User processing method based on linear regression model

Abstract

The invention relates to the technical field of electric digital data processing, in particular to a user processing method based on a linear regression model. The method comprises the steps of obtaining a training user set; obtaining a preset linear regression model; solving a preset linear regression model by using the standardized data and insurance premium of the training users included in the training user set to obtain a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9; acquiring initial data of a to-be-insured user; standardizing the initial data of the to-be-insured user to obtain standardized data of the to-be-insured user; predicting the premium of the to-be-insured user according to the standardized data of the to-be-insured user and the solved value; and if the difference value between the predicted insurance premium and the target insurance premium of the user to be insured is greater than or equal to a preset difference threshold value, outputting preset information for indicating abnormity. According to the invention, abnormity judgment can be carried out on the insurance premium of the to-be-insured user determined manually.

Description

Technical Field

[0001] The invention relates to the technical field of electronic digital data processing, and in particular to a user processing method based on a linear regression model. Background Art

[0002] In the prior art, the premium of the user to be insured is usually determined manually based on the personal information of the user to be insured. The premiums of different users to be insured will be different due to the different personal information of the users to be insured. Generally speaking, different personal information corresponds to different claims risks. In order to increase the willingness of people with lower claims risks to insure and reduce the compensation costs of people with higher claims risks, higher premiums are usually set for users to be insured with higher claims risks, and lower premiums are set for users to be insured with lower claims risks. However, when the premium of the user to be insured is determined manually based on the personal information of the user to be insured, errors may occur, that is, the personal information of the user to be insured does not match the premium of the user to be insured determined manually. How to make abnormal judgments on the premium of the user to be insured determined manually is a problem that needs to be solved urgently. Summary of the invention

[0003] The purpose of the present invention is to provide a user processing method based on a linear regression model to make abnormal judgments on the premiums of manually determined users to be insured.

[0004] According to the present invention, a user processing method based on a linear regression model is provided, the processing method comprising the following steps:

[0005] S100, obtaining a training user set; the training user set includes a number of training users, and the standardized data of each training user includes the corresponding training user's standardized age, a first label for characterizing gender, a standardized body mass index, a second label for characterizing the presence or absence of chronic diseases, a standardized smoking frequency, a standardized drinking frequency, a standardized exercise frequency, a third label for characterizing whether the occupation is a high-risk occupation, and a standardized income; the premium of each training user is obtained based on the actual number of claims and the amount of claims settled for the training user.

[0006] S200, obtain a preset linear regression model A=a0+w1×a1+w2×a2+w3×a3+w4×a4+w5×a5+w6×a6+w7×a7+w8×a8+w9×a9+b0, where A is the premium, a0 is the intercept, w1, w2, w3, w4, w5, w6, w7, w8 and w9 are the first coefficient, the second coefficient, the third coefficient, the fourth coefficient, the fifth coefficient, the sixth coefficient, the seventh coefficient, the eighth coefficient and the ninth coefficient respectively, a1, a2, a3, a4, a5, a6, a7, a8 and a9 are the standardized age, the first label for characterizing gender, the standardized body mass index, the second label for characterizing the presence or absence of chronic diseases, the standardized smoking frequency, the standardized drinking frequency, the standardized exercise frequency, the third label for characterizing whether the occupation is a high-risk occupation and the standardized income respectively, and b0 is the error.

[0007] S300, using the standardized data and premiums of the training users included in the training user set to solve the preset linear regression model, to obtain a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9.

[0008] S400, obtaining initial data of the user to be insured; the initial data includes age, gender, body mass index, chronic disease history, smoking habits, drinking habits, exercise frequency, occupation type and income.

[0009] S500, standardizing the initial data of the user to be insured to obtain standardized data of the user to be insured.

[0010] S600, predicting the premium of the user to be insured based on the standardized data of the user to be insured and the solved a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9.

[0011] S700: If the difference between the predicted premium of the user to be insured and the target premium of the user to be insured is greater than or equal to a preset difference threshold, output preset information for indicating an abnormality.

