Method, device, apparatus and computer-readable medium for verifying insurance policies

Abstract

The present invention discloses a method, device, equipment and computer-readable medium for verifying insurance policies, and relates to the field of computer technology. A specific implementation of the method includes: extracting disease combinations from the historical insurance policy claims data of the user to be verified, and establishing a disease impact matrix; determining the influencing diseases of the user to be verified according to the disease impact matrix; extracting user feature vectors and the historical disease vectors of the user to be verified from the historical insurance policy claims data, and constructing the claims vector of the user to be verified; based on the claims vector of the user to be verified, determining similar users to the user to be verified, so as to use the average claims time interval of the similar users as the claims time of the user to be verified; verifying the insurance policy of the user to be verified according to the influencing diseases and the claims time of the user to be verified. This implementation can reduce the error rate of underwriting insurance policies.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and computer-readable medium for verifying an insurance policy. Background Art

[0002] The process of verifying the insurance policy involves predicting the diseases that the user may develop in the future to determine whether the insurance is successful. At present, the risk of a certain disease is manually judged based on the user's physical examination indicators, hospital medical records and other information.

[0003] In the process of realizing the present invention, the inventors found that there are at least the following problems in the prior art: due to the poor accuracy of predicting disease risks, the error rate in underwriting insurance policies is high. Summary of the invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, device and computer-readable medium for verifying insurance policies, which can reduce the error rate of underwriting insurance policies.

[0005] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for verifying an insurance policy is provided, comprising:

[0006] Extract disease combinations from historical insurance policy claims data of the user to be verified and establish a disease impact matrix;

[0007] Determining the influencing disease of the user to be verified according to the disease impact matrix;

[0008] Extracting a user feature vector and a historical disease vector of the user to be verified from the historical insurance policy claim data, and constructing a claim vector of the user to be verified;

[0009] Based on the claim settlement vector of the user to be verified, determine users similar to the user to be verified, and use the average claim settlement time interval of the similar users as the claim settlement time of the user to be verified;

[0010] According to the influencing disease and the claim settlement time of the user to be verified, the insurance policy of the user to be verified is verified.

[0011] The extracting of disease combinations from the historical insurance policy claims data of the user to be verified and establishing a disease impact matrix includes:

[0012] Extract disease combinations from the historical insurance claims data of the user to be verified, and determine the vector of each disease based on the International Classification of Diseases standard;

[0013] A weight vector for each disease is constructed based on the vector of each disease, and the weight vector for each disease is arranged in rows to obtain the disease impact matrix.

[0014] Determining the influencing disease of the user to be verified according to the disease impact matrix includes:

[0015] After the historical disease vector of the user to be checked is multiplied by the disease influence matrix, the difference between the historical disease vector and the disease influence matrix is ​​the influence matrix, and the influencing disease of the user to be checked is determined according to the element values ​​in the influence matrix.

[0016] The determining, based on the claim settlement vector of the user to be verified, a user similar to the user to be verified comprises:

[0017] Based on the user feature vector and historical disease vector in the claim vector of the user to be verified, and the user feature vector and historical disease vector in the matching vector of other users, the vector distance between the user to be verified and the other users is calculated to determine similar users to the user to be verified among the other users.

[0018] The determining of users similar to the user to be verified from among the other users includes:

[0019] In the ascending order of the vector distance, a preset number of similar users or a preset proportion of users are selected from the other users as similar users to the user to be checked.

[0020] According to the influencing disease and the claim settlement time of the user to be verified, verifying the insurance policy of the user to be verified includes:

[0021] According to the influencing disease and the claim settlement time of the user to be verified, check whether the insurance policy of the user to be verified is abnormal.

[0022] According to the influencing disease and the claim settlement time of the user to be verified, verifying the insurance policy of the user to be verified includes:

[0023] According to the influencing disease and the claim settlement time of the user to be verified, the insurance policy of the user to be verified is checked to determine whether a reminder message needs to be sent.

