Insurance package pushing method and device, server and storage medium
By building and training neural network models, generating and verifying insurance package numbers, the problem that insurance packages in the existing technology cannot meet customer needs is solved, and personalized matching and improving customer experience is achieved.
Patent Information
- Application Number
- CN202411670293.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-20
- Publication Date
- 2025-02-07
AI Technical Summary
In the prior art, choosing insurance packages by users spontaneously logging into the insurance platform may result in the insurance packages being unable to meet customers' personalized needs and reduce customer experience.
By obtaining customer account data, building and training neural network models, generating insurance package numbers to be verified, and verifying the effectiveness of insurance packages. If passed, push it to the client.
It realizes personalized matching of insurance packages, can meet customer needs more accurately and improve customer experience.
Smart Images
Figure CN119398943A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of financial technology, and in particular to an insurance package push method, device, server and storage medium. Background Art
[0002] An insurance package is a combination of multiple insurance products, designed to meet the different needs of customers and simplify the insurance purchase process. Insurance packages are usually designed by insurance companies based on customer needs and risk profiles, and may include life insurance, medical insurance, accident insurance, and other types of insurance. Therefore, the choice of an insurance package is particularly important.
[0003] In the prior art, the traditional insurance package push method is that the user spontaneously logs into the insurance platform to select the insurance package and completes the purchase of the insurance package.
[0004] However, in the prior art, the selection of insurance packages by users spontaneously logging into the insurance platform may result in the insurance packages not being able to meet the personalized needs of customers, thereby reducing customer experience. Summary of the invention
[0005] The present application provides an insurance package push method, device, server and storage medium to solve the problem that when users spontaneously log in to an insurance platform to select an insurance package, the insurance package may not meet the actual needs of the customer, thereby reducing the customer experience.
[0006] In a first aspect, the present application provides an insurance package push method, which is applied to a server and includes:
[0007] Obtain customer account data and build the first neural network model;
[0008] determining an insurance adjustment parameter, and adjusting the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data;
[0009] Training the first neural network model according to the adjusted customer account data to obtain a second neural network model;
[0010] In response to a triggering operation of any customer on the client, generating an insurance package request;
[0011] Inputting the insurance package request into the second neural network model to generate an insurance package number to be verified;
[0012] Determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package;
[0013] If the insurance package passes the validity verification, the insurance package is pushed to the client.
[0014] In one possible design, obtaining customer account data includes: obtaining the customer's insurance package address; retrieving the corresponding insurance package script unit according to each insurance package address, wherein the insurance package script unit includes one or more insurance package scripts; parsing each insurance package script unit to obtain all insurance package scripts; determining a correlation coefficient of each insurance package script, and deleting the insurance package scripts whose correlation coefficient exceeds a preset critical value to obtain the remaining insurance package scripts; and determining the remaining insurance package scripts as the customer account data.
[0015] In one possible design, the insurance adjustment parameters are a full set of insurance adjustment parameters, and the adjusted customer account data are parameter selection instruction data; determining the insurance adjustment parameters includes: constructing a third neural network model, and inputting the customer account data into the third neural network model to obtain a full set of insurance adjustment parameters; accordingly, adjusting the customer account data according to the insurance adjustment parameters to obtain the adjusted customer account data includes: determining the corresponding insurance package number according to the full set of insurance adjustment parameters, and generating the corresponding parameter selection instruction data according to the insurance package number; accordingly, training the first neural network model according to the adjusted customer account data to obtain the second neural network model includes: training the first neural network model according to the parameter selection instruction data to obtain the second neural network model.
[0016] In one possible design, the training of the first neural network model according to the parameter selection instruction data to obtain the second neural network model includes: dividing the parameter selection instruction data into a data set to obtain a training set, a validation set and a test set; adjusting the first neural network model according to the training set to obtain an adjusted first neural network model; verifying the adjusted first neural network model according to the validation set to determine whether the error requirement is met; if the adjusted first neural network model meets the error requirement, testing the adjusted first neural network model according to the test set; if the adjusted first neural network model passes the test, determining the adjusted first neural network model as the second neural network model.
[0017] In one possible design, the insurance adjustment parameter is an explanatory insurance adjustment parameter, and the adjusted customer account data is an adjustment data set; determining the insurance adjustment parameter includes: classifying the insurance package numbers of the customer account data into levels to obtain the insurance packages to be processed after each level classification; determining the corresponding explanatory insurance adjustment parameter according to the insurance packages to be processed after each level classification; accordingly, adjusting the customer account data according to the insurance adjustment parameter to obtain the adjusted customer account data includes: adjusting the customer account data according to each explanatory insurance adjustment parameter to obtain the adjustment data set; accordingly, training the first neural network model according to the adjusted customer account data to obtain the second neural network model includes: training the first neural network model according to the adjustment data set to obtain the second neural network model.