[0012] Compared with the prior art, the present invention has at least the following beneficial effects:

[0013] In the present invention, the standardized data of each training user includes information related to age, gender, body mass index, presence or absence of chronic diseases, smoking frequency, drinking frequency, exercise frequency, occupation and income. The preset linear regression model includes independent variables corresponding to this information. Based on the standardized data of the training user set and the corresponding premium, the unknown intercept and coefficient in the preset linear regression model can be obtained. On this basis, the premium of the user to be insured can be predicted based on the standardized data of the user to be insured and the obtained intercept and coefficient; in the present invention, the above information is more relevant to the risk of claims settlement, and the premium of the user obtained based on the above information is more accurate and more matched with the user. If the predicted premium of the user to be insured is significantly different from the target premium (i.e., the premium of the user to be insured determined manually), preset information for indicating the abnormality is output to prompt that the manually determined premium may not match the corresponding personal information. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0015] Figure 1 A flowchart of a user processing method based on a linear regression model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0017] According to this embodiment, Figure 1 As shown, a user processing method based on a linear regression model is provided, and the processing method comprises the following steps:

[0018] S100, obtaining a training user set; the training user set includes a number of training users, and the standardized data of each training user includes the corresponding training user's standardized age, a first label for characterizing gender, a standardized body mass index, a second label for characterizing the presence or absence of chronic diseases, a standardized smoking frequency, a standardized drinking frequency, a standardized exercise frequency, a third label for characterizing whether the occupation is a high-risk occupation, and a standardized income; the premium of each training user is obtained based on the actual number of claims and the amount of claims settled for the training user.

[0019] Optionally, the values ​​of the first label, the second label and the third label are 0 or 1; when the first label is 0, it indicates that the gender is male; when the first label is 1, it indicates that the gender is female; when the second label is 0, it indicates that there is no chronic disease; when the second label is 1, it indicates that there is a chronic disease; when the third label is 0, it indicates that the occupation is a low-risk occupation; when the third label is 1, it indicates that the occupation is a high-risk occupation.

[0020] Optionally, the value range of the standardized age, standardized body mass index, standardized smoking frequency, standardized drinking frequency, standardized exercise frequency and standardized income are all 0-1; the standardized age is the ratio of the user's age to a preset maximum age, the standardized body mass index is the ratio of the user's body mass index to a preset maximum body mass index, the standardized smoking frequency is the ratio of the number of days the user smokes per week to 7, the standardized drinking frequency is the ratio of the number of days the user drinks per week to 7, the standardized exercise frequency is the ratio of the number of days the user exercises per week to 7, and the standardized income is the ratio of the user's income to the preset maximum income.

[0021] As a preferred specific implementation method, the premium of each training user is basic premium + claim number coefficient × actual number of claims + claim amount coefficient × claim amount, the basic premium includes a preset minimum premium, a preset average profit for each insurance policy and a preset management fee for each insurance policy, the claim number coefficient is the ratio of the average claim amount of all users in the historical time period to the average number of claims, the claim amount coefficient is the ratio of the additional expenses expected by the insurance company to the total claim amount, the additional expenses expected by the insurance company refer to the expenses that the insurance company hopes to preset in addition to the basic premium in order to cover additional risks and costs, the actual number of claims is the total number of claims of the corresponding training user in the historical time period, the claim amount is the total amount of claims of the corresponding training user in the historical time period, and optionally, the historical time period is the most recent 1 year.

[0022] As a preferred specific implementation, the process of acquiring the training user set includes:

[0023] S010, obtaining a historical user set, the historical user set includes several historical users, and the standardized data of each historical user includes the corresponding historical user's standardized age, a first label for characterizing gender, a standardized body mass index, a second label for characterizing the presence or absence of chronic diseases, a standardized smoking frequency, a standardized drinking frequency, a standardized exercise frequency, a third label for characterizing whether the occupation is a high-risk occupation, and standardized income.

[0024] S020, grouping historical users in the historical user set according to corresponding standardized data, so that the historical users in each group have standardized ages that belong to the same preset age range, the same first label for characterizing gender, the same body mass index that belongs to the same preset body mass index range, the same second label for characterizing the presence or absence of chronic diseases, the same smoking frequency that belongs to the same preset smoking frequency range, the same drinking frequency that belongs to the same preset drinking frequency range, the same exercise frequency that belongs to the same preset exercise frequency range, the same third label for characterizing whether the occupation is a high-risk occupation, and the same standardized income that belongs to the same preset income range.