[0024] According to a second aspect of an embodiment of the present invention, there is provided a device for verifying an insurance policy, comprising:

[0025] The extraction module is used to extract the disease combinations in the historical insurance policy claims data of the user to be verified and establish the disease impact matrix;

[0026] A disease module, used for determining the influencing disease of the user to be verified according to the disease impact matrix;

[0027] A construction module, used to extract the user feature vector and the historical disease vector of the user to be verified from the historical insurance policy claim data, and construct a claim vector of the user to be verified;

[0028] A similarity module, configured to determine similar users to the user to be verified based on the claim settlement vector of the user to be verified, so as to use the average claim settlement time interval of the similar users as the claim settlement time of the user to be verified;

[0029] The verification module is used to verify the insurance policy of the user to be verified according to the influencing disease and the claim settlement time of the user to be verified.

[0030] According to a third aspect of an embodiment of the present invention, there is provided an electronic device for checking an insurance policy, comprising:

[0031] one or more processors;

[0032] a storage device for storing one or more programs,

[0033] When the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0034] According to a fourth aspect of an embodiment of the present invention, there is provided a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the method as described above is implemented.

[0035] One embodiment of the above invention has the following advantages or beneficial effects: extracting disease combinations from the historical insurance policy claims data of the user to be verified, and establishing a disease impact matrix; determining the influencing diseases of the user to be verified based on the disease impact matrix; extracting user feature vectors and the historical disease vectors of the user to be verified from the historical insurance policy claims data, and constructing the claims vector of the user to be verified; based on the claims vector of the user to be verified, determining similar users to the user to be verified, so as to use the average claims time interval of the similar users as the claims time of the user to be verified; verifying the insurance policy of the user to be verified according to the influencing diseases and the claims time of the user to be verified. On the one hand, the influencing diseases are determined based on the historical insurance policy claims data; on the other hand, the claims time is determined from the perspective of similar users. The accuracy of verification can be improved according to the influencing diseases and the claims time, thereby reducing the error rate of verifying insurance policies.

[0036] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with specific implementation examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0038] Figure 1 is a schematic diagram of the main process of the method for verifying an insurance policy according to an embodiment of the present invention;

[0039] Figure 2 is a schematic diagram of a process of establishing a disease impact matrix according to an embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of obtaining a weight vector according to an embodiment of the present invention;

[0041] Figure 4 is a schematic diagram of obtaining a disease impact matrix according to an embodiment of the present invention;

[0042] Figure 5 is a schematic diagram of an interface for sending reminder information to a user according to an embodiment of the present invention;

[0043] Figure 6 is a schematic diagram of an interface for checking policy anomalies according to an embodiment of the present invention;

[0044] Figure 7 is a schematic diagram of the main structure of a device for verifying an insurance policy according to an embodiment of the present invention;

[0045] Figure 8 is a schematic diagram of an application of a device for verifying an insurance policy according to an embodiment of the present invention;

[0046] Fig. 9 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0047] Fig.10 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0049] At present, the process of verifying insurance policies involves predicting diseases that users may develop in the future. It is mainly based on the user's physical examination indicators, hospital medical records and other information, and a data model is established for a single disease to predict the risk of a certain disease.

[0050] The technology that uses physical examination and medical record data as the basis for prediction has high requirements for the integrity of data acquisition and is difficult to obtain structured data. It is difficult to give a high-accuracy prediction result based on the quality of existing data. Moreover, different models need to be trained for different diseases, which is a large workload. The prediction results of different diseases cannot be integrated, and the application process is complicated. More importantly, the result of the model prediction is only the probability of a certain disease, and there is no prompt for the key time node of the onset, which has certain limitations on the reference value for the decision-making of verifying insurance policies. Therefore, the error rate of underwriting insurance policies is relatively high.

[0051] In order to reduce the error rate in underwriting insurance policies, the following technical solutions in the embodiments of the present invention may be adopted.

[0052] See also Figure 1 , Figure 1 Schematic diagram of the main process of the method for verifying insurance policies according to an embodiment of the present invention, using the disease impact matrix to determine multiple influencing diseases at one time, and then verifying the insurance policies. Figure 1 As shown, the specific steps include:

[0053] S101. Extract disease combinations from historical insurance policy claims data of the user to be verified and establish a disease impact matrix.

[0054] In the embodiment of the present invention, it can be implemented in one or more devices such as a mobile terminal, a personal computer or a server. For example, an APP is set in the mobile terminal, and the result of the insurance policy verification can be directly obtained through the client of the APP. Of course, the client can also be set in a personal computer and a server.