[0018] In one possible design, the insurance adjustment parameter is an insurance adjustment parameter of an insurance package with a complaint history, and the adjusted customer account data is customer account data with records, and determining the insurance adjustment parameter includes: performing complaint verification on the insurance package number of the customer account data to obtain the insurance adjustment parameter of the insurance package with a complaint history; accordingly, adjusting the customer account data according to the insurance adjustment parameter to obtain the adjusted customer account data includes: adjusting the customer account data according to the insurance adjustment parameter of the insurance package with a complaint history to obtain the customer account data with records; accordingly, training the first neural network model according to the adjusted customer account data to obtain a second neural network model; training the first neural network model according to the customer account data with records to obtain a second neural network model.
[0019] In one possible design, the complaint verification of the insurance package number of the customer account data to obtain the insurance adjustment parameters of the insurance package with a complaint history includes: collecting a verification script data set corresponding to the customer account data, wherein the verification script data set includes multiple verification scripts, each verification script is used to identify the customer who generated the complaint, the verification rule type and the insurance type; determining the number of verification rule types and the number of insurance types with complaints for each customer according to the verification script data set; calculating the total number of verification rule types and the number of insurance types with complaints for each customer; judging whether the total number exceeds a preset threshold; if the total number does not exceed the preset threshold, determining that the customer is an invalid complaint, and storing the customer account data corresponding to the customer in a preset negative database; if the total number exceeds the preset threshold, determining that the customer is a valid complaint, and determining the insurance adjustment parameters of the insurance package with a complaint history according to the customer account data corresponding to the customer.
[0020] In a second aspect, the present application provides an insurance package push device, which is applied to a server, including:
[0021] An acquisition module, used to acquire customer account data and build a first neural network model;
[0022] a determination module, configured to determine an insurance adjustment parameter, and adjust the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data;
[0023] A training module, configured to train the first neural network model according to the adjusted customer account data to obtain a second neural network model;
[0024] A generating module, for generating an insurance package request in response to a triggering operation of any customer on the client end;
[0025] An input module, used for inputting the insurance package request into the second neural network model to generate an insurance package number to be verified;
[0026] A verification module, used to determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package;
[0027] A push module is used to push the insurance package to the client if the insurance package passes the validity verification.
[0028] In a third aspect, the present application provides a server, comprising: at least one processor and a memory;
[0029] The memory stores computer-executable instructions;
[0030] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the insurance package pushing method described in the first aspect and various possible designs of the first aspect.
[0031] In a fourth aspect, the present application provides a computer storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the insurance package push method described in the first aspect and various possible designs of the first aspect is implemented.
[0032] The insurance package push method, device, server and storage medium provided in the present application obtain customer account data and construct a first neural network model; determine insurance adjustment parameters, and adjust the customer account data according to the insurance adjustment parameters to obtain adjusted customer account data; train the first neural network model according to the adjusted customer account data to obtain a second neural network model; generate an insurance package request in response to a trigger operation of any customer on a client; input the insurance package request into the second neural network model to generate an insurance package number to be verified; determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package; if the insurance package passes the validity verification, push the insurance package to the client, so that the insurance package can meet the personalized needs of the customer and improve the customer experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0034] Figure 1 A schematic diagram of an application scenario of the insurance package push method provided in an embodiment of the present application;
[0035] Figure 2 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 1 ;
[0036] Figure 3 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 2 ;
[0037] Figure 4 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 3 ;
[0038] Figure 5 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 4 ;
[0039] Figure 6 A schematic diagram of the structure of the insurance package push device provided in the embodiment of the present application;
[0040] Figure 7 A schematic diagram of the hardware structure of the server provided in an embodiment of the present application. DETAILED DESCRIPTION
[0041] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0042] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of financial data or user data and other information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals. It should be noted that in the embodiments of this application, some existing solutions in the industry such as certain software, components, models, etc. may be mentioned, and they should be considered as exemplary. Their purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.
[0043] An insurance package refers to a combination product that includes multiple types of insurance, which is designed to meet the different needs of customers and simplify the insurance purchase process. Insurance packages are usually designed by insurance companies based on customer needs and risk conditions. They may include various types of insurance such as life insurance, medical insurance, and accident insurance. They have the characteristics of comprehensive protection, convenience, and preferential prices. Therefore, the correct choice of insurance packages is particularly important. In the prior art, the traditional insurance package push method is to select an insurance package and complete the purchase of the insurance package by the user's spontaneous login to the insurance platform. However, in the prior art, the selection of insurance packages by the user's spontaneous login to the insurance platform may result in the insurance package not meeting the personalized needs of the customer, thereby reducing the customer experience.