[0025] As a preferred specific implementation, the value range of each standardized data is divided into several sub-ranges. If the sub-ranges to which all the standardized data of two historical users belong are the same, then the two historical users belong to the same group. For example, the value range of each standardized data is divided into two sub-ranges, so the historical users in the historical user set can be divided into two sub-ranges. 9 Based on this preferred specific implementation, the preliminary division of historical users can be achieved by simply judging whether each standardized data of each historical user conforms to the corresponding sub-range, with a relatively fast division speed and a relatively high division efficiency.

[0026] S030, dividing the historical users in each group into a number of clusters according to the distance of the standardized data of the historical users in the group.

[0027] In this embodiment, each historical user in the same group is taken as an object to be clustered, and the k-means method is used to realize the division of historical users in the same group, where k is an empirical value; in the process of using the k-means method to divide historical users in the same group, the distance between any two historical users needs to be obtained. As a preferred specific implementation method, the historical users in each group are divided into several clusters according to the distance of the standardized data of the historical users in the group, including:

[0028] S031, obtaining standardized data of a first historical user; the first historical user is any user in any group.

[0029] S032, obtaining standardized data of a second historical user; the second historical user and the first historical user belong to the same group, and the second historical user is different from the first historical user.

[0030] S033, obtaining the distance between the standardized data of the first historical user and the second historical user according to the standardized data of the first historical user and the standardized data of the second historical user; the distance between the standardized data of the first historical user and the second historical user is obtained according to the preset first weight, the standardized age difference value between the first historical user and the second historical user, the preset second weight, the first label difference value between the first historical user and the second historical user, the preset third weight, the standardized weight index difference value between the first historical user and the second historical user, the preset fourth weight, the second label difference value between the first historical user and the second historical user, the preset fifth weight, the standardized smoking frequency difference value between the first historical user and the second historical user, the preset sixth weight, the standardized drinking frequency difference value between the first historical user and the second historical user, the preset seventh weight, the standardized exercise frequency between the first historical user and the second historical user, the preset eighth weight, the third label difference value between the first historical user and the second historical user, the preset ninth weight and the standardized income difference value between the first historical user and the second historical user.

[0031] In this embodiment, the distance between the standardized data of the first historical user and the second historical user is obtained by using a weighted sum formula, and the difference value refers to the absolute value of the difference. As a preferred specific implementation, the process of obtaining the preset first weight, the preset second weight, the preset third weight, the preset fourth weight, the preset fifth weight, the preset sixth weight, the preset seventh weight, the preset eighth weight, and the preset ninth weight includes:

[0032] S0331, obtain the first Pearson correlation coefficient between the standardized age of historical users and the premiums of historical users.

[0033] Those skilled in the art know that any method for obtaining the Pearson correlation coefficient in the prior art falls within the protection scope of the present invention.

[0034] S0332, obtaining the second Pearson correlation coefficient between the first label of the historical user and the premium of the historical user.

[0035] S0333, obtaining the third Pearson correlation coefficient between the standardized body mass index of the historical user and the insurance premium of the historical user.

[0036] S0334, obtaining the fourth Pearson correlation coefficient between the second label of the historical user and the premium of the historical user.

[0037] S0335, obtain the fifth Pearson correlation coefficient between the standardized smoking frequency of historical users and the premium of historical users.

[0038] S0336, obtain the sixth Pearson correlation coefficient between the standardized drinking frequency of the historical user and the premium of the historical user.

[0039] S0337, obtain the seventh Pearson correlation coefficient between the standardized exercise frequency of the historical user and the premium of the historical user.

[0040] S0338, obtaining the eighth Pearson correlation coefficient between the third tag of the historical user and the premium of the historical user.

[0041] S0339, obtain the ninth Pearson correlation coefficient between the standardized income of historical users and the premiums of historical users.

