087] For example, the preset medical information of the applicant can be obtained by the insurance business platform from a medical database based on the identity information of the applicant, and the preset medical information of the applicant can be used to verify whether the health information of the applicant is true. For example, the insurance business platform can determine the query condition of the medical database according to the identity card number of the applicant and the key information corresponding to the health information of the applicant, so that the medical database outputs the preset medical information of the applicant according to the query condition. The source of the medical database can be medical insurance data, hospital electronic health records, etc. Before the insurance business platform obtains the preset medical information of the applicant from the medical database, the authorization of the applicant needs to be obtained in advance. For example, before obtaining the identity information of the applicant, the insurance business platform asks the applicant whether to authorize the insurance business platform to query the preset medical information of the applicant. Of course, the way to obtain the authorization of the applicant in advance is not limited to this, which is not limited herein.

[0088] Taking the applicant Zhang San as an example, the insurance business platform can query whether the hospitalization record of Zhang San in 2020 is stored in the medical database according to the identity card number of Zhang San. The query condition includes at least one of the identity card number of Zhang San, the time range, and the disease name. The time range can include the whole year of 2020. The disease name can include acute appendicitis. The medical database can output the preset medical information of Zhang San according to the query condition.

[0089] For example, in the case that the medical database outputs the preset medical information of the applicant, the insurance business platform can verify the health information of the applicant according to the matching result between the preset medical information of the applicant and the health information of the applicant, and obtain the health verification information of the applicant. In the case that the preset medical information is consistent with the health information according to the matching result, it can be determined that the health verification information of the applicant is true. In the case that the preset medical information is partially inconsistent with the health information according to the matching result, it can be determined that the health verification information of the applicant is not true. In the case that the preset medical information is completely inconsistent with the health information according to the matching result, it can be determined that the health verification information of the applicant is concealment of medical history.

[0090] Taking the policyholder as Zhang San as an example, in a case that the matching result obtained by comparing the preset medical information of Zhang San with the health information of Zhang San is that the preset medical information is consistent with the health information, it can be determined that the health verification information of Zhang San is that the health information of Zhang San is true. For example, in a case that there is a record of Zhang San being hospitalized for acute appendicitis in 2020 in the medical database, and the diagnosis result and the treatment method are consistent, it can be determined that the preset medical information of Zhang San is consistent with the health information. In a case that the matching result obtained by comparing the preset medical information of Zhang San with the health information of Zhang San is that the preset medical information is inconsistent with the health information, it can be determined that the health verification information of Zhang San is that the health information of Zhang San is not true. For example, in a case that there is a record of Zhang San being treated in 2020 in the medical database, but at least one of the diagnosis result and the treatment method is inconsistent, it can be determined that the preset medical information of Zhang San is inconsistent with the health information. For example, the diagnosis result of Zhang San is chronic appendicitis, rather than acute appendicitis. For another example, the treatment method of Zhang San is conservative treatment, rather than hospitalization. For another example, in a case that there is no record of Zhang San being treated in 2020 in the medical database, it can be determined that the health verification information of Zhang San is that Zhang San conceals medical history.

[0091] For example, the preset medical information of the policyholder can also be obtained by the insurance business platform based on at least one of the medical records provided by the policyholder and the medical database, so that the preset medical information of the policyholder can be used to verify whether the health information of the policyholder is true. For example, the insurance business platform can perform information analysis and processing on the medical records provided by the policyholder to obtain the preset medical information of the policyholder.

[0092] Taking the policyholder as Li Si as an example, in a case that Li Si provides his own medical records, the insurance business platform can use the NLP model to perform information analysis and processing on the medical records provided by Li Si to obtain the preset health information of Li Si. For example, the preset health information shown in the medical records of Li Si is “the patient was diagnosed with hypertension in March 2019 and has been taking antihypertensive drugs for a long time”. In a case that the health information informed by Li Si to the insurance business platform is “no history of hypertension”, the insurance business platform can use the NLP model to extract the key information corresponding to the health information of Li Si as denying the history of hypertension. Correspondingly, the insurance business platform can query the medical database according to the ID number of Li Si. In a case that there is a record of Li Si being diagnosed with hypertension in March 2019 in the medical database, and there are multiple purchase records of antihypertensive drugs, it can be determined that the preset medical information of Li Si is inconsistent with the health information of Li Si, and further it can be determined that the health verification information of Li Si is at least one of concealing medical history or the health information being not true.

[0093] In an exemplary embodiment, the insurance business platform can perform entity recognition on the health information provided by the applicant according to a named entity recognition (NER) model, such as recognizing at least one of the disease name, time range, diagnosis result, treatment method, and the like included in the health information, to determine the key information corresponding to the health information. Of course, it is not limited thereto, and is not limited herein.

[0094] In the process of comparing the preset medical information of the applicant with the health information of the applicant to determine the matching result between the preset medical information and the health information, and then verifying the health information of the applicant to obtain the health verification information of the applicant, the insurance business platform can not only compare whether the diseases involved in the health information exist in the preset medical information, but also compare whether the time, diagnosis result, treatment method, and the like involved in the health information exist in the preset medical information, which is beneficial to improve the determination accuracy of the health verification information of the applicant.