[0044] In order to solve the above technical problems, the embodiments of the present application propose the following technical concepts: the inventor takes into account the trained second neural network model, and based on the insurance package request generated by the client, uses the second neural network model to process the insurance package request to generate an insurance package number; if the insurance package corresponding to the insurance package number passes the validity verification, the insurance package is pushed to the client, so that the insurance package can meet the customer's personalized needs and improve the customer experience.
[0045] Figure 1 It is a schematic diagram of an application scenario of the insurance package push method provided in an embodiment of the present application.
[0046] like Figure 1 As shown, the scenario includes: a client 101 and a server 102 .
[0047] The client 101 may be a display screen or a terminal such as a personal computer.
[0048] The server 102 may be an independent server or a cluster consisting of multiple servers.
[0049] The server 102 obtains customer account data and constructs a first neural network model; determines insurance adjustment parameters, and adjusts customer account data according to the insurance adjustment parameters to obtain adjusted customer account data; trains the first neural network model according to the adjusted customer account data to obtain a second neural network model; generates an insurance package request in response to any customer triggering operation on the client 101; inputs the insurance package request into the second neural network model to generate an insurance package number to be verified; determines the insurance package according to the insurance package number to be verified, and if it is determined that the insurance package passes the validity verification, pushes the insurance package to the client 101 for display. The following is a detailed description using a detailed embodiment.
[0050] Figure 2 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 1 , the execution subject of this embodiment can be Figure 1 The server in the embodiment shown is not particularly limited in this embodiment. Figure 2 As shown, the method includes:
[0051] S201: Obtain customer account data and build a first neural network model.
[0052] In this embodiment, the first neural network model is obtained by constructing corresponding fully connected layers, convolutional layers, pooling layers, activation functions, batch normalization layers, and discard layers.
[0053] Specifically, obtaining customer account data in step S201 specifically includes:
[0054] S2011: Get the customer's insurance package address.
[0055] S2012: Retrieve the corresponding insurance package script unit according to each insurance package address, wherein the insurance package script unit includes one or more insurance package scripts.
[0056] S2013: Parse each insurance package script unit to obtain all insurance package scripts.
[0057] S2014: Determine the correlation coefficient of each insurance package script, and delete the insurance package scripts whose correlation coefficient exceeds a preset critical value to obtain the remaining insurance package scripts.
[0058] S2015: Determine the remaining insurance package scripts as customer account data.
[0059] In this embodiment, the remaining insurance package scripts are insurance package scripts required by the registered customer accounts.
[0060] S202: Determine insurance adjustment parameters, and adjust customer account data according to the insurance adjustment parameters to obtain adjusted customer account data.
[0061] S203: Train the first neural network model according to the adjusted customer account data to obtain a second neural network model.
[0062] Specifically, step S203 includes:
[0063] S2031: Divide the adjusted customer account data into a data set to obtain a training set, a validation set, and a test set.
[0064] S2032: Setting different training parameters for the training set, validation set, and test set according to the number of insurance package scripts in the adjusted customer account data.
[0065] S2033: Processing the first neural network model according to the training set, validation set and test set with different training parameters to obtain a second neural network model.
[0066] S204: Generate an insurance package request in response to a triggering operation of any customer on the client.
[0067] S205: Input the insurance package request into the second neural network model to generate an insurance package number to be verified.
[0068] Specifically, the insurance package request is input into the second neural network model for insurance package matching to obtain the number corresponding to the insurance package to be verified.
[0069] S206: Determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package.
[0070] Exemplarily, the corresponding insurance package is retrieved according to the insurance package number to be verified; the customer's credibility is obtained, and the feasibility and effectiveness of executing the insurance package under the customer's credibility are determined.
[0071] S207: If the insurance package passes the validity verification, the insurance package is pushed to the client.
[0072] Exemplarily, if the insurance package passes the validity verification, the insurance package is pushed to the customer's mobile phone for display.
[0073] In summary, the insurance package push method provided in this embodiment obtains customer account data and constructs a first neural network model; determines insurance adjustment parameters, and adjusts customer account data according to the insurance adjustment parameters to obtain adjusted customer account data; trains the first neural network model according to the adjusted customer account data to obtain a second neural network model; generates an insurance package request in response to a trigger operation of any customer on the client; inputs the insurance package request into the second neural network model to generate an insurance package number to be verified; determines the insurance package according to the insurance package number to be verified, and verifies the validity of the insurance package; if the insurance package passes the validity verification, pushes the insurance package to the client, so that the insurance package can meet the personalized needs of the customer and improve the customer experience.