[0055] As an example, health management service personnel can obtain the verification results of the user's insurance policy through the client, helping them to complete the formulation of health management plans and the implementation of health management services. The verification results of insurance policies are embedded in the automatic claims rules of the claims system to support intelligent claims decision-making.

[0056] In the embodiment of the present invention, the user whose insurance policy is to be checked is called the user to be checked, that is, the user to be checked is the user whose insurance policy is to be checked. As an example, the user to be checked needs to buy critical illness insurance. In order to avoid unnecessary losses, the insurance staff needs to check the insurance policy for the user to be checked.

[0057] Specifically, extract the disease combinations in the historical insurance policy claims data of the user to be verified and establish a disease impact matrix. For the user to be verified, it is necessary to obtain the historical insurance policy claims data from the data server, and then establish a disease impact matrix based on the disease combinations in the historical insurance policy claims data. The disease impact matrix is ​​used to characterize the impact of the user's existing diseases on the user. Due to the different disease combinations in the historical insurance policy claims data, the disease impact matrix of each user to be verified is different.

[0058] See also Figure 2 , Figure 2 1 is a flow chart of establishing a disease impact matrix according to an embodiment of the present invention, which specifically includes the following steps:

[0059] S201. Extract disease combinations from historical insurance policy claims data of the user to be verified, and determine the vector of each disease based on the International Classification of Diseases standard.

[0060] The disease combination data of the user to be verified is extracted from the historical claims database. Based on the International Classification of Diseases (ICD), each disease combination D[d] forms a corresponding disease vector V, and

[0061]

[0062] The dimension n of vector V is equal to the number of diseases classified by the International Classification of Diseases.

[0063] In other words, the disease vector is a parameter obtained based on the International Classification of Diseases standard. The International Classification of Diseases standard can accurately describe the disease and will not be affected by the personal influence of the verifier, so the disease vector can fully represent the disease of the user to be verified.

[0064] S202. Construct a weight vector for each disease based on the vector of each disease, and arrange the weight vector of each disease in rows to obtain a disease impact matrix.

[0065] When calculating the weight vector of disease d, it is necessary to select the vector V with V[d] = 1 from the disease vector set {V} d The vector set {V d}.

[0066] As an example, there are n diseases in the ICD standard. For a disease combination [Disease 1, Disease 2, Disease 4] in historical claims data, the corresponding disease vector V is (1, 1, 0, 1, 0, ...), where V is n-dimensional.

[0067] Set {V d}, then the weight vector W corresponding to disease d can be d It is expressed as:

[0068] W d =∑V d / numd formula 2

[0069] See also Figure 3 , Figure 3 is a schematic diagram of obtaining a weight vector according to an embodiment of the present invention. Figure 3The left image in the middle is the n-dimensional vector V, and the middle image is the vector set of disease d {V d}, the right figure is the weight vector corresponding to disease d.

[0070] As an example, when calculating the weight of disease 1, it is necessary to select the vectors with the value of 1 in the first dimension from the vector V of the historical insurance claim data of the user to be verified to form a vector set. The number of vectors in the vector set is the number of all vectors that meet the conditions in the historical insurance claim data. Then, based on the vectors with the value of 1 in all dimensions, {V d}, and finally the weight vector of each disease is obtained.

[0071] That is to say, the weight vector of each disease is obtained based on the vector of each disease. As an example, in the historical insurance claim data, the weight vector of the vector with 0 elements in the disease {V} is 0; the weight vector corresponding to the vector with 1 elements in the disease {V} in the historical insurance claim data is the ratio of the sum of the 1 elements of the disease to numd. For example, the sum of the 1 elements of disease 3 is 3, numd is 10, and the weight vector of disease 3 is 0.3.

[0072] The weight vectors of each disease are arranged in rows and integrated to obtain the disease impact matrix.

[0073] See also Figure 4 , Figure 4 is a schematic diagram of obtaining a disease impact matrix according to an embodiment of the present invention. The left figure is a weight vector W of an n-dimensional disease d , the right picture is the integration W d Get the disease impact matrix T. T is an n*n dimensional matrix.

[0074] exist Figure 2 In an embodiment, various diseases in historical insurance policy claim data are used to represent the various existing diseases of the user to be verified using a matrix, and then a disease impact matrix is ​​used to represent the various existing diseases of the user to be verified.

[0075] S102. Determine the influencing diseases of the user to be verified according to the disease impact matrix.