[0095] In the case where the insurance business platform can determine the identity verification information of the applicant according to the preset identity information of the applicant and the identity information of the applicant, it is beneficial to improve the verification convenience of the identity information of the applicant. In the case where the insurance business platform can determine the health verification information of the applicant according to the preset medical information of the applicant and the health information of the applicant, it is beneficial to improve the verification convenience of the health information of the applicant. The insurance application verification information of the applicant can include at least one of the identity verification information of the applicant and the health verification information of the applicant, and the insurance application verification information of the applicant can be used for subsequent evaluation of whether the insurance application behavior of the applicant belongs to the insurance black and gray production behavior, which is beneficial to improve the monitoring convenience of the insurance application behavior, and further improve the monitoring convenience of the insurance black and gray production behavior.

[0096] In some embodiments, the insurance application feature information is input into a risk assessment model, so that the risk assessment model assesses the influence coefficient of the insurance application feature information on the insurance black and gray production behavior according to the feature information; and the risk score of the insurance application behavior of the applicant belonging to the insurance black and gray production behavior is determined according to the sum of the influence coefficients corresponding to all the insurance application feature information.

[0097] For example, when insurance application characteristics are input into a risk assessment model, the model can assess whether the application involves insurance-related illicit activities. If the model determines that the application does not involve such activities, then the impact of these characteristics on whether the application constitutes insurance-related illicit activities is negative, and the influence coefficient of these characteristics on whether the application constitutes such activities can be negative. Conversely, if the model determines that the application involves insurance-related illicit activities, then the impact of these characteristics on whether the application constitutes such activities is positive, and the influence coefficient of these characteristics on whether the application constitutes such activities can be positive.

[0098] The insurance business platform can synthesize the influence coefficients corresponding to all insurance application characteristics, determine the sum of the influence coefficients for each characteristic, and then obtain a risk score indicating whether the insurance application behavior constitutes illegal or unethical activities in the insurance industry. Accordingly, the platform can store at least one of the following in an unstructured database: the policyholder's application characteristics, the influence coefficients corresponding to those characteristics, and the risk score for the application behavior. This data can be used for subsequent analysis to determine whether the policyholder's application behavior constitutes illegal or unethical activities in the insurance industry.

[0099] Let's take Zhang San as an example. Zhang San's social behavior information can be determined based on his social media posts, such as the life updates he shares on Weibo. These posts might include things like "Having dinner with friends at Haidilao this weekend #food", "Checking in at my new gym! Three workouts a week #healthylife", and "Working overtime until 3 AM #programmerdaylife". Zhang San's insurance information could include his insurance records, such as having purchased critical illness insurance.

[0100] The insurance business platform can perform feature extraction processing on the social behavior information and the insurance application behavior information of Zhang San, and obtain corresponding insurance application feature information. The insurance application feature information includes, for example, "overtime work" and "gym". In the case of inputting the insurance application feature information into the risk assessment model, the risk assessment model can determine that "overtime work" will have an adverse effect on the health of Zhang San, and thus determine that the influence coefficient of the insurance application feature information "overtime work" on the insurance application behavior belonging to the insurance black and gray production behavior is a positive number, such as 0.2. The risk assessment model can determine that "gym" will have a beneficial effect on the health of Zhang San, and thus determine that the influence coefficient of the insurance application feature information "gym" on the insurance application behavior belonging to the insurance black and gray production behavior is a negative number, such as -0.1. For another example, in the case of performing feature extraction processing on the social behavior information of Zhang San and determining that Zhang San has worked overtime for three consecutive weeks, it can be determined that the corresponding insurance application feature information is "three consecutive weeks of overtime work", and accordingly, the influence coefficient of the insurance application feature information "three consecutive weeks of overtime work" on the insurance application behavior belonging to the insurance black and gray production behavior is a positive number, such as 0.3. Of course, the insurance application feature information and the influence coefficient corresponding to the insurance application feature information are not limited to this, and are not limited herein.

[0101] The insurance business platform can determine the risk score of the insurance application behavior of Zhang San belonging to the insurance black and gray production behavior by synthesizing the sum of the influence coefficients corresponding to all the insurance application feature information of Zhang San. For example, according to the influence coefficient corresponding to "overtime work" being 0.2, the influence coefficient corresponding to "gym" being -0.1, and the influence coefficient corresponding to "three consecutive weeks of overtime work" being 0.3, it can be determined that the risk score of the insurance application behavior of Zhang San belonging to the insurance black and gray production behavior is 0.4. The risk score of the insurance application behavior of Zhang San belonging to the insurance black and gray production behavior can be used by the insurance business platform to determine the processing strategy for the insurance application behavior of Zhang San.

[0102] In the case of using the risk assessment model, combining the insurance application feature information corresponding to the insurance application behavior of the insurance applicant, determining the influence coefficient of the insurance application feature information on the insurance application behavior belonging to the insurance black and gray production behavior, and then combining the sum of the influence coefficients of all the insurance application feature information on the insurance application behavior belonging to the insurance black and gray production behavior, the risk score of the insurance application behavior of the insurance applicant belonging to the insurance black and gray production behavior can be determined, which is beneficial to improving the convenience of determining the risk score of the insurance application behavior belonging to the insurance black and gray production behavior. The risk score of the insurance application behavior belonging to the insurance black and gray production behavior can be used by the insurance business platform to determine the processing strategy for the insurance application behavior of the insurance applicant, which is beneficial to improving the convenience of processing the insurance application behavior, and further improving the convenience of processing the insurance black and gray production behavior.