[0042] S03310, normalize the absolute value of the first Pearson correlation coefficient, the absolute value of the second Pearson correlation coefficient, the absolute value of the third Pearson correlation coefficient, the absolute value of the fourth Pearson correlation coefficient, the absolute value of the fifth Pearson correlation coefficient, the absolute value of the sixth Pearson correlation coefficient, the absolute value of the seventh Pearson correlation coefficient, the absolute value of the eighth Pearson correlation coefficient, and the absolute value of the ninth Pearson correlation coefficient, and determine the absolute value of the first Pearson correlation coefficient after the normalization as a preset first weight, determine the absolute value of the second Pearson correlation coefficient after the normalization as a preset second weight, and determine the absolute value of the second Pearson correlation coefficient after the normalization as a preset second weight. The absolute value of the third Pearson correlation coefficient is determined as the preset third weight, the absolute value of the fourth Pearson correlation coefficient after normalization is determined as the preset fourth weight, the absolute value of the fifth Pearson correlation coefficient after normalization is determined as the preset fifth weight, the absolute value of the sixth Pearson correlation coefficient after normalization is determined as the preset sixth weight, the absolute value of the seventh Pearson correlation coefficient after normalization is determined as the preset seventh weight, the absolute value of the eighth Pearson correlation coefficient after normalization is determined as the preset eighth weight, and the absolute value of the ninth Pearson correlation coefficient after normalization is determined as the preset ninth weight.

[0043] In this embodiment, the absolute value of the first Pearson correlation coefficient after normalization is the ratio of the absolute value of the first Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the second Pearson correlation coefficient after normalization is the ratio of the absolute value of the second Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the third Pearson correlation coefficient after normalization is the ratio of the absolute value of the third Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the fourth Pearson correlation coefficient after normalization is the ratio of the absolute value of the fourth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, and the absolute value of the fifth Pearson correlation coefficient after normalization is the ratio of the absolute value of the fifth Pearson correlation coefficient to the sum of the absolute values ​​of the fifth Pearson correlation coefficient. The absolute value of the sixth Pearson correlation coefficient after normalization is the ratio of the absolute value of the sixth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the seventh Pearson correlation coefficient after normalization is the ratio of the absolute value of the seventh Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the eighth Pearson correlation coefficient after normalization is the ratio of the absolute value of the eighth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, and the absolute value of the ninth Pearson correlation coefficient after normalization is the ratio of the absolute value of the ninth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients.

[0044] Based on this preferred specific implementation, the weight corresponding to the standardized data that is more relevant to the premium is larger. Therefore, the premiums of historical users in the same cluster are closer, which is conducive to improving the accuracy of subsequent screening of historical users with abnormal premiums based on the premiums of historical users in the cluster.

[0045] S040, screen out abnormal historical users in the cluster according to the premiums of historical users in each cluster; the premium of any historical user is positively correlated with the actual number of claims of the historical user, and the premium of any historical user is positively correlated with the claim amount of the historical user.

[0046] In this embodiment, the historical users with abnormal premiums in the cluster refer to historical users with a large premium difference from other historical users in the cluster. As a preferred specific implementation method, screening out abnormal historical users in the cluster according to the premiums of each historical user in the cluster includes:

[0047] S041, obtaining the number of historical users in a first cluster; the first cluster is any cluster obtained by dividing historical users in any group.

[0048] S042, obtaining the variance of the premiums of historical users in the first cluster.

[0049] Those skilled in the art know that any method for obtaining variance in the prior art falls within the protection scope of the present invention.

[0050] S043, matching the number of historical users in the first cluster, the average number of historical users in all clusters, the variance of the premiums of historical users in the first cluster, and the average variance of the premiums of historical users in all clusters in a preset screening ratio table; the preset screening ratio table includes a plurality of entries, each entry includes a quantity range, an average quantity range, a variance range, an average variance range, and a screening ratio.

[0051] In this embodiment, the average number of historical users of all clusters is the ratio of the number of historical users included in the historical user set to the number of clusters, and the average variance of the premiums of historical users of all clusters is the ratio of the sum of the variances of the premiums of historical users of all clusters to the number of clusters.