[0074] Figure 3 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 2 In the embodiment of the present application, Figure 2 Based on the embodiment provided, the specific implementation method of determining the insurance adjustment parameters in step S202 and the subsequent steps is described in detail. The insurance adjustment parameters are a full set of insurance adjustment parameters, and the adjusted customer account data is the parameter selection instruction data, such as Figure 3 As shown, the method includes:
[0075] S301: Obtain customer account data and build a first neural network model.
[0076] In this embodiment, the discussion about the customer account data and the first neural network has been discussed in detail in step S201 and will not be repeated here.
[0077] S302: Construct a third neural network model, and input the customer account data into the third neural network model to obtain a full set of insurance adjustment parameters.
[0078] In this embodiment, the third neural network model is obtained by constructing corresponding fully connected layers, convolutional layers, pooling layers, activation functions, batch normalization layers, and discard layers.
[0079] S303: Determine the corresponding insurance package number according to the full set of insurance adjustment parameters, and generate corresponding parameter selection instruction data according to the insurance package number.
[0080] In this embodiment, the parameter selection instruction data generally refers to the behavior of using parameters in a command line to select or configure a command.
[0081] S304: Train the first neural network model according to the parameter selection instruction data to obtain a second neural network model.
[0082] Specifically, step S304 includes:
[0083] S3041: Divide the parameter selection instruction data into data sets to obtain a training set, a validation set, and a test set.
[0084] S3042: Adjust the first neural network model according to the training set to obtain an adjusted first neural network model.
[0085] S3043: Verify the adjusted first neural network model according to the verification set to determine whether the error requirement is met.
[0086] In this embodiment, the error requirement is to be less than a preset error threshold, where the error threshold can be any value among 5%, 10% or 15%, or other values.
[0087] S3044: If the adjusted first neural network model meets the error requirement, the adjusted first neural network model is tested according to the test set.
[0088] S3045: If the adjusted first neural network model passes the test, the adjusted first neural network model is determined as the second neural network model.
[0089] S305: Generate an insurance package request in response to a triggering operation of any customer on the client.
[0090] S306: Input the insurance package request into the second neural network model to generate an insurance package number to be verified.
[0091] In this embodiment, the discussion on generating the insurance package number to be verified has been described in detail in step S205 and will not be repeated here.
[0092] S307: Determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package.
[0093] In this embodiment, the discussion on validity verification has been described in detail in step S206 and will not be repeated here.
[0094] S308: If the insurance package passes the validity verification, the insurance package is pushed to the client.
[0095] In summary, the insurance package push method provided in this embodiment constructs a third neural network model and inputs the customer account data into the third neural network model to obtain a full set of insurance adjustment parameters; determines the corresponding insurance package number according to the full set of insurance adjustment parameters, and generates corresponding parameter selection instruction data according to the insurance package number; trains the first neural network model according to the parameter selection instruction data to obtain the second neural network model, and trains the neural network model through the parameter selection instruction data, so that the subsequently determined insurance package can meet the customer's personalized needs and improve the customer experience.
[0096] Figure 4 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 3 In the embodiment of the present application, Figure 2 Based on the provided embodiment, the specific implementation method of determining the insurance adjustment parameter in step S202 and the subsequent steps is described in detail. The insurance adjustment parameter is an explanation of the insurance adjustment parameter, and the adjusted customer account data is an adjustment data set, such as Figure 4 As shown, the method includes:
[0097] S401: Obtain customer account data and build a first neural network model.
[0098] In this embodiment, the discussion about the customer account data and the first neural network has been discussed in detail in step S201 and will not be repeated here.
[0099] S402: Classify the insurance package numbers of the customer account data into different levels to obtain insurance packages to be processed after classification.
[0100] S403: Determine corresponding explanation insurance adjustment parameters according to the to-be-processed insurance packages divided into different levels.
[0101] S404: Adjust the customer account data according to the insurance adjustment parameters of each explanation to obtain an adjustment data set.
[0102] S405: Train the first neural network model according to the adjusted data set to obtain a second neural network model.
[0103] In this embodiment, step S405 specifically includes:
[0104] S4051: Divide the adjusted data set into a training set, a validation set, and a test set.
[0105] S4052: Adjust the first neural network model according to the training set to obtain an adjusted first neural network model.
[0106] S4053: Verify the adjusted first neural network model according to the verification set to determine whether the error requirement is met.
[0107] In this embodiment, the error requirement is to be less than a preset error threshold, where the error threshold can be any value among 5%, 10% or 15%, or other values.