[0103] In some embodiments, the risk assessment model is used to evaluate at least one of the identity authenticity, the insurance application behavior rationality, and the fund flow rationality of the insurance applicant according to the insurance application feature information, and obtain the influence coefficient of the insurance application feature information on the insurance application behavior belonging to the insurance black and gray production behavior.

[0104] In the case that the insurance application feature information is obtained by performing feature extraction processing on at least one of the insurance applicant's social behavior information, insurance application behavior information, and insurance application verification information, the insurance application feature information corresponding to the insurance application verification information can be used by the risk assessment model to assess the identity authenticity of the insurance applicant, and the insurance application feature information corresponding to the social behavior information and the insurance application behavior information can be used by the risk assessment model to assess the rationality of the insurance application behavior of the insurance applicant. Correspondingly, the insurance application behavior information can be used by the risk assessment model to assess the rationality of the flow of funds of the insurance applicant.

[0105] For example, the risk assessment model can assess the identity authenticity of the insurance applicant according to the identity verification information of the insurance applicant. In the case that the identity verification information of the insurance applicant is not true, the risk assessment model can determine that the identity authenticity of the insurance applicant is not true. In the case that the identity authenticity of the insurance applicant is not true, the influence coefficient of this insurance application feature information on the insurance application behavior belonging to the insurance black and gray production behavior is, for example, 0.3. Correspondingly, in the case that the identity verification information of the insurance applicant is true, the risk assessment model can determine that the identity authenticity of the insurance applicant is true. In the case that the identity authenticity of the insurance applicant is true, the influence coefficient of this insurance application feature information on the insurance application behavior belonging to the insurance black and gray production behavior is, for example, -0.3. Of course, it is not limited to this, and is not limited herein.

[0106] For example, in the case that the risk assessment model assesses the rationality of the insurance application behavior of the insurance applicant according to the insurance application feature information, the insurance application feature information can include at least one of the following: at least one of the basic insurance application data obtained by feature extraction processing on the social behavior information of the insurance applicant and the insurance application behavior information of the insurance applicant, time dimension data, financial correlation data, correlation party data, behavior anomaly data, and external verification data. The basic insurance application data can include at least one of the insurance application frequency, the type of insurance, the amount of insurance, and the payment method of the insurance applicant. The basic insurance application data can be determined by the insurance application behavior information obtained through a business platform related to the insurance business platform, such as a core business system of the insurance business platform. The time dimension data can include at least one of the insurance application time point, the interval period of the insurance policy, and the renewal timeliness. The time dimension data can be obtained by time stamp analysis on the insurance policy included in the insurance application behavior information. The financial correlation data can include at least one of the premium income proportion, the insurance policy loan record, and the refund rate of the insurance policy. The financial correlation data can be determined by monitoring the fund flow through a business platform related to the insurance business platform, such as a financial system. The correlation party data can include at least one of the beneficiary relationship, the correlation degree of multi-person joint insurance, and the cross-company insurance application record. The correlation party data can be obtained by relationship graph analysis on the insurance application behavior information and the social behavior information of the insurance applicant. The behavior anomaly data can include at least one of the number of health information modifications, the information conflict rate, and the frequency of underwriting urging. The behavior anomaly data can be obtained by operation log auditing on the insurance application behavior information of the insurance applicant. The external verification data can include at least one of the authenticity of the income proof, the asset-liability matching degree, and the social media consumption level. The external verification data can be obtained by cross-verification on the insurance application behavior information, such as tax data and bank data, and the social behavior information of the insurance applicant. Of course, the insurance application feature information and the feature extraction method corresponding to the insurance application feature information are not limited to this, and are not limited herein.

[0107] The risk assessment model can perform at least one of a basic insurance mode analysis and a time series analysis in the process of assessing the rationality of the insurance application behavior of the insurance applicant according to the insurance application feature information. For example, in the process of performing the basic insurance mode analysis on the rationality of the insurance application behavior of the insurance applicant according to the insurance application feature information, if the risk assessment model determines that the insurance application density of the insurance applicant is more than 5 times the preset insurance application density threshold value and the single insurance type concentration rate is 100%, the risk assessment model can determine that the rationality of the insurance application behavior of the insurance applicant is unreasonable, and thus the possibility that the insurance application behavior of the insurance applicant belongs to the insurance black and gray production behavior is relatively high. In the process of performing the time series analysis on the rationality of the insurance application behavior of the insurance applicant according to the insurance application feature information, if the risk assessment model determines that the insurance application behavior of the insurance applicant has at least one of the following situations: the average insurance application interval corresponding to the insurance application behavior of the insurance applicant is less than the preset insurance application interval threshold value, the total number of insurance application behaviors of the insurance applicant in the same preset time period is greater than or equal to the preset number threshold value, and the coincidence rate between the insurance application time of the insurance applicant and the issuing time corresponding to the preset medical information of the insurance applicant is greater than or equal to the preset coincidence rate threshold value, the risk assessment model can determine that the rationality of the insurance application behavior of the insurance applicant is unreasonable, and thus the possibility that the insurance application behavior of the insurance applicant belongs to the insurance black and gray production behavior is relatively high. For example, the average insurance application interval corresponding to the insurance application behavior of the insurance applicant is only 14 days, and the preset insurance application interval threshold value is 90 days. For another example, the insurance applicant applies for insurance at 10-11 am on weekdays, which has a high possibility of mechanical operation. For another example, the insurance applicant learns that the medical examination report shows lung cancer indicators on July 25, 2023, and purchases a critical illness insurance on the same day. Of course, it is not limited to this, and no limitation is made herein.