[0076] The disease impact matrix can reflect the impact of existing diseases. Then, the influencing diseases of the user to be verified are determined according to the disease impact matrix and the historical disease vector of the user to be verified. The historical disease vector of the user to be verified is used to represent the existing diseases of the user to be verified.

[0077] Specifically, the historical disease vector V of the user to be verified c, is a vector obtained by sorting the number of diseases n in the ICD standard. If the user to be verified suffers from the disease, the corresponding element of the disease is 1; if the user does not suffer from the disease, the corresponding element of the disease is 0. Secondly, the historical disease vector of the user to be verified is multiplied by the disease impact matrix to obtain V c * . V c * It represents the risk of contracting the corresponding disease in the future. The greater the risk value, the greater the possibility of contracting the disease.

[0078] ΔR=V c * -V c =V c TV c Formula 3

[0079] The element values ​​in ΔR directly reflect the tendency of disease evolution and the level of risk. That is, after the historical disease vector of the user to be verified is multiplied by the disease impact matrix, the difference between the historical disease vector and the disease impact matrix is ​​the impact matrix ΔR, and the element values ​​in the impact matrix determine the impact disease of the user to be verified.

[0080] As an example, for the user c to be verified, firstly organize the historical diseases of the user to be verified into V c Then, the disease impact matrix of the user to be verified is compared with V c Multiply to get V c * . And calculate V c * With V c The difference is ΔR.

[0081] It should be noted that the disease impact matrix is ​​an n*n matrix, and the i-th column represents the impact weight of disease i on other diseases, that is, the contribution rate. For the user to be verified, the historical disease vector is 1*n, the element corresponding to the disease has been suffered is 1, and the elements corresponding to other diseases that have not been suffered are 0.

[0082] The historical disease vector of the user to be verified is multiplied by the disease impact matrix to obtain a vector of 1*n, which is then subtracted from the disease impact matrix to obtain the impact matrix ΔR. In the impact matrix, the element values ​​corresponding to other disease dimensions will change from 0 to new risk values. A risk value that is not 0 indicates that the corresponding disease may be contracted in the future. The larger the risk value, the greater the possibility of contracting the disease.

[0083] S103: extracting user feature vectors and historical disease vectors of the user to be verified from historical insurance policy claims data, and constructing a claims vector for the user to be verified.

[0084] Generally speaking, it takes some time for a disease to develop. If the disease occurs within a period of time after purchasing the policy, the claim will be paid according to the policy. After obtaining the disease impact matrix of the user to be verified, it is necessary to construct the claim vector of the user to be verified to obtain the claim time of the policy.

[0085] Specifically, the user feature vector and the historical disease vector of the user to be verified are extracted from the historical insurance claim data to construct the claim vector of the user to be verified. That is, the user feature vector F is constructed based on the user features obtained from the database, and then combined with the historical disease vector V of the user to be verified c , establish the claim vector P(F, V c ). As an example, the user feature vector includes user parameters of multiple dimensions, such as age, gender, height, and weight.

[0086] S104. Based on the claim vector of the user to be verified, determine users similar to the user to be verified, and use the average claim time interval of the similar users as the claim time of the user to be verified.

[0087] The claim time of the user to be checked can be determined by similar users to the user to be checked. Specifically, the user to be checked is represented by a claim vector, and similarly, other users can also be represented by a matching vector. Different from the claim vector features of the user to be checked, the matching vector includes the user feature vector, the historical disease vector, and the claim time. That is: the matching vector Q(F,V c ,KT), KT represents the claim settlement time, and KT is the number of days between the current claim settlement and the next claim settlement. If there is only one claim settlement, KT is infinite.

[0088] Based on the claim vector of the user to be verified, similar users to the user to be verified are determined, and the average claim time interval of similar users is used as the claim time of the user to be verified. In other words, similar users are used as a bridge, and the average claim time interval of similar users is used as the claim time of the user to be verified.

[0089] In one embodiment of the present invention, based on the vector distance between the claim vector of the user to be verified and the matching vectors of other users, similar users to the user to be verified are determined. c , and the matching vectors F and V c The vector distance between them is used to determine the similar users to the user to be verified. Accordingly, the smaller the vector distance is, the greater the similarity between the user to be verified and other users is. As an example, the user corresponding to the minimum vector distance is selected as the similar user to the user to be verified.