[0108] S4054: If the adjusted first neural network model meets the error requirement, the adjusted first neural network model is tested according to the test set.
[0109] S4055: If the adjusted first neural network model passes the test, the adjusted first neural network model is determined as the second neural network model.
[0110] S406: Generate an insurance package request in response to a triggering operation of any customer on the client.
[0111] S407: Input the insurance package request into the second neural network model to generate an insurance package number to be verified.
[0112] In this embodiment, the discussion on generating the insurance package number to be verified has been described in detail in step S205 and will not be repeated here.
[0113] S408: Determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package.
[0114] In this embodiment, the discussion on validity verification has been described in detail in step S206 and will not be repeated here.
[0115] S409: If the insurance package passes the validity verification, the insurance package is pushed to the client.
[0116] In summary, the insurance package push method provided in this embodiment divides the insurance package numbers of the customer account data into levels to obtain the insurance packages to be processed after each level division; determines the corresponding explanation insurance adjustment parameters according to the insurance packages to be processed after each level division; adjusts the customer account data according to the insurance adjustment parameters of each explanation to obtain an adjusted data set; trains a first neural network model according to the adjusted data set to obtain a second neural network model, and trains the neural network model by adjusting the data set, so that the insurance packages determined subsequently can meet the personalized needs of the customers and improve the customer experience.
[0117] Figure 5 Schematic diagram of the insurance package push method provided in the embodiment of the present application Figure 4 In the embodiment of the present application, Figure 2Based on the provided embodiment, the specific implementation method of determining the insurance adjustment parameter in step S202 and the subsequent steps is described in detail. The insurance adjustment parameter is the insurance adjustment parameter of the insurance package with the complaint history record, and the adjusted customer account data is the customer account data with the record, such as Figure 5 As shown, the method includes:
[0118] S501: Obtain customer account data and build a first neural network model.
[0119] In this embodiment, the discussion about the customer account data and the first neural network has been discussed in detail in step S201 and will not be repeated here.
[0120] S502: Complaint verification is performed on the insurance package number of the customer account data to obtain insurance adjustment parameters of the insurance package with complaint history.
[0121] Specifically, step S502 includes:
[0122] S5021: Collect a verification script data set corresponding to the customer account data, where the verification script data set includes multiple verification scripts, each verification script is used to identify the customer who generated the complaint, the type of verification rule, and the type of insurance.
[0123] Exemplarily, the types of verification rules include electronic policy verification rules, paper policy verification rules, online verification rules, official telephone verification rules, and other verification rules.
[0124] Exemplarily, the types of insurance include life insurance, property insurance, and other insurance.
[0125] S5022: Determine the number of verification rule types and insurance types for each customer with complaints based on the verification script data set.
[0126] S5023: Calculate the total number of verification rule types and insurance types for each customer with complaints.
[0127] S5024: Determine whether the total number exceeds a preset threshold.
[0128] In this embodiment, the preset threshold may be any value among 2, 3 or 5, or other values.
[0129] S5025: If the total number does not exceed the preset threshold, the customer's complaint is determined to be invalid, and the customer account data corresponding to the customer is stored in the preset negative database.
[0130] S5026: If the total number exceeds the preset threshold, the customer is judged to have made a valid complaint, and the insurance adjustment parameters of the insurance package with the complaint history are determined based on the customer account data corresponding to the customer.
[0131] S503: Adjust the customer account data according to the insurance adjustment parameters of the insurance package with the complaint history to obtain the customer account data with the record.
[0132] S504: Train the first neural network model according to the recorded customer account data to obtain a second neural network model.
[0133] In this embodiment, step S504 specifically includes:
[0134] S5041: Divide the recorded customer account data to obtain a training set, a validation set, and a test set.
[0135] S5042: Adjust the first neural network model according to the training set to obtain an adjusted first neural network model.
[0136] S5043: Verify the adjusted first neural network model according to the verification set to determine whether the error requirement is met.
[0137] In this embodiment, the error requirement is to be less than a preset error threshold, where the error threshold can be any value among 5%, 10% or 15%, or other values.
[0138] S5044: If the adjusted first neural network model meets the error requirement, the adjusted first neural network model is tested according to the test set.
[0139] S5045: If the adjusted first neural network model passes the test, the adjusted first neural network model is determined as the second neural network model.
[0140] S505: Generate an insurance package request in response to a triggering operation of any customer on the client.
[0141] S506: Input the insurance package request into the second neural network model to generate an insurance package number to be verified.
[0142] In this embodiment, the discussion on generating the insurance package number to be verified has been described in detail in step S205 and will not be repeated here.
[0143] S507: Determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package.