[0108] In the case where the rationality of the insurance application behavior of the insurance applicant is unreasonable, the influence coefficient of the insurance application feature information on the insurance application behavior belonging to the insurance black and gray production behavior is, for example, 0.4. Correspondingly, in the case where the rationality of the insurance application behavior of the insurance applicant is reasonable, the influence coefficient of the insurance application feature information on the insurance application behavior belonging to the insurance black and gray production behavior is, for example, -0.4. Of course, it is not limited to this, and no limitation is made herein.

[0109] For example, the risk assessment model assesses the rationality of the insured person's fund flow according to the insured characteristic information. The insured characteristic information can include at least one of the insured person's account basic data, transaction flow data, associated party graph, behavior matching data, and external risk data. The account basic data can include at least one of the opening bank, account type, account activity, and historical balance fluctuation. The account basic data can be determined by a business platform related to the insurance business platform, such as a bank system or a network system. The transaction flow data can include at least one of the transaction time, transaction amount, counterparty, payment channel, and transaction positioning. The transaction flow data can be determined by transaction flow indicated by a business platform related to the insurance business platform, such as at least one of bank flow, third-party payment records. The associated party graph can include at least one of the fund closed-loop path, multi-layer transfer relationship, and abnormal counterparty characteristics. The associated party graph can be determined by a graph database related to the insurance business platform. The behavior matching data can include at least one of the transaction frequency, transaction amount and income matching degree, and insured time correlation. The behavior matching data can be obtained by cross-verification of the insured person's insured behavior information such as tax information and social security data. The external risk data can include at least one of the anti-money laundering blacklist, account library involved in the case, and virtual currency address. The external risk data can be determined by at least one of a business platform related to the insurance business platform, such as the central bank anti-money laundering system and the block chain browser. Of course, the insured characteristic information and the feature extraction method corresponding to the insured characteristic information are not limited to this, and are not limited herein.

[0110] In the process of assessing the rationality of the insured person's fund flow according to the insured characteristic information, the risk assessment model can assess whether the insured person has fund flow abnormal account and other fund flow unreasonable behaviors, and further determine the rationality of the insured person's fund flow. In the case of the insured person's fund flow rationality being unreasonable, the influence coefficient of the insured characteristic information on the insured person's insurance black and gray production behavior is, for example, 0.3. Correspondingly, in the case of the insured person's insured behavior rationality being reasonable, the influence coefficient of the insured characteristic information on the insured behavior belonging to the insurance black and gray production behavior is, for example, -0.3. Of course, it is not limited to this, and is not limited herein.

[0111] In the case that the risk assessment model is used to evaluate at least one of the identity authenticity of the applicant, the rationality of the application behavior, and the rationality of the fund flow according to the application feature information, and an influence coefficient of the application feature information on the application behavior belonging to the insurance black and gray production behavior is obtained, the influence coefficient corresponding to the application feature information can be used for subsequent determination of the risk score of the application behavior belonging to the insurance black and gray production behavior, thereby facilitating the convenience of determining the risk score corresponding to the application behavior. The higher the risk score corresponding to the application behavior, the higher the possibility of the application behavior belonging to the insurance black and gray production behavior, and the risk score corresponding to the application behavior can be used for subsequent determination of the processing strategy for the application behavior.

[0112] In some embodiments, when the risk score is less than or equal to a first score threshold, the insurance black and gray production identification strategy for the application behavior is maintained as automatic monitoring; when the risk score is greater than the first score threshold and less than or equal to a second score threshold, the insurance black and gray production identification strategy for the application behavior is maintained as automatic monitoring, and the transaction behavior related to the application behavior is restricted; the second score threshold is greater than the first score threshold; when the risk score is greater than the second score threshold, the insurance black and gray production identification strategy for the application behavior is switched to manual review.

[0113] In a case where the risk score is less than or equal to a first score threshold, the insurance business platform can determine that the possibility of the insurance application behavior of the applicant belonging to the insurance black production behavior is low, and can maintain the insurance black production identification strategy for the insurance application behavior as automatic monitoring. In a case where the risk score threshold is greater than the first score threshold and less than or equal to a second score threshold, the insurance business platform can determine that the possibility of the insurance application behavior of the applicant belonging to the insurance black production behavior is medium, i.e., the insurance application behavior may or may not belong to the insurance black production behavior, and can maintain the insurance black production identification strategy for the insurance application behavior as automatic monitoring and limit the transaction behavior related to the insurance application behavior. Limiting the transaction behavior, for example, includes limiting at least one of the insurance premium and the insurance amount of the applicant. Of course, it is not limited thereto, and is not limited herein. By limiting the transaction behavior, the economic loss caused by the insurance application behavior turning into the insurance black production behavior can be reduced. In a case where the risk score threshold is greater than the second score threshold, the insurance business platform can determine that the possibility of the insurance application behavior of the applicant belonging to the insurance black production behavior is high, i.e., the insurance application behavior can be regarded as the insurance black production behavior, and can switch the insurance black production identification strategy for the insurance application behavior to manual review. For example, the insurance business platform can send an audit task corresponding to the insurance application behavior to an audit system corresponding to a staff of the insurance business platform to prompt the staff to manually review the insurance application behavior. Of course, it is not limited thereto, and is not limited herein. The first score threshold, for example, includes 0.5, and the second score threshold, for example, includes 0.8. The first score threshold and the second score threshold can be pre-set or set by the staff, and are not limited herein.