[0052] In this embodiment, the preset screening ratio table is a pre-constructed table. As a preferred specific implementation, the table satisfies the following conditions: for two entries with the same quantity range, the same average quantity range and the same variance range, the screening ratio of the entry with the larger average variance corresponding to the average variance range is smaller than the screening ratio of the entry with the smaller average variance corresponding to the average variance range; for two entries with the same quantity range, the same average quantity range and the same average variance range, the screening ratio of the entry with the larger variance corresponding to the variance range is greater than the screening ratio of the entry with the smaller variance corresponding to the variance range; for two entries with the same quantity range, the same variance range and the same average variance range, the screening ratio of the entry with the larger average quantity corresponding to the average quantity range is smaller than the screening ratio of the entry with the smaller average quantity corresponding to the average quantity range; for two entries with the same average quantity range, the same variance range and the same average variance range, the screening ratio of the entry with the larger quantity corresponding to the quantity range is greater than the screening ratio of the entry with the smaller quantity corresponding to the quantity range. Based on this preferred specific implementation, while taking into account the amount of data for training users, the difference in the number of historical users in each cluster after subsequent screening can be made smaller, and the premiums of the historical users in each retained cluster are more accurate, which is conducive to improving the accuracy of subsequent linear regression models in predicting different users to be insured.

[0053] S044: If the match is successful, determine the screening ratio of the successfully matched entries as the screening ratio of the first cluster.

[0054] In this embodiment, S044 also includes a process for determining whether the match is successful, which includes: if the quantity range included in a certain entry in the preset screening ratio table includes the number of historical users in the first cluster, the average quantity range included in the entry includes the average number of historical users of all clusters, the variance range included in the entry includes the variance of the premiums of the historical users in the first cluster and the average variance range included in the entry includes the average variance of the premiums of the historical users of all clusters, then the match is determined to be successful, and the entry is determined to be a successfully matched entry.

[0055] In this embodiment, if the matching is unsuccessful, historical users with abnormal premiums in the cluster will not be screened out.

[0056] S045, sorting the historical users in the first cluster in ascending order of corresponding premiums, and screening out the first percentage of historical users with the smallest premiums and the first percentage of historical users with the largest premiums in the first cluster from the first cluster; the first percentage is half of the screening ratio of the first cluster.

[0057] In this embodiment, the top first percentage of historical users with the smallest premiums and the top first percentage of historical users with the largest premiums in the first cluster are determined as historical users with abnormal premiums in the first cluster.

[0058] S050, determining a historical user set consisting of the remaining historical users in all clusters as a training user set.

[0059] The training user set obtained based on S010-S050 can improve the accuracy of the linear regression model's prediction of different users to be insured.

[0060] S200, obtain a preset linear regression model A=a0+w1×a1+w2×a2+w3×a3+w4×a4+w5×a5+w6×a6+w7×a7+w8×a8+w9×a9+b0, where A is the premium, a0 is the intercept, w1, w2, w3, w4, w5, w6, w7, w8 and w9 are the first coefficient, the second coefficient, the third coefficient, the fourth coefficient, the fifth coefficient, the sixth coefficient, the seventh coefficient, the eighth coefficient and the ninth coefficient respectively, a1, a2, a3, a4, a5, a6, a7, a8 and a9 are the standardized age, the first label for characterizing gender, the standardized body mass index, the second label for characterizing the presence or absence of chronic diseases, the standardized smoking frequency, the standardized drinking frequency, the standardized exercise frequency, the third label for characterizing whether the occupation is a high-risk occupation and the standardized income respectively, and b0 is the error.

[0061] In this embodiment, A in the preset linear regression model is the dependent variable, a1, a2, a3, a4, a5, a6, a7, a8 and a9 are independent variables, and a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9 are all unknowns to be solved. Based on the preset linear regression model, this embodiment can subsequently predict the premium of the user to be insured more accurately.

[0062] S300, using the standardized data and premiums of the training users included in the training user set to solve the preset linear regression model, to obtain a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9.

[0063] Those skilled in the art know that any method for solving a linear regression model in the prior art falls within the protection scope of the present invention.