[0090] That is, based on the user feature vector and historical disease vector in the claim vector of the user to be verified, and the user feature vector and historical disease vector in the matching vector of other users, the vector distance between the user to be verified and other users is calculated to determine users similar to the user to be verified among other users.

[0091] In one embodiment of the present invention, in order to improve the accuracy of similar users, similar users to the user to be verified can be determined according to a preset similarity number and vector distance. That is, in the order of vector distance from small to large, a preset number of similar users or a preset proportion of users are selected from other users as similar users to the user to be verified.

[0092] As an example, 20 users are determined in the order of vector distance from small to large. The preset similarity number is 10, and the first 10 users can be used as similar users of the user to be verified. The average claim settlement time of the above 10 similar users is used as the claim settlement time of the user to be verified. As another example, users with the top 10% of vector distances are used as similar users. The average claim settlement time of the above similar users is used as the claim settlement time of the user to be verified.

[0093] S105. Check the insurance policies of the users to be checked according to the claim settlement time of the affected diseases and the users to be checked.

[0094] After knowing the influencing disease and claim time of the user to be verified, you can start to verify the insurance policy. As an example, according to the influencing disease, query the claim time threshold. If the claim time threshold is less than the claim time of the user to be verified, it is judged that the insurance policy of the user to be verified has not passed the verification; if the claim time threshold is greater than or equal to the claim time of the user to be verified, it is judged that the insurance policy of the user to be verified has passed the verification.

[0095] In one embodiment of the present invention, the insurance policy of the user to be checked is checked to see if it is abnormal according to the influencing disease and the claim settlement time of the user to be checked. In other words, it is possible to determine if the insurance policy of the user to be checked is abnormal from the two perspectives of the influencing disease and the claim settlement time. If an abnormality is determined, a reminder can be provided in the form of text annotation, color identification, and prompt information.

[0096] In one embodiment of the present invention, the insurance policy of the user to be checked is checked according to the influencing disease and the claim settlement time of the user to be checked to determine whether a reminder message needs to be sent. In the process of checking the insurance policy of the user to be checked, if the influencing disease is likely to occur at the claim settlement time of the user to be checked, then it can be determined whether to send a reminder message to the user to be checked based on the checking result.

[0097] As an example, if the verification result is passed, it means that the disease has little impact on the user, and there is no need to send a reminder message; if the verification result is failed, it means that the disease has a great impact on the user. In order to remind the user to seek medical treatment in time and change life habits, a reminder message can be sent to the user.

[0098] In the above embodiment, disease combinations are extracted from the historical insurance policy claims data of the user to be verified, and a disease impact matrix is ​​established; based on the disease impact matrix, the influencing diseases of the user to be verified are determined; the user feature vector and the historical disease vector of the user to be verified are extracted from the historical insurance policy claims data to construct the claim vector of the user to be verified; based on the claim vector of the user to be verified, similar users to the user to be verified are determined, so as to use the average claim time interval of the similar users as the claim time of the user to be verified; the insurance policy of the user to be verified is verified according to the influencing diseases and the claim time of the user to be verified. On the one hand, the influencing diseases are determined based on the historical insurance policy claims data; on the other hand, the claim time is determined from the perspective of similar users. The accuracy of the verification can be improved according to the influencing diseases and the claim time, thereby reducing the error rate of verifying the insurance policy.

[0099] The technical solutions in the embodiments of the present invention are exemplarily described below with reference to specific examples.

[0100] In the database of historical insurance claims data, obtain the user information and pre-existing disease information of user A. User A's user information {age, gender, occupation, region...}, pre-existing disease information [disease 1].

[0101] User A’s historical disease vector is V A =(1,0,0,0,…).

[0102] User feature vector F of user A A =(49,1,0,0,1,0,…)

[0103] Among them, the first column is [age-numeric variable-49], the second column is [sex female-01 variable-yes], the third column is [sex male-01 variable-no], etc.

[0104] User A’s claim vector B A (V A , F A )=(49,1,0,0,1,0,…,1,0,0,0,…).

[0105] According to S101 and S102, the influencing diseases of user A are determined, and the disease influencing matrix of user A is:

[0106]

[0107] ΔR A =V A * -V A =V A TV A =(1,0,0.5,0,…)-(1,0,0,0,…)=(0,0,0.5,0,…)

[0108] The value of the third column is 0.5, and the values ​​of the other columns are 0, indicating that disease 3 has the highest risk. A Similar users of user A are identified, and it is learned that the claim settlement period is 108 days.