[0114] Of course, it is not limited thereto, and the insurance business platform can also determine the risk prompt strategy corresponding to the insurance application behavior according to the risk score in the process of processing the insurance application behavior according to the risk score corresponding to the insurance application behavior. For example, in a case where the risk score is less than or equal to the first score threshold, the insurance business platform can not output any prompt information. In a case where the risk score is greater than the first score threshold and less than or equal to the second score threshold, the insurance business platform can output warning information corresponding to the insurance application behavior to warn the insurance application behavior of the applicant suspected of being the insurance black production behavior and prompt the limitation of the transaction behavior of the applicant. In a case where the risk score is greater than the second score threshold, the insurance business platform can output manual review information corresponding to the insurance application behavior to inform the insurance application behavior of the applicant involving the insurance black production behavior and prompt the manual review of the insurance application behavior of the applicant. This is not limited herein.

[0115] Correspondingly, the insurance business platform can also record the processing log corresponding to the insurance application behavior in the process of processing the insurance application behavior according to the risk score corresponding to the insurance application behavior, for subsequent review of whether the insurance application behavior belongs to the insurance black and gray production behavior. Of course, it is not limited to this, and is not limited here.

[0116] Taking the applicant Li Si as an example, the applicant Li Si is 35 years old, and applies for a 2 million insurance amount of critical illness insurance. The insurance business platform can obtain or determine at least one of the social behavior information, the insurance application behavior information and the insurance application verification information of Li Si.

[0117] a) The social behavior information of Li Si includes:

[0118] - Social media 1: recently frequently publishes “continuous overtime to the early morning”, “cervical spondylosis recurrence” and the like;

[0119] - Social media 2: collects a large number of “thyroid nodule diet conditioning” notes.

[0120] b) The insurance application behavior information of Li Si includes:

[0121] - Insurance application record in the past 3 years: 5 insurance companies of critical illness insurance (cumulative insurance amount of 8 million);

[0122] - Claim record: none;

[0123] - Cancellation record: 3 short-term medical insurance policies were cancelled within 2 months.

[0124] c) The insurance application verification information of Li Si includes:

[0125] - Identity verification information:

[0126] -- Face comparison pass rate 92% (lower than 97% threshold);

[0127] -- Voiceprint matching fails (recording background has traces of others guiding);

[0128] - Health verification information:

[0129] -- Claims “no physical examination abnormalities”, but the medical database shows: 1, 2023-06-01, B ultrasound in a first-class hospital: thyroid nodule 4a; 2, 2023-08-15, tumor marker CA199 is elevated;

[0130] The insurance business platform can perform feature extraction processing on at least one of the social behavior information, the insurance application behavior information and the insurance application verification information of Li Si, to obtain insurance application feature information associated with the insurance application behavior of Li Si.

[0131] The insurance feature information associated with Li Si's insurance application behavior can include at least one of insurance feature information corresponding to social behavior information, insurance feature information corresponding to insurance application behavior information, and insurance feature information corresponding to insurance verification information. The insurance feature information corresponding to the social behavior information includes, for example, "overtime frequency", "disease keywords", and the like. The insurance feature information corresponding to the insurance application behavior information includes, for example, "short-term cancellation rate", "multiple insurance", and the like. The insurance feature information corresponding to the insurance verification information includes, for example, "face mismatch", "voiceprint mismatch", "health information mismatch", "nodule level", "tumor marker abnormality", and the like.

[0132] The insurance business platform can determine, by using the risk assessment model, a risk score of the insurance application behavior of Li Si according to the insurance feature information corresponding to Li Si. For example, the risk score of the insurance application behavior of Li Si output by the risk assessment model is 0.87.

[0133] The insurance business platform can process the insurance application behavior of Li Si according to the risk score corresponding to the insurance application behavior of Li Si.

[0134] For example, the insurance business platform can determine, according to a preset risk level division rule, a risk level to which the risk score corresponding to the insurance application behavior of Li Si belongs.

[0135] The risk level division rule includes, for example, when the risk score is less than or equal to a first score threshold, the risk level corresponding to the insurance application behavior is low; when the risk score is greater than the first score threshold and less than or equal to a second score threshold, the risk level corresponding to the insurance application behavior is medium; and when the risk score is greater than the second score threshold, the risk level corresponding to the insurance application behavior is high.

[0136] If the first score threshold is 0.5 and the second score threshold is 0.8, it can be determined that the risk level to which the risk score corresponding to the insurance application behavior of Li Si belongs is high, and it can be determined that the processing strategy for the insurance application behavior of Li Si is to switch the insurance gray production identification strategy of the insurance application behavior to manual review.

[0137] For example, the automatic processing flow of the insurance business platform for the insurance application behaviors of different insurance applicants includes obtaining the risk scores corresponding to the insurance application behaviors of different insurance applicants; and triggering the processing strategy for the insurance application behavior according to the risk score corresponding to the insurance application behavior. In this way, the insurance application behaviors of different insurance applicants are processed in batches.

