Insurance renewal prediction method and device, electronic equipment and storage medium
By obtaining user feature data and using pre-trained renewal prediction model to predict, the problems of cumbersome processing of insurance renewal prediction data and large deviations in the estimation results in the prior art are solved, and the accuracy of renewal prediction and the effectiveness of business strategies are improved.
Patent Information
- Application Number
- CN202510124455.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-26
- Publication Date
- 2025-05-23
AI Technical Summary
In the insurance renewal forecast, the prior art has problems such as cumbersome data processing, complex calculation logic, easy software crash, slow processing speed and large deviations in the estimation results, resulting in insufficient accuracy of renewal forecasts.
By obtaining user feature data, using a pre-trained renewal prediction model (based on neural network model, obtained through historical transaction data training), it automatically counts the policy renewal rate, refund rate and other indicators of relevant transaction users with unique attributes such as different age groups, genders, and regions, and uses a large model to predict future effects based on actual statistical effects.
Improve the accuracy of insurance renewal forecasting, and through automated data statistics and large-scale prediction, better business direction strategies are provided, reducing the deviation of manual estimation.
Smart Images
Figure CN120031668A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and more specifically, to a method, device, electronic device, and storage medium for predicting insurance renewal. Background Art
[0002] The calculation of profitable premiums in the insurance business has different profit settlement methods based on different product characteristics and sales links. Users of different age groups, genders, and regions have different indicators such as renewal rates, surrender rates, and savings rates for related policies. Large quantities of policies use office tools to calculate relevant indicators based on detailed statistics. Half of the data output needs to be manually implemented through office software such as WPS. The large amount of data and complex calculation logic will cause software crashes and slow processing speeds. Estimation can only be made based on a small number of limited rules and experience, and the estimation results are highly biased. Therefore, how to improve the accuracy of insurance renewal predictions is an urgent problem that needs to be solved. Summary of the Invention
[0003] The purpose of some embodiments of the present application is to provide a method, device, electronic device and storage medium for predicting insurance renewal. Through the technical solutions of the embodiments of the present application, user feature data is obtained; based on a pre-trained renewal prediction model, a prediction result corresponding to the user feature data is determined; wherein, the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data; based on the prediction result, an adjustment strategy corresponding to the prediction result is determined. The embodiments of the present application automatically count the renewal rate, surrender rate and other indicators of the policies of relevant transaction users with unique attributes such as different age groups, genders, and regions through code according to different products, sales links and transaction user characteristics, and based on actual statistical results, a large model is used to predict future effects to provide better direction strategies for the business, thereby improving the accuracy of insurance renewal predictions.
[0004] In a first aspect, some embodiments of the present application provide a method for predicting insurance renewal, including:
[0005] Obtain user feature data;
[0006] Determining a prediction result corresponding to the user feature data based on a pre-trained renewal prediction model; wherein the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data;
[0007] According to the prediction result, an adjustment strategy corresponding to the prediction result is determined.
[0008] Some embodiments of the present application use code to automatically calculate the renewal rate, surrender rate and other indicators of the policies of relevant transaction users with unique attributes such as different age groups, genders, and regions based on different products, sales links and transaction user characteristics. Based on the actual statistical results, a large model is used to predict future effects to provide the business with better direction strategies and improve the accuracy of insurance renewal predictions.
[0009] Optionally, determining an adjustment strategy corresponding to the prediction result according to the prediction result includes:
[0010] If the prediction result does not meet the preset profit requirements, the intermediate indicators are adjusted, wherein the intermediate indicators at least include the connection rate, the connection transaction rate and the product price for each age group.
[0011] Some embodiments of the present application compare the prediction results with the preset profit requirements. If the prediction results do not meet the profit requirements, the intermediate indicators, namely the connection rate, the connection transaction rate and the product prices for each age group, are adjusted, and then predictions are made based on the adjusted indicators until the profit requirements are met.
[0012] Optionally, the pre-trained renewal prediction model is obtained by:
[0013] Acquire the historical transaction data, the historical transaction data including at least user operation data and profit data, the user operation data including at least sales records and user order data, and the profit data including at least user payment record data;
[0014] Performing feature extraction on the historical transaction data to obtain a feature vector corresponding to the historical transaction data;
[0015] Training the neural network model according to the feature vector to obtain training results, wherein the training results at least include renewal rate, cancellation rate and surrender rate;
[0016] The renewal prediction model is determined based on the training results and the preset transaction rate, wherein the preset transaction rate at least includes a preset renewal rate, a preset cancellation rate and a preset surrender rate.