[0109] See Table 1, which is a table showing matching vectors of similar users of user A, vector examples with user A, and a schematic diagram of claim settlement time intervals.

[0110] Table 1

[0111]

[0112] Based on the impact disease: high-risk disease 3, claim time: 108 days, check the insurance policy of user A. In addition, you can also set the corresponding health management plan, such as the dietary taboos corresponding to disease 3, the health indicators that need to be paid attention to, etc., and refer to the key time prediction results to determine the health management effect feedback time and monitoring plan. The interface displayed to the user on the front end is as follows: Figure 5 shown.

[0113] See also Figure 5 , Figure 5 It is a schematic diagram of an interface for sending reminder information to a user according to an embodiment of the present invention. Figure 5 Included are diet plans, exercise plans, health checks and other tips for disease 3.

[0114] The technical solution in the embodiment of the present invention is exemplified below with reference to another specific example.

[0115] User B reports the claim after the accident at time t, and implements the technical solution of the embodiment of the present invention in combination with the historical insurance policy claim data, and learns that the affected disease is: disease 3, and the claim time is 108 days.

[0116] At the same time, the disease combination provided by customer B in this report is retrieved, and the time interval between this claim and the most recent claim in history is calculated. The disease combination this time is disease 2 and disease 4, and the claim settlement time is 40 days.

[0117] By checking the policy analysis data: Disease 3, the claim time is 108 days, and the accident data: Disease 2 and Disease 4, the claim time is 40 days, it is preliminarily determined that user B's report is abnormal. The claim system sends a warning message of abnormal risk of customer claim disease to the claim personnel.

[0118] See also Figure 6 , Figure 6 It is a schematic diagram of an interface for checking policy anomalies according to an embodiment of the present invention. Figure 6 The report includes the case reports of three users. User B is judged to be abnormal after the above analysis. The prompt message is marked: abnormal disease risk, please pay attention.

[0119] The embodiments of the present invention deeply explore the value of historical policy claims data and establish a disease impact matrix and similarity calculation. It allows users with a history of medical conditions to predict the evolution of the disease at key time points in the future, which is of great help in strengthening the health management services of insurance companies and the implementation of smart claims strategies.

[0120] In addition, the dimensional construction and updating of the disease impact matrix are consistent with the types of diseases in the International Classification of Diseases (ICD) developed by the WHO, which can directly form standards and can also be exported as technical capabilities.

[0121] From a business perspective, it is applied to customer health management services and smart claims. (1) By predicting the future changes in customers' diseases, it helps formulate customer health management plans, improves the company's health management platform functions, and enhances customer experience; (2) When making claims, it embeds automatic claims rules, compares the predicted results with the actual reports of customers, reduces claims costs, realizes smart claims, and promotes the development of the smart two-core strategy.

[0122] See also Figure 7 , Figure 7 is a schematic diagram of the main structure of an apparatus for verifying an insurance policy according to an embodiment of the present invention. The apparatus for verifying an insurance policy can implement a method for verifying an insurance policy, such as Figure 7 As shown, the device for verifying the insurance policy specifically includes:

[0123] Extraction module 701, used to extract disease combinations from historical insurance policy claim data of the user to be verified and establish a disease impact matrix;

[0124] The disease module 702 is used to determine the influencing disease of the user to be verified according to the disease influence matrix;

[0125] A construction module 703 is used to extract a user feature vector and a historical disease vector of the user to be verified from the historical insurance policy claim data, and construct a claim vector of the user to be verified;

[0126] A similarity module 704 is used to determine similar users to the user to be verified based on the claim settlement vector of the user to be verified, so as to use the average claim settlement time interval of the similar users as the claim settlement time of the user to be verified;

[0127] The verification module 705 is used to verify the insurance policy of the user to be verified according to the influencing disease and the claim settlement time of the user to be verified.

[0128] See also Figure 8 , Figure 8 2 is a schematic diagram of an application of a device for verifying an insurance policy according to an embodiment of the present invention. Figure 8 The claims system data server is used to store historical policy claims data.