[0138] It can be seen that in the above embodiments, for an entity under an insurance business, such as an insurance business platform, social behavior information and insurance behavior information of an applicant can be acquired; insurance verification processing is performed on the insurance behavior information of the applicant to obtain insurance verification information of the applicant; at least one of the social behavior information, the insurance behavior information, and the insurance verification information is subjected to feature extraction processing to obtain insurance feature information associated with the insurance behavior of the applicant; based on a risk assessment model, a risk score of the insurance behavior belonging to an insurance black and gray production behavior is determined according to the insurance feature information; and the insurance behavior is processed according to the risk score corresponding to the insurance behavior.

[0139] In the case where the insurance behavior information of the applicant is acquired, the insurance business platform can perform insurance verification processing on the insurance behavior information to obtain insurance verification information. Accordingly, the insurance business platform can comprehensively judge the possibility of the insurance behavior of the applicant belonging to an insurance black and gray production behavior based on the social behavior information, the insurance behavior information, and the insurance verification information of the applicant. For example, the insurance business platform can perform feature extraction processing on at least one of the social behavior information, the insurance behavior information, and the insurance verification information to obtain insurance feature information associated with the insurance behavior of the applicant, and utilize a risk assessment model to determine a risk score of the insurance behavior belonging to an insurance black and gray production behavior according to the insurance feature information. The higher the risk score is, the greater the possibility of the insurance behavior of the applicant belonging to an insurance black and gray production behavior is, and then the insurance business platform can process the insurance behavior according to the risk score corresponding to the insurance behavior. In the process of identifying whether the insurance behavior of the applicant belongs to an insurance black and gray production behavior, artificial auditing is not required, which is conducive to improving the monitoring convenience of the insurance behavior and further improving the monitoring convenience of the insurance black and gray production behavior. Accordingly, in the case where the insurance business platform can comprehensively identify whether the insurance behavior of the applicant belongs to an insurance black and gray production behavior based on different types of data such as the social behavior information, the insurance behavior information, and the insurance verification information of the applicant, it is conducive to improving the monitoring accuracy of the insurance behavior and further improving the monitoring accuracy of the insurance black and gray production behavior. Based on the improvement of the monitoring accuracy of the insurance black and gray production behavior, it is conducive to promoting the sustainable development of financial business and medical and health business.

[0140] In the process of determining the risk score of the insurance application behavior belonging to the insurance black production behavior according to the monitoring method of the insurance application behavior, and processing the insurance application behavior according to the risk score corresponding to the insurance application behavior, the processing strategy of the insurance application behavior can be adapted to the risk score corresponding to the insurance application behavior, which is beneficial to improve the monitoring convenience and accuracy of the insurance application behavior, and then prompt the monitoring convenience and accuracy of the insurance black production behavior. Accordingly, the acquisition of social behavior information and insurance application behavior information involved in the monitoring method of the insurance application behavior, the verification of the insurance application behavior information, the feature extraction of the social behavior information, the insurance application behavior information and the insurance verification information, the determination of the risk score of the insurance application behavior belonging to the insurance black production behavior and the determination of the insurance application behavior processing strategy can be realized by means of artificial intelligence (AI), which is beneficial to reduce the labor cost while improving the monitoring efficiency of the insurance application behavior.

[0141] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0142] In an embodiment, a monitoring device for insurance application behavior is provided, which corresponds to the monitoring method for insurance application behavior in the above embodiment. As shown in the figure, the monitoring device for insurance application behavior includes a data acquisition module 110, a data verification module 120, a feature extraction module 130, a risk assessment module 140 and an insurance application behavior processing module 150. The functions of each module are described in detail as follows: Figure 3

[0143] The data acquisition module 110 is configured to acquire the social behavior information and the insurance application behavior information of the insurance applicant.

[0144] The data verification module 120 is configured to perform insurance verification processing on the insurance application behavior information of the insurance applicant to obtain insurance verification information of the insurance applicant.

[0145] The feature extraction module 130 is configured to perform feature extraction processing on at least one of the social behavior information, the insurance application behavior information and the insurance verification information to obtain insurance feature information associated with the insurance application behavior of the insurance applicant.

[0146] The risk assessment module 140 is configured to determine, based on a risk assessment model, a risk score of the insurance application behavior belonging to the insurance black production behavior according to the insurance feature information.

[0147] The insurance application behavior processing module 150 is configured to process the insurance application behavior according to the risk score corresponding to the insurance application behavior.

[0148] In an embodiment, the data acquisition module 110 is configured to:​

[0149] According to the insurance behavior information, obtain social media information associated with the applicant;

[0150] According to the social media information, determine the social behavior information of the applicant.

[0151] In an embodiment, the data acquisition module 110 is configured to:

[0152] perform information analysis processing on the social media information to obtain at least one of the living habit, the social relationship, and the consumption habit of the applicant;

[0153] According to the at least one of the living habit, the social relationship, and the consumption habit of the applicant, determine the social behavior information of the applicant.

[0154] In an embodiment, the insurance behavior information at least includes at least one of the identity information and the health information of the applicant; the data verification module 120 is configured to:

[0155] According to the matching result between the preset identity information of the applicant and the identity information of the applicant, verify the identity information of the applicant to obtain the identity verification information of the applicant;

[0156] According to the matching result between the preset medical information of the applicant and the health information of the applicant, verify the health information of the applicant to obtain the health verification information of the applicant.

[0157] In an embodiment, the risk assessment module 140 is configured to:

[0158] input the insurance feature information into the risk assessment model, so that the risk assessment model assesses the influence coefficient of the insurance feature information on the insurance black and gray production behavior of the insurance behavior according to the feature information;

[0159] According to the sum of the respective influence coefficients of all the insurance feature information, determine the risk score of the insurance behavior of the applicant belonging to the insurance black and gray production behavior.