[0144] In this embodiment, the discussion on validity verification has been described in detail in step S206 and will not be repeated here.
[0145] S508: If the insurance package passes the validity verification, the insurance package is pushed to the client.
[0146] In summary, the insurance package push method provided in this embodiment performs complaint verification on the insurance package number of the customer account data to obtain insurance adjustment parameters of the insurance package with a complaint history record; adjusts the customer account data according to the insurance adjustment parameters of the insurance package with the complaint history record to obtain customer account data with records; trains a first neural network model according to the customer account data with records to obtain a second neural network model, and trains the neural network model by using the customer account data with records, so that the subsequently determined insurance package can meet the customer's personalized needs and improve the customer experience.
[0147] In addition, the insurance package push method provided in this embodiment can determine whether the customer's complaint is invalid by judging whether the total number exceeds a preset threshold, thereby screening the customers to a certain extent and improving the data quality of the customer account data, so that the subsequently determined insurance packages can more accurately meet the customer's personalized needs.
[0148] It should be noted that, in the insurance package push method provided by the present application, step S502 can also be specifically as follows: calculate the data volume of the negative type identified by each verification script in the verification script data set to obtain the negative data volume, where the customer information identified by each verification script in the verification script data set is the customer ID and the customer number of the denied object identified by each verification script is the customer number of the denied object; calculate the data volume of the verification rule type identified by each verification script in the verification script data set to obtain the verification rule data volume; construct a negative database based on the negative data volume and the verification rule data volume, and when a negative database exists, the customer information is the customer ID of the primary invalid negative action.
[0149] Figure 6 This is a schematic diagram of the structure of the insurance package push device provided in the embodiment of the present application. Figure 6 As shown, the insurance package pushing device includes: an acquisition module 601, a determination module 602, a training module 603, a generation module 604, an input module 605, a verification module 606 and a pushing module 607.
[0150] An acquisition module 601 is used to acquire customer account data and construct a first neural network model;
[0151] A determination module 602, configured to determine an insurance adjustment parameter, and adjust the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data;
[0152] A training module 603, configured to train the first neural network model according to the adjusted customer account data to obtain a second neural network model;
[0153] A generating module 604, for generating an insurance package request in response to a triggering operation of any customer on the client;
[0154] An input module 605, configured to input the insurance package request into the second neural network model to generate an insurance package number to be verified;
[0155] A verification module 606, configured to determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package;
[0156] The push module 607 is used to push the insurance package to the client if the insurance package passes the validity verification.
[0157] In a possible implementation, the acquisition module 601 specifically includes:
[0158] The acquisition unit 6011 is used to acquire the insurance package address of the customer;
[0159] A retrieving unit 6012 is used to retrieve a corresponding insurance package script unit according to each insurance package address, wherein the insurance package script unit includes one or more insurance package scripts;
[0160] The parsing unit 6013 is used to parse each insurance package script unit to obtain all insurance package scripts;
[0161] The first determining unit 6014 is used to determine the correlation coefficient of each insurance package script, and delete the insurance package script whose correlation coefficient exceeds a preset critical value to obtain the remaining insurance package scripts;
[0162] The second determining unit 6015 is used to determine the remaining insurance package scripts as customer account data.
[0163] In a possible implementation, the insurance adjustment parameter is a full set of insurance adjustment parameters, and the adjusted customer account data is parameter selection instruction data; the determination module 602 specifically includes:
[0164] A construction unit 6021 is used to construct a third neural network model and input the customer account data into the third neural network model to obtain a full set of insurance adjustment parameters;
[0165] The first determining unit 6022 is used to determine the corresponding insurance package number according to the full set of insurance adjustment parameters, and generate corresponding parameter selection instruction data according to the insurance package number;
[0166] Accordingly, the training module 603 specifically includes:
[0167] The training unit 6031 is used to train the first neural network model according to the parameter selection instruction data to obtain a second neural network model.
[0168] In a possible implementation, the training unit 6031 specifically includes:
[0169] A division unit 60311 is used to divide the parameter selection instruction data into a data set to obtain a training set, a validation set and a test set;
[0170] An adjusting unit 60312, configured to adjust the first neural network model according to the training set to obtain an adjusted first neural network model;
[0171] A verification unit 60313 is used to verify the adjusted first neural network model according to a verification set to determine whether the error requirement is met;
[0172] A testing unit 60314 is used to test the adjusted first neural network model according to a test set if the adjusted first neural network model meets the error requirement;
[0173] The determination unit 60315 is configured to determine the adjusted first neural network model as the second neural network model if the adjusted first neural network model passes the test.