[0064] S400, obtaining initial data of the user to be insured; the initial data includes age, gender, body mass index, chronic disease history, smoking habits, drinking habits, exercise frequency, occupation type and income.

[0065] S500, standardizing the initial data of the user to be insured to obtain standardized data of the user to be insured.

[0066] S600, predicting the premium of the user to be insured based on the standardized data of the user to be insured and the solved a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9.

[0067] In this embodiment, the premium of the user to be insured is obtained by summing the following values: the product of a0, w1 and the standardized age of the user to be insured, the product of w2 and the first label used to characterize gender, the product of w3 and the standardized body mass index, the product of w4 and the second label used to characterize the presence or absence of chronic diseases, the product of w5 and the standardized smoking frequency, the product of w6 and the standardized drinking frequency, the product of w7 and the standardized exercise frequency, the product of w8 and the third label used to characterize whether the occupation is a high-risk occupation, and the product of w9 and the standardized income.

[0068] S700: If the difference between the predicted premium of the user to be insured and the target premium of the user to be insured is greater than or equal to a preset difference threshold, output preset information for indicating an abnormality.

[0069] In this embodiment, the target premium is the manually determined premium of the user to be insured.

[0070] In this embodiment, the preset difference threshold is an empirical value.

[0071] In this embodiment, the standardized data of each training user includes information related to age, gender, body mass index, presence or absence of chronic diseases, smoking frequency, drinking frequency, exercise frequency, occupation and income. The preset linear regression model includes independent variables corresponding to this information. The unknown intercept and coefficient in the preset linear regression model can be obtained based on the standardized data of the training user set and the corresponding premium. On this basis, the premium of the user to be insured can be predicted based on the standardized data of the user to be insured and the obtained intercept and coefficient. In this embodiment, the above information is more relevant to the risk of claims settlement, and the premium of the user obtained based on the above information is more accurate and more matched with the user. If the predicted premium of the user to be insured is significantly different from the target premium (i.e., the premium of the user to be insured determined manually), preset information for indicating the abnormality is output to prompt that the manually determined premium may not match the corresponding personal information.

[0072] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A user processing method based on a linear regression model, characterized in that: The processing method comprises the following steps: S100, obtaining a training user set; the training user set includes a number of training users, and the standardized data of each training user includes the corresponding training user's standardized age, a first label for representing gender, a standardized body mass index, a second label for representing the presence or absence of chronic diseases, a standardized smoking frequency, a standardized drinking frequency, a standardized exercise frequency, a third label for representing whether the occupation is a high-risk occupation, and a standardized income; the premium of each training user is obtained according to the actual number of claims and the amount of claims of the training user; S200, obtaining a preset linear regression model A=a0+w1×a1+w2×a2+w3×a3+w4×a4+w5×a5+w6×a6+w7×a7+w8×a8+w9×a9+b0, where A is the premium, a0 is the intercept, w1, w2, w3, w4, w5, w6, w7, w8 and w9 are the first coefficient, the second coefficient, the third coefficient, the fourth coefficient, the fifth coefficient, the sixth coefficient, the seventh coefficient, the eighth coefficient and the ninth coefficient respectively, a1, a2, a3, a4, a5, a6, a7, a8 and a9 are the standardized age, the first label for representing gender, the standardized body mass index, the second label for representing the presence or absence of chronic diseases, the standardized smoking frequency, the standardized drinking frequency, the standardized exercise frequency, the third label for representing whether the occupation is a high-risk occupation and the standardized income respectively, and b0 is the error; S300, solving a preset linear regression model using standardized data and premiums of training users included in the training user set to obtain a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9; S400, obtaining initial data of the user to be insured; the initial data includes age, gender, body mass index, chronic disease history, smoking habits, drinking habits, exercise frequency, occupation type and income; S500, standardizing the initial data of the user to be insured to obtain standardized data of the user to be insured; S600, predicting the insurance premium of the user to be insured based on the standardized data of the user to be insured and the solved a0, w1, w2, w3, w4, w5, w6, w7, w8 and w9; S700: If the difference between the predicted premium of the user to be insured and the target premium of the user to be insured is greater than or equal to a preset difference threshold, output preset information for indicating an abnormality.