[0160] In an embodiment, the risk assessment module 140 is configured to:

[0161] According to the insurance feature information, the risk assessment model is configured to assess at least one of the identity authenticity, the insurance behavior rationality, and the fund flow rationality of the applicant to obtain the influence coefficient of the insurance feature information on the insurance black and gray production behavior of the insurance behavior.

[0162] In an embodiment, the insurance behavior processing module 150 is configured to:

[0163] when the risk score is less than or equal to a first score threshold, maintaining an insurance black production identification strategy for the insurance application behavior as automatic monitoring;

[0164] when the risk score is greater than the first score threshold and less than or equal to a second score threshold, maintaining the insurance black production identification strategy for the insurance application behavior as automatic monitoring, and limiting transaction behavior related to the insurance application behavior; the second score threshold is greater than the first score threshold;

[0165] when the risk score is greater than the second score threshold, switching the insurance black production identification strategy for the insurance application behavior to manual review.

[0166] The application provides a monitoring device for an insurance application behavior, which obtains social behavior information of an insurance applicant and insurance application behavior information; performs insurance application verification processing on the insurance application behavior information of the insurance applicant to obtain insurance application verification information of the insurance applicant; performs feature extraction processing on at least one of the social behavior information, the insurance application behavior information, and the insurance application verification information to obtain insurance application feature information associated with the insurance application behavior of the insurance applicant; determines, based on a risk assessment model, a risk score of the insurance application behavior belonging to an insurance black production behavior according to the insurance application feature information; and processes the insurance application behavior according to the risk score corresponding to the insurance application behavior.

[0167] The monitoring device for an insurance application behavior can determine, by a computer device, a risk score of an insurance application behavior of an insurance applicant belonging to an insurance black production behavior, and process the insurance application behavior according to the risk score corresponding to the insurance application behavior, so that the processing strategies of different insurance application behaviors can be adapted to the risk scores corresponding to the insurance application behaviors, thereby improving the monitoring convenience and accuracy of the insurance application behavior, and improving the monitoring convenience and accuracy of the insurance black production behavior, and improving the processing convenience and accuracy of the insurance application behavior, and improving the processing convenience and accuracy of the insurance black production behavior. Based on the improvement of the monitoring convenience and accuracy of the insurance application behavior and the monitoring convenience and accuracy of the insurance black production behavior, the sustainable development of financial business and medical and health business is promoted.

[0168] For example, the monitoring device for the insurance application behavior can be used in different business scenarios to monitor and process whether the insurance application behavior in different business scenarios belongs to the insurance black production behavior. For example, it can be used in a financial business scenario to evaluate whether the insurance application behavior related to one of the insurance policies involved in the financial business scenario, such as one of the enterprise property insurance, home property insurance, network risk insurance, pension insurance, and the like, belongs to the insurance black production behavior. For another example, it can be used in a medical health business scenario to evaluate whether the insurance application behavior related to one of the insurance policies involved in the medical health business scenario, such as one of the medical insurance, major disease insurance, disability income loss insurance, long-term care insurance, accidental injury insurance, and group health insurance, belongs to the insurance black production behavior. Accordingly, the processing strategy for the foregoing insurance application behavior can also be determined. Based on this, the monitoring device for the insurance application behavior can improve the monitoring convenience and accuracy of the insurance application behavior involved in the financial business scenario or the medical health business scenario, thereby improving the monitoring convenience and accuracy of the insurance black production behavior, and thereby facilitating the sustainable development of the financial business and the medical health business.

[0169] The specific limitations of the monitoring device for the insurance application behavior can be referred to the limitations of the monitoring method for the insurance application behavior in the foregoing, which will not be described herein. Each module in the monitoring device for the insurance application behavior can be realized by software, hardware, and a combination thereof in whole or in part. Each module can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0170] In one embodiment, a computer device is provided, which can be a server, and an internal structure diagram thereof can be as shown in Figure 4 The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client through a network connection. The computer program is executed by the processor to implement the functions or steps of the server side of the monitoring method for the insurance application behavior.

[0171] In one embodiment, a computer device is provided, which can be a client, and an internal structure diagram thereof can be as shown in Figure 5As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The network interface of the computer device is used to communicate with the external server through the network connection. The computer program is executed by the processor to realize the function or step of the client side of the monitoring method of the insurance application behavior.

[0172] In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the computer program to implement the following steps:

[0173] Obtain the social behavior information and the insurance application behavior information of the insurance applicant;

[0174] Perform insurance verification processing on the insurance application behavior information of the insurance applicant to obtain insurance verification information of the insurance applicant;

[0175] Perform feature extraction processing on at least one of the social behavior information, the insurance application behavior information, and the insurance verification information to obtain insurance feature information associated with the insurance application behavior of the insurance applicant;

[0176] Based on a risk assessment model, determine a risk score of the insurance application behavior belonging to an insurance black and gray production behavior according to the insurance feature information;

[0177] Process the insurance application behavior according to the risk score corresponding to the insurance application behavior.