[0174] In a possible implementation, the insurance adjustment parameter is an explanation insurance adjustment parameter, and the adjusted customer account data is an adjustment data set; the determination module 602 specifically includes:
[0175] A classification unit 6023 is used to classify the insurance package numbers of the customer account data into different levels to obtain insurance packages to be processed after each level classification;
[0176] The second determining unit 6024 is used to determine the corresponding explanation insurance adjustment parameter according to the to-be-processed insurance packages divided into different levels;
[0177] A first adjustment unit 6025, configured to adjust the customer account data according to each explanation insurance adjustment parameter to obtain an adjustment data set;
[0178] Correspondingly, the training module 603 is specifically used to: train the first neural network model according to the adjustment data set to obtain a second neural network model.
[0179] In a possible implementation, the insurance adjustment parameter is an insurance adjustment parameter of an insurance package with a complaint history record, and the adjusted customer account data is customer account data with a record. The determination module 602 specifically includes:
[0180] A verification unit 6026, for performing complaint verification on the insurance package number of the customer account data to obtain insurance adjustment parameters of the insurance package with complaint history records;
[0181] The second adjustment unit 6027 is used to adjust the customer account data according to the insurance adjustment parameter of the insurance package with complaint history record to obtain the customer account data with record;
[0182] Correspondingly, the training module 603 is specifically used to: train the first neural network model according to the recorded customer account data to obtain a second neural network model.
[0183] In a possible implementation, the verification unit 6026 specifically includes:
[0184] A collection unit 60261 is used to collect a verification script data set corresponding to the customer account data, wherein the verification script data set includes multiple verification scripts, each verification script is used to identify the customer who generated the complaint, the type of verification rule, and the type of insurance;
[0185] The first determining unit 60262 is used to determine the number of verification rule types and the number of insurance types with complaints for each customer according to the verification script data set;
[0186] A calculation unit 60263 is used to calculate the total number of verification rule types and insurance types for each customer with complaints;
[0187] A judging unit 60264 is used to judge whether the total number exceeds a preset threshold;
[0188] The storage unit 60265 is used to determine that the customer's complaint is invalid if the total number does not exceed a preset threshold, and store the customer account data corresponding to the customer in a preset negative database;
[0189] The second determination unit 60266 is used to determine that the customer has made a valid complaint if the total number exceeds a preset threshold, and determine the insurance adjustment parameters of the insurance package with the complaint history record based on the customer account data corresponding to the customer.
[0190] The device provided in this embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be repeated here.
[0191] Figure 7 The hardware structure diagram of the server provided in the embodiment of the present application is shown in FIG. Figure 7 As shown, the server of this embodiment includes: a processor 701 and a memory 702; the memory stores computer-executable instructions; at least one processor executes the computer-executable instructions stored in the memory, so that at least one processor executes the above insurance package push method.
[0192] Optionally, the memory 702 may be independent or integrated with the processor 701 .
[0193] When the memory 702 is independently provided, the server further includes a bus 703 for connecting the memory 702 and the processor 701 .
[0194] The embodiment of the present application also provides a computer storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the insurance package push method as described above is implemented.
[0195] The embodiment of the present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the insurance package push method as described above is implemented.
[0196] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.
[0197] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to implement the solution of this embodiment.
[0198] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The above-mentioned module-composed unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
[0199] The above-mentioned integrated module implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform some steps of the method described in each embodiment of the present application.
[0200] It should be understood that the above processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the invention can be directly embodied as being executed by a hardware processor, or can be executed by a combination of hardware and software modules in the processor.
[0201] The memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory, and may also be a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk or an optical disk, etc.
[0202] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.
[0203] The above storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The storage medium may be any available medium that can be accessed by a general or special purpose computer.
[0204] An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the storage medium can also exist as discrete components in an electronic device or a main control device.
[0205] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.
[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.
Claims
1. A method for pushing an insurance package, characterized in that: Applicable to servers, including: Obtain customer account data and build the first neural network model; determining an insurance adjustment parameter, and adjusting the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data; Training the first neural network model according to the adjusted customer account data to obtain a second neural network model; In response to a triggering operation of any customer on the client, generating an insurance package request; Inputting the insurance package request into the second neural network model to generate an insurance package number to be verified; Determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package; If the insurance package passes the validity verification, the insurance package is pushed to the client.
2. The method according to claim 1, characterized in that The obtaining of customer account data includes: Get the customer's insurance package address; Retrieving the corresponding insurance package script unit according to each insurance package address, wherein the insurance package script unit includes one or more insurance package scripts; Parse each insurance package script unit to obtain all insurance package scripts; Determine the correlation coefficient of each insurance package script, and delete the insurance package scripts whose correlation coefficient exceeds a preset critical value to obtain the remaining insurance package scripts; The remaining insurance package scripts are determined as customer account data.