[0129] The basic data processing server includes an extraction module 701, which obtains historical insurance policy claims data from the claims system data server through the data dedicated network. The data training model server includes a disease module 702 and a construction module 703. The prediction calculation server includes a similarity module 704 and a verification module 705. The web server is used to send the verification results to the result real-time communication server through the service dedicated network.

[0130] That is to say, Figure 7 The device can be Figure 8 Implemented in the server.

[0131] In one embodiment of the present invention, the extraction module 701 is specifically used to extract disease combinations in the historical insurance policy claim data of the user to be verified, and determine the vector of each disease;

[0132] The disease impact matrix is ​​established based on the vector of each disease.

[0133] In one embodiment of the present invention, the disease module 702 is specifically configured to determine the influencing disease of the user to be verified according to the disease influence matrix and the historical disease vector of the user to be verified.

[0134] In one embodiment of the present invention, the similarity module 704 is specifically configured to determine users similar to the user to be verified based on the vector distance between the claim vector of the user to be verified and the matching vectors of other users.

[0135] In one embodiment of the present invention, the similarity module 704 is specifically configured to determine similar users to the user to be verified according to a preset similarity number and the vector distance.

[0136] In one embodiment of the present invention, the verification module 705 is specifically used to verify whether the insurance policy of the user to be verified is abnormal according to the influencing disease and the claim settlement time of the user to be verified.

[0137] In one embodiment of the present invention, the verification module 705 is specifically used to verify the insurance policy of the user to be verified according to the influencing disease and the claim settlement time of the user to be verified to determine whether it is necessary to send a reminder message.

[0138] Fig. 9 An exemplary system architecture 900 is shown to which the method for verifying insurance policies or the apparatus for verifying insurance policies according to the embodiments of the present invention may be applied.

[0139] like Fig. 9 As shown, system architecture 900 may include terminal devices 901, 902, 903, network 904 and server 905. Network 904 is used to provide a medium for communication links between terminal devices 901, 902, 903 and server 905. Network 904 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0140] Users can use terminal devices 901, 902, 903 to interact with server 905 through network 904 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 901, 902, 903, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0141] The terminal devices 901 , 902 , and 903 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0142] The server 905 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 901, 902, and 903. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as target push information, product information - only an example) to the terminal device.

[0143] It should be noted that the method for verifying the insurance policy provided in the embodiment of the present invention is generally executed by the server 905 , and accordingly, the device for verifying the insurance policy is generally set in the server 905 .

[0144] It should be understood that Fig. 9 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0145] Reference below Fig.10, which shows a schematic diagram of the structure of a computer system 1000 of a terminal device suitable for implementing an embodiment of the present invention. Fig.10 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0146] like Fig.10 As shown, the computer system 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage part 1008 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the system 1000 are also stored. The CPU 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0147] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that a computer program read therefrom is installed into the storage section 1008 as needed.

[0148] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1009, and / or installed from the removable medium 1011. When the computer program is executed by the central processing unit (CPU) 1001, the above-mentioned functions defined in the system of the present invention are executed.

[0149] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0150] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0151] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The modules described may also be provided in a processor, for example, may be described as: a processor including an extraction module, a disease module, a construction module, a similarity module and a verification module. Among them, the names of these modules do not constitute a limitation on the module itself in certain cases, for example, the extraction module may also be described as "used to extract disease combinations in the historical insurance policy claims data of the user to be verified, and establish a disease impact matrix".

[0152] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes:

[0153] Extract disease combinations from historical insurance policy claims data of the user to be verified and establish a disease impact matrix;

[0154] Determining the influencing disease of the user to be verified according to the disease impact matrix;

[0155] Extracting a user feature vector and a historical disease vector of the user to be verified from the historical insurance policy claim data, and constructing a claim vector of the user to be verified;

[0156] Based on the claim settlement vector of the user to be verified, determine users similar to the user to be verified, and use the average claim settlement time interval of the similar users as the claim settlement time of the user to be verified;

[0157] According to the influencing disease and the claim settlement time of the user to be verified, the insurance policy of the user to be verified is verified.