2. The user processing method based on the linear regression model according to claim 1, characterized in that: The process of obtaining the training user set includes: S010, obtaining a historical user set, wherein the historical user set includes a number of historical users, and the standardized data of each historical user includes the corresponding historical user's standardized age, a first label for representing gender, a standardized body mass index, a second label for representing the presence or absence of chronic diseases, a standardized smoking frequency, a standardized drinking frequency, a standardized exercise frequency, a third label for representing whether the occupation is a high-risk occupation, and a standardized income; S020, grouping historical users in the historical user set according to corresponding standardized data, so that the standardized ages of the historical users included in each group belong to the same preset age range, the first labels used to characterize gender are the same, the standardized body mass index belongs to the same preset body mass index range, the second labels used to characterize the presence or absence of chronic diseases are the same, the standardized smoking frequency belongs to the same preset smoking frequency range, the standardized drinking frequency belongs to the same preset drinking frequency range, the standardized exercise frequency belongs to the same preset exercise frequency range, the third labels used to characterize whether the occupation is a high-risk occupation are the same, and the standardized income belongs to the same preset income range; S030, dividing the historical users in each group into a number of clusters according to the distance of the standardized data of the historical users in the group; S040, based on the premiums of historical users in each cluster, screen out historical users with abnormal premiums in the cluster; the premium of any historical user is positively correlated with the actual number of claims of the historical user, and the premium of any historical user is positively correlated with the claim amount of the historical user; S050, determining a historical user set consisting of the remaining historical users in all clusters as a training user set.

3. The user processing method based on the linear regression model according to claim 2, characterized in that: Screening out historical users with abnormal premiums in a cluster based on the premiums of historical users in each cluster includes: S041, obtaining the number of historical users in a first cluster; the first cluster is any cluster obtained by dividing historical users in any group; S042, obtaining the variance of the premiums of historical users in the first cluster; S043, matching the number of historical users in the first cluster, the average number of historical users in all clusters, the variance of the premiums of historical users in the first cluster, and the average variance of the premiums of historical users in all clusters in a preset screening ratio table; the preset screening ratio table includes a plurality of entries, each entry includes a quantity range, an average quantity range, a variance range, an average variance range, and a screening ratio; S044, if the match is successful, determining the screening ratio of the successfully matched entries as the screening ratio of the first cluster; S045, sorting the historical users in the first cluster in ascending order of corresponding premiums, and screening out the first percentage of historical users with the smallest premiums and the first percentage of historical users with the largest premiums in the first cluster from the first cluster; the first percentage is half of the screening ratio of the first cluster.

4. The user processing method based on the linear regression model according to claim 2, characterized in that: The historical users in each group are divided into several clusters according to the distance of the standardized data of the historical users in the group, including: S031, obtaining standardized data of a first historical user; the first historical user is any user in any group; S032, obtaining standardized data of a second historical user; the second historical user and the first historical user belong to the same group, and the second historical user is different from the first historical user; S033, obtaining the distance between the standardized data of the first historical user and the second historical user according to the standardized data of the first historical user and the standardized data of the second historical user; the distance between the standardized data of the first historical user and the second historical user is obtained according to the preset first weight, the standardized age difference value between the first historical user and the second historical user, the preset second weight, the first label difference value between the first historical user and the second historical user, the preset third weight, the standardized weight index difference value between the first historical user and the second historical user, the preset fourth weight, the second label difference value between the first historical user and the second historical user, the preset fifth weight, the standardized smoking frequency difference value between the first historical user and the second historical user, the preset sixth weight, the standardized drinking frequency difference value between the first historical user and the second historical user, the preset seventh weight, the standardized exercise frequency between the first historical user and the second historical user, the preset eighth weight, the third label difference value between the first historical user and the second historical user, the preset ninth weight and the standardized income difference value between the first historical user and the second historical user.