[0017] Some embodiments of the present application use historical transaction data to train a neural network model to obtain a renewal prediction model, and use a large model to predict future effects to provide better direction strategies for the business.
[0018] Optionally, determining the renewal prediction model according to the training result and a preset transaction rate includes:
[0019] If the difference between the renewal rate and the preset renewal rate is less than a first preset value, the difference between the cancellation rate and the preset cancellation rate is less than a second preset value, and the difference between the surrender rate and the preset surrender rate is less than a third preset value, the neural network model corresponding to the training result is determined as the renewal prediction model.
[0020] Optionally, the preset transaction rate is obtained by:
[0021] Determine the actual premiums collected, cancellation data, and surrender data for the user's insurance policy within a preset time period based on the historical transaction data and the preset renewal rate rules;
[0022] The preset renewal rate, the preset cancellation rate and the preset surrender rate are determined based on the actually collected premiums, the cancellation data, the surrender data and the total premiums.
[0023] Some embodiments of the present application process historical transaction data and preset renewal rate rules to obtain preset transaction rates, namely, preset renewal rates, preset cancellation rates and preset surrender rates, which are used to compare the training results in model training and improve the accuracy of model training.
[0024] In a second aspect, some embodiments of the present application provide an insurance renewal prediction device, including:
[0025] Acquisition module, used to obtain user feature data;
[0026] a determination module, configured to determine a prediction result corresponding to the user characteristic data based on a pre-trained renewal prediction model; wherein the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data;
[0027] The prediction module is used to determine an adjustment strategy corresponding to the prediction result based on the prediction result.
[0028] Some embodiments of the present application use code to automatically calculate the renewal rate, surrender rate and other indicators of the policies of relevant transaction users with unique attributes such as different age groups, genders, and regions based on different products, sales links and transaction user characteristics. Based on the actual statistical results, a large model is used to predict future effects to provide the business with better direction strategies and improve the accuracy of insurance renewal predictions.
[0029] Optionally, the prediction module is used to:
[0030] If the prediction result does not meet the preset profit requirements, the intermediate indicators are adjusted, wherein the intermediate indicators at least include the connection rate, the connection transaction rate and the product price for each age group.
[0031] Some embodiments of the present application compare the prediction results with the preset profit requirements. If the prediction results do not meet the profit requirements, the intermediate indicators, namely the connection rate, the connection transaction rate and the product prices for each age group, are adjusted, and then predictions are made based on the adjusted indicators until the profit requirements are met.
[0032] Optionally, the pre-trained renewal prediction model is obtained by:
[0033] Acquire the historical transaction data, the historical transaction data including at least user operation data and profit data, the user operation data including at least sales records and user order data, and the profit data including at least user payment record data;
[0034] Performing feature extraction on the historical transaction data to obtain a feature vector corresponding to the historical transaction data;
[0035] Training the neural network model according to the feature vector to obtain training results, wherein the training results at least include renewal rate, cancellation rate and surrender rate;
[0036] The renewal prediction model is determined based on the training results and the preset transaction rate, wherein the preset transaction rate at least includes a preset renewal rate, a preset cancellation rate and a preset surrender rate.
[0037] Some embodiments of the present application use historical transaction data to train a neural network model to obtain a renewal prediction model, and use a large model to predict future effects to provide better direction strategies for the business.
[0038] Optionally, determining the renewal prediction model according to the training result and a preset transaction rate includes:
[0039] If the difference between the renewal rate and the preset renewal rate is less than a first preset value, the difference between the cancellation rate and the preset cancellation rate is less than a second preset value, and the difference between the surrender rate and the preset surrender rate is less than a third preset value, the neural network model corresponding to the training result is determined as the renewal prediction model.
[0040] Optionally, the preset transaction rate is obtained by:
[0041] Determine the actual premiums collected, cancellation data, and surrender data for the user's insurance policy within a preset time period based on the historical transaction data and the preset renewal rate rules;
[0042] The preset renewal rate, the preset cancellation rate and the preset surrender rate are determined based on the actually collected premiums, the cancellation data, the surrender data and the total premiums.