[0178] In one embodiment, a computer readable storage medium is provided, having a computer program stored thereon, the computer program being executed by a processor to implement the following steps:

[0179] Obtain the social behavior information and the insurance application behavior information of the insurance applicant;

[0180] Perform insurance verification processing on the insurance application behavior information of the insurance applicant to obtain insurance verification information of the insurance applicant;

[0181] Perform feature extraction processing on at least one of the social behavior information, the insurance application behavior information, and the insurance verification information to obtain insurance feature information associated with the insurance application behavior of the insurance applicant;

[0182] Based on a risk assessment model, determine a risk score of the insurance application behavior belonging to an insurance black and gray production behavior according to the insurance feature information;

[0183] According to the risk score corresponding to the insurance application behavior, processing the insurance application behavior.

[0184] It should be noted that the functions or steps described above with respect to the computer-readable storage medium or the computer device can correspond to the relevant descriptions of the server side and the client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0185] A person of ordinary skill in the art can understand that all or part of the processes in the foregoing method embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, the processes of the foregoing embodiments of the method can be included. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0186] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0187] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for monitoring insurance purchase behavior, characterized in that, include: To obtain information about the policyholder's social behavior and insurance application behavior; The policyholder's insurance application behavior information is processed for insurance application verification to obtain the policyholder's insurance application verification information; At least one of the social behavior information, the insurance purchase behavior information, and the insurance purchase verification information is subjected to feature extraction processing to obtain insurance purchase feature information associated with the policyholder's insurance purchase behavior; Based on the risk assessment model and the insurance characteristics information, a risk score is determined to indicate that the insurance purchase behavior belongs to the black and gray market activities of insurance. The insurance purchase behavior is processed based on the risk score corresponding to the insurance purchase behavior.

2. The monitoring method according to claim 1, characterized in that, The acquisition of the policyholder's social behavior information includes: Based on the insurance purchase information, obtain the social media information associated with the policyholder; Based on the social media information, the social behavior information of the policyholder is determined.

3. The monitoring method according to claim 2, characterized in that, The step of determining the policyholder's social behavior information based on the social media information includes: The social media information is analyzed and processed to obtain at least one of the policyholder's lifestyle habits, social relationships, and consumption habits; The social behavior information of the insured is determined based on at least one of the insured's lifestyle habits, social relationships, and consumption habits.

4. The monitoring method according to claim 1, characterized in that, The insurance application information includes at least one of the following: the policyholder's identity information and health information; The process of verifying the policyholder's insurance application behavior information to obtain the policyholder's insurance application verification information includes at least one of the following: Based on the matching result between the policyholder's preset identity information and the policyholder's identity information, the policyholder's identity information is verified to obtain the policyholder's identity verification information; Based on the matching results between the policyholder's preset medical information and the policyholder's health information, the policyholder's health information is verified to obtain the policyholder's health verification information.

5. The monitoring method according to any one of claims 1 to 4, characterized in that, The risk assessment model, based on the insurance purchase characteristic information, determines the risk score of whether the insurance purchase behavior belongs to the black and gray market of insurance, including: The insurance application feature information is input into the risk assessment model so that the risk assessment model can evaluate the impact coefficient of the insurance application feature information on whether the insurance application behavior belongs to the black and gray market of insurance based on the feature information; Based on the sum of the influence coefficients corresponding to each of the aforementioned insurance characteristics, a risk score is determined to indicate whether the insured's insurance purchase behavior falls under the category of black and gray insurance activities.

6. The monitoring method according to claim 5, characterized in that, The step of inputting the insurance application characteristic information into the risk assessment model, so that the risk assessment model can evaluate the impact coefficient of the insurance application characteristic information on whether the insurance application behavior belongs to the black and gray market of insurance, includes: The risk assessment model evaluates at least one of the following based on the insurance characteristics: the authenticity of the policyholder's identity, the rationality of the insurance behavior, and the rationality of the fund flow. This yields the influence coefficient of the insurance characteristics on whether the insurance behavior constitutes a black or gray market activity in the insurance industry.

7. The monitoring method according to any one of claims 1 to 4, characterized in that, The process of processing the insurance purchase behavior based on the risk score corresponding to the insurance purchase behavior includes: When the risk score is less than or equal to the first score threshold, the insurance black and gray market identification strategy for the insurance purchase behavior remains in automated monitoring. When the risk score is greater than the first scoring threshold and less than or equal to the second scoring threshold, the insurance black and gray market identification strategy for the insurance purchase behavior remains in automated monitoring, and transactions related to the insurance purchase behavior are restricted; the second scoring threshold is greater than the first scoring threshold. When the risk scoring threshold is greater than the second scoring threshold, the insurance black and gray market identification strategy for the insurance purchase behavior will be switched to manual review.

8. A monitoring device for insurance purchase behavior, characterized in that, The monitoring device includes: The data acquisition module is used to acquire the policyholder's social behavior information and insurance application behavior information; The data verification module is used to process the policyholder's insurance application behavior information to obtain the policyholder's insurance application verification information; The feature extraction module is used to perform feature extraction processing on at least one of the social behavior information, the insurance purchase behavior information and the insurance purchase verification information to obtain insurance purchase feature information associated with the policyholder's insurance purchase behavior; The risk assessment module is used to determine the risk score of the insurance purchase behavior as belonging to the black and gray market activities of insurance based on the risk assessment model and the insurance purchase characteristic information. The insurance application behavior processing module is used to process the insurance application behavior based on the risk score corresponding to the insurance application behavior.

9. A computer device, characterized in that, The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the monitoring method for insurance purchase behavior as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the monitoring method for insurance purchase behavior as described in any one of claims 1 to 7.

Images