5. The user processing method based on the linear regression model according to claim 4 is characterized in that: The process of obtaining the preset first weight, the preset second weight, the preset third weight, the preset fourth weight, the preset fifth weight, the preset sixth weight, the preset seventh weight, the preset eighth weight, and the preset ninth weight includes: S0331, obtain the first Pearson correlation coefficient between the standardized age of the historical user and the premium of the historical user; S0332, obtaining the second Pearson correlation coefficient between the first label of the historical user and the premium of the historical user; S0333, obtain the third Pearson correlation coefficient between the standardized body mass index of the historical user and the premium of the historical user; S0334, obtaining the fourth Pearson correlation coefficient between the second tag of the historical user and the premium of the historical user; S0335, obtain the fifth Pearson correlation coefficient between the standardized smoking frequency of historical users and the premium of historical users; S0336, obtain the sixth Pearson correlation coefficient between the standardized drinking frequency of the historical user and the historical user's premium; S0337, obtain the seventh Pearson correlation coefficient between the standardized exercise frequency of the historical user and the premium of the historical user; S0338, obtaining the eighth Pearson correlation coefficient between the third tag of the historical user and the premium of the historical user; S0339, obtain the ninth Pearson correlation coefficient between the standardized income of historical users and the premiums of historical users; S03310, normalize the absolute value of the first Pearson correlation coefficient, the absolute value of the second Pearson correlation coefficient, the absolute value of the third Pearson correlation coefficient, the absolute value of the fourth Pearson correlation coefficient, the absolute value of the fifth Pearson correlation coefficient, the absolute value of the sixth Pearson correlation coefficient, the absolute value of the seventh Pearson correlation coefficient, the absolute value of the eighth Pearson correlation coefficient, and the absolute value of the ninth Pearson correlation coefficient, and determine the absolute value of the first Pearson correlation coefficient after the normalization as a preset first weight, determine the absolute value of the second Pearson correlation coefficient after the normalization as a preset second weight, and determine the absolute value of the second Pearson correlation coefficient after the normalization as a preset second weight. The absolute value of the third Pearson correlation coefficient is determined as the preset third weight, the absolute value of the fourth Pearson correlation coefficient after normalization is determined as the preset fourth weight, the absolute value of the fifth Pearson correlation coefficient after normalization is determined as the preset fifth weight, the absolute value of the sixth Pearson correlation coefficient after normalization is determined as the preset sixth weight, the absolute value of the seventh Pearson correlation coefficient after normalization is determined as the preset seventh weight, the absolute value of the eighth Pearson correlation coefficient after normalization is determined as the preset eighth weight, and the absolute value of the ninth Pearson correlation coefficient after normalization is determined as the preset ninth weight.

6. The user processing method based on the linear regression model according to claim 5, characterized in that: The absolute value of the first Pearson correlation coefficient after normalization is the ratio of the absolute value of the first Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the second Pearson correlation coefficient after normalization is the ratio of the absolute value of the second Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the third Pearson correlation coefficient after normalization is the ratio of the absolute value of the third Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the fourth Pearson correlation coefficient after normalization is the ratio of the absolute value of the fourth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, and the absolute value of the fifth Pearson correlation coefficient after normalization is the ratio of the absolute value of the fifth Pearson correlation coefficient. The absolute value of the relationship number is the ratio of the absolute value of the relationship number to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the sixth Pearson correlation coefficient after normalization is the ratio of the absolute value of the sixth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the seventh Pearson correlation coefficient after normalization is the ratio of the absolute value of the seventh Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, the absolute value of the eighth Pearson correlation coefficient after normalization is the ratio of the absolute value of the eighth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients, and the absolute value of the ninth Pearson correlation coefficient after normalization is the ratio of the absolute value of the ninth Pearson correlation coefficient to the sum of the absolute values ​​of all Pearson correlation coefficients.

7. The user processing method based on the linear regression model according to claim 3, characterized in that: S044 also includes a process for determining whether the match is successful, which includes: if the quantity range included in a certain entry in the preset screening ratio table includes the number of historical users in the first cluster, the average quantity range included in the entry includes the average number of historical users in all clusters, the variance range included in the entry includes the variance of the premiums of the historical users in the first cluster and the average variance range included in the entry includes the average variance of the premiums of the historical users in all clusters, then the match is determined to be successful, and the entry is determined to be a successfully matched entry.

Images