[0043] Some embodiments of the present application process historical transaction data and preset renewal rate rules to obtain preset transaction rates, namely, preset renewal rates, preset cancellation rates and preset surrender rates, which are used to compare the training results in model training and improve the accuracy of model training.
[0044] In a third aspect, some embodiments of the present application provide an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the insurance renewal prediction method as described in any embodiment of the first aspect can be implemented.
[0045] In a fourth aspect, some embodiments of the present application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the insurance renewal prediction method as described in any embodiment of the first aspect.
[0046] In a fifth aspect, some embodiments of the present application provide a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, it can implement the insurance renewal prediction method as described in any embodiment of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions of some embodiments of the present application, the following is a brief introduction to the drawings required for use in some embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.
[0048] Figure 1 A flowchart of a method for predicting insurance renewal provided in an embodiment of the present application;
[0049] Figure 2 A flowchart of another insurance renewal prediction method provided in an embodiment of the present application;
[0050] Figure 3 A schematic diagram of the structure of an insurance renewal prediction device provided in an embodiment of the present application;
[0051] Figure 4 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0052] The technical solutions in some embodiments of the present application will be described below in conjunction with the drawings in some embodiments of the present application.
[0053] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.
[0054] The calculation of insurance business profit premium has different profit settlement methods according to different product characteristics and sales links. Users of different age groups, genders, and regions have different indicators such as renewal rate, surrender rate, and savings rate of related transaction policies. A large number of policies use office tools to calculate relevant indicators based on detailed statistics. Half of the data output needs to be manually implemented through office software such as wps. Large amounts of data and complex calculation logic will cause software crashes and slow processing speeds. Estimation can only be made based on a small number of limited rules and experience, and the estimation results have large deviations. Therefore, how to improve the accuracy of insurance renewal prediction is an urgent problem to be solved. In view of this, some embodiments of the present application provide an insurance renewal prediction method. A prediction method, the method comprising obtaining user feature data; determining a prediction result corresponding to the user feature data based on a pre-trained renewal prediction model; wherein the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data; and determining an adjustment strategy corresponding to the prediction result based on the prediction result. In the embodiment of the present application, based on different products, sales links and transaction user characteristics, the code is used to automatically count the renewal rate, surrender rate and other indicators of the policies of relevant transaction users with unique attributes such as different age groups, genders, and regions, and based on the actual statistical results, a large model is used to predict future results to provide better direction strategies for the business, thereby improving the accuracy of insurance renewal predictions.
[0055] like Figure 1 As shown, an embodiment of the present application provides a method for predicting insurance renewal, the method comprising:
[0056] S101, obtaining user feature data;
[0057] Specifically, the terminal platform, i.e., the terminal device, obtains user characteristic data, which may be input by the user or obtained from a third-party platform. The user characteristic data includes at least product attributes or user attributes. User attributes include at least the user's age group, and product attributes include at least the product name and product price.
[0058] S102. Determine a prediction result corresponding to the user feature data based on a pre-trained renewal prediction model; wherein the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data;
[0059] A renewal prediction model is pre-stored on the terminal device. The renewal prediction model is used to predict whether the next cycle will be renewed based on the product name or user attributes. The renewal prediction model is obtained by training a neural network model using historical transaction data.
[0060] After acquiring the user characteristic data, the terminal device converts the user characteristic data into a feature vector, inputs the feature vector into the renewal prediction model, and obtains a prediction result corresponding to the user characteristic data, which at least includes profit data.
[0061] S103: Determine an adjustment strategy corresponding to the prediction result based on the prediction result.
[0062] Specifically, after obtaining the prediction result, the terminal device judges the prediction result. If the prediction result does not meet the profit standard, an adjustment strategy corresponding to the prediction result is determined, for example, the parameters in the prediction process need to be adjusted.
[0063] Some embodiments of the present application use code to automatically calculate the renewal rate, surrender rate and other indicators of the policies of relevant transaction users with unique attributes such as different age groups, genders, and regions based on different products, sales links and transaction user characteristics. Based on the actual statistical results, a large model is used to predict future effects to provide the business with better direction strategies and improve the accuracy of insurance renewal predictions.