3. The method according to claim 1, characterized in that The insurance adjustment parameters are a full set of insurance adjustment parameters, and the adjusted customer account data are parameter selection instruction data; the insurance adjustment parameters are determined, including: Constructing a third neural network model, and inputting the customer account data into the third neural network model to obtain a full set of insurance adjustment parameters; Accordingly, the step of adjusting the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data includes: Determine the corresponding insurance package number according to the full set of insurance adjustment parameters, and generate corresponding parameter selection instruction data according to the insurance package number; Accordingly, training the first neural network model according to the adjusted customer account data to obtain a second neural network model includes: The first neural network model is trained according to the parameter selection instruction data to obtain a second neural network model.
4. The method according to claim 3, characterized in that The step of training the first neural network model according to the parameter selection instruction data to obtain a second neural network model includes: Dividing the parameter selection instruction data into a data set to obtain a training set, a validation set and a test set; Adjusting the first neural network model according to the training set to obtain an adjusted first neural network model; Verifying the adjusted first neural network model according to the verification set to determine whether the error requirement is met; If the adjusted first neural network model meets the error requirement, testing the adjusted first neural network model according to the test set; If the adjusted first neural network model passes the test, the adjusted first neural network model is determined as the second neural network model.
5. The method according to any one of claims 1 to 4, characterized in that: The insurance adjustment parameter is an explanation insurance adjustment parameter, and the adjusted customer account data is an adjustment data set; and determining the insurance adjustment parameter includes: Classifying the insurance package numbers of the customer account data into different levels to obtain insurance packages to be processed after each level classification; Determine corresponding explanation insurance adjustment parameters according to the to-be-processed insurance packages divided into various levels; Accordingly, the step of adjusting the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data includes: Adjusting the customer account data according to each interpretation insurance adjustment parameter to obtain an adjusted data set; Accordingly, the first neural network model is trained according to the adjusted customer account data to obtain a second neural network model, including: The first neural network model is trained according to the adjusted data set to obtain a second neural network model.
6. The method according to claim 1, characterized in that The insurance adjustment parameter is an insurance adjustment parameter of an insurance package with a complaint history record, the adjusted customer account data is customer account data with a record, and determining the insurance adjustment parameter includes: Performing complaint verification on the insurance package number of the customer account data to obtain insurance adjustment parameters of the insurance package with complaint history records; Accordingly, the step of adjusting the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data includes: Adjusting the customer account data according to the insurance adjustment parameters of the insurance package with complaint history records to obtain customer account data with records; Accordingly, the first neural network model is trained according to the adjusted customer account data to obtain a second neural network model; The first neural network model is trained according to the recorded customer account data to obtain a second neural network model.
7. The method according to claim 6, characterized in that The complaint verification of the insurance package number of the customer account data to obtain the insurance adjustment parameter of the insurance package with the complaint history record includes: Collecting a verification script data set corresponding to the customer account data, wherein the verification script data set includes a plurality of verification scripts, each verification script being used to identify a customer who has generated a complaint, a verification rule type, and an insurance type; Determine the number of verification rule types and the number of insurance types for each customer with complaints based on the verification script data set; Calculate the total number of verification rule types and insurance types with complaints for each customer; Determining whether the total quantity exceeds a preset threshold; If the total number does not exceed the preset threshold, the customer is determined to have made an invalid complaint, and the customer account data corresponding to the customer is stored in a preset negative database; If the total number exceeds a preset threshold, the customer is determined to have made a valid complaint, and the insurance adjustment parameters of the insurance package with the complaint history are determined based on the customer account data corresponding to the customer.
8. An insurance package push device, characterized in that: Applicable to servers, including: An acquisition module, used to acquire customer account data and build a first neural network model; a determination module, configured to determine an insurance adjustment parameter, and adjust the customer account data according to the insurance adjustment parameter to obtain adjusted customer account data; A training module, configured to train the first neural network model according to the adjusted customer account data to obtain a second neural network model; A generating module, for generating an insurance package request in response to a triggering operation of any customer on the client end; An input module, used for inputting the insurance package request into the second neural network model to generate an insurance package number to be verified; A verification module, used to determine the insurance package according to the insurance package number to be verified, and verify the validity of the insurance package; A push module is used to push the insurance package to the client if the insurance package passes the validity verification.
9. A server, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the insurance package pushing method as described in any one of claims 1 to 7.
10. A computer storage medium, characterized in that: The computer storage medium stores computer execution instructions, and when the processor executes the computer execution instructions, the insurance package push method as described in any one of claims 1 to 7 is implemented.