[0158] According to the technical solution of the embodiment of the present invention, the disease combination in the historical insurance policy claim data of the user to be verified is extracted to establish a disease impact matrix; based on the disease impact matrix, the influencing diseases of the user to be verified are determined; the user feature vector and the historical disease vector of the user to be verified are extracted from the historical insurance policy claim data to construct the claim vector of the user to be verified; based on the claim vector of the user to be verified, similar users to the user to be verified are determined, so as to use the average claim time interval of the similar users as the claim time of the user to be verified; the insurance policy of the user to be verified is verified according to the influencing diseases and the claim time of the user to be verified. On the one hand, the influencing diseases are determined based on the historical insurance policy claim data; on the other hand, the claim time is determined from the perspective of similar users. According to the influencing diseases and the claim time, the accuracy of the verification can be improved, thereby reducing the error rate of verifying the insurance policy.

[0159] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for verifying an insurance policy, characterized in that: include: Extract disease combinations from historical insurance policy claims data of the user to be verified and establish a disease impact matrix; Determining the influencing disease of the user to be verified according to the disease impact matrix; Extracting a user feature vector and a historical disease vector of the user to be verified from the historical insurance policy claim data, and constructing a claim vector of the user to be verified; Based on the claim settlement vector of the user to be verified, determine users similar to the user to be verified, and use the average claim settlement time interval of the similar users as the claim settlement time of the user to be verified; Verify the insurance policy of the user to be verified according to the influencing disease and the claim settlement time of the user to be verified; The determining, based on the claim settlement vector of the user to be verified, a user similar to the user to be verified comprises: Based on the user feature vector and the historical disease vector in the claim vector of the user to be verified, and the user feature vector and the historical disease vector in the matching vector of other users, the vector distance between the user to be verified and the other users is calculated to determine users similar to the user to be verified among the other users; The determining of users similar to the user to be verified from among the other users includes: In the ascending order of the vector distance, a preset number of similar users or a preset proportion of users are selected from the other users as similar users to the user to be checked.

2. The method for verifying an insurance policy according to claim 1, characterized in that: The extracting of disease combinations from the historical insurance policy claims data of the user to be verified and establishing a disease impact matrix includes: Extract disease combinations from the historical insurance claims data of the user to be verified, and determine the vector of each disease based on the International Classification of Diseases standard; A weight vector for each disease is constructed based on the vector of each disease, and the weight vector for each disease is arranged in rows to obtain the disease impact matrix.

3. The method for verifying an insurance policy according to claim 1, characterized in that: Determining the influencing disease of the user to be verified according to the disease impact matrix includes: After the historical disease vector of the user to be checked is multiplied by the disease influence matrix, the difference between the historical disease vector and the disease influence matrix is ​​the influence matrix, and the influencing disease of the user to be checked is determined according to the element values ​​in the influence matrix.

4. The method for verifying an insurance policy according to claim 1, characterized in that: According to the influencing disease and the claim settlement time of the user to be verified, verifying the insurance policy of the user to be verified includes: According to the influencing disease and the claim settlement time of the user to be verified, check whether the insurance policy of the user to be verified is abnormal.

5. The method for verifying an insurance policy according to claim 1, characterized in that: According to the influencing disease and the claim settlement time of the user to be verified, verifying the insurance policy of the user to be verified includes: According to the influencing disease and the claim settlement time of the user to be verified, the insurance policy of the user to be verified is checked to determine whether a reminder message needs to be sent.

6. A device for verifying an insurance policy, characterized in that: include: The extraction module is used to extract the disease combinations in the historical insurance policy claims data of the user to be verified and establish the disease impact matrix; A disease module, used for determining the influencing disease of the user to be verified according to the disease impact matrix; A construction module, used to extract the user feature vector and the historical disease vector of the user to be verified from the historical insurance policy claim data, and construct a claim vector of the user to be verified; A similarity module, configured to determine similar users to the user to be verified based on the claim settlement vector of the user to be verified, so as to use the average claim settlement time interval of the similar users as the claim settlement time of the user to be verified; And, based on the user feature vector and the historical disease vector in the claim vector of the user to be verified, and the user feature vector and the historical disease vector in the matching vector of other users, calculating the vector distance between the user to be verified and the other users, so as to determine users similar to the user to be verified among the other users; and, selecting a preset number of similar users or a preset proportion of users from the other users in the ascending order of the vector distances as similar users to the user to be verified; The verification module is used to verify the insurance policy of the user to be verified according to the influencing disease and the claim settlement time of the user to be verified.

7. An electronic device for checking an insurance policy, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

8. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images