[0064] Another embodiment of the present application further supplements the insurance renewal prediction method provided in the above embodiment.
[0065] Optionally, determining an adjustment strategy corresponding to the prediction result according to the prediction result includes:
[0066] If the forecast result does not meet the preset profit requirements, the intermediate indicators will be adjusted, where the intermediate indicators include at least the connection rate, the connection transaction rate and the product prices for each age group.
[0067] Specifically, if the forecast results fall short of targets or even indicate a potential loss, adjustments will be needed to related intermediate indicators, such as the connection rate, the connection conversion rate, and product prices for different age groups. Once profitability is achieved, the adjustment strategy can be managed based on these intermediate indicators. For example, if the actual collection rate meets targets but the annualized premiums are low or the total collected premiums are insufficient, the premium specifications for the product can be selectively increased in subsequent operations, or the age of the affected users can be increased, as premiums for older users of the same product are higher than those for younger users. Strategies can be adjusted in advance based on the corresponding forecasts to provide proactive services to insured users, ensuring their best interests and generating profitability.
[0068] Some embodiments of the present application compare the prediction results with the preset profit requirements. If the prediction results do not meet the profit requirements, the intermediate indicators, namely the connection rate, the connection transaction rate and the product prices for each age group, are adjusted, and then predictions are made based on the adjusted indicators until the profit requirements are met.
[0069] Optionally, a pre-trained renewal prediction model is obtained as follows:
[0070] Obtain historical transaction data, which includes at least user operation data and profit data. User operation data includes at least sales records and user order data, and profit data includes at least user payment record data.
[0071] Specifically, the terminal device obtains user-related operational data and profit data. The operational data refers to the user's call records, sales records manually recorded by the agent, the user's age, gender, region, order submission records, and payment records. Profit data refers to the user's payment records, which are insurance policies paid. The corresponding policies have annualized premiums, renewal rates (the corresponding premiums are deducted monthly, and the number of user payment periods has a corresponding renewal rate indicator), corresponding expenditure costs, manual agent costs, and server costs. The gross profit and actual income are calculated by integration.
[0072] Perform feature extraction on historical transaction data to obtain feature vectors corresponding to the historical transaction data;
[0073] Training the neural network model based on the feature vector to obtain training results, where the training results include at least renewal rate, cancellation rate, and surrender rate;
[0074] The feature vectors include the number of user calls, the text and tags used during the user communication, the duration of the communication, and the feature vectors of user order submission and user payment.
[0075] The feature vector is input into the neural network model to obtain a training result. The neural network model at least includes a model obtained by training based on a machine learning network. The machine learning model includes a convolutional neural network model, a recurrent neural network model and other network models, which are not specifically limited in the embodiments of the present application.
[0076] In this embodiment of the application, based on the historical renewal rates of users of different ages, genders, regions, and behaviors, the fluctuation range of renewal rates during historical months, and the premiums of products for corresponding age groups, it is possible to predict the expected actual collection rate and actual premium collected by the transaction users in the next year. Then, based on the same rules for estimating related costs, it is possible to directly predict whether there will be profitable output in the next year.
[0077] By calculating actual profit indicators such as the renewal rate of insurance users, and based on the renewal rate rules and the actual annualized premiums of user policies, the actual collected premiums can be calculated. The actual profit indicator is the difference between the collected premiums and the actual income calculated from various costs.
[0078] A renewal prediction model is determined based on the training results and the preset transaction rate, wherein the preset transaction rate includes at least a preset renewal rate, a preset cancellation rate and a preset surrender rate.
[0079] In the embodiment of the present application, a neural network model is trained based on these features, and the neural network model will give predictions of subsequent renewal rates, cancellation rates, surrender rates and other indicators based on different user characteristics.
[0080] Some embodiments of the present application use historical transaction data to train a neural network model to obtain a renewal prediction model, and use a large model to predict future effects to provide better direction strategies for the business.
[0081] Optionally, based on the training results and the preset transaction rate, a renewal prediction model is determined, including:
[0082] If the difference between the renewal rate and the preset renewal rate is less than the first preset value, the difference between the cancellation rate and the preset cancellation rate is less than the second preset value, and the difference between the surrender rate and the preset surrender rate is less than the third preset value, the neural network model corresponding to the training result will be determined as the renewal prediction model.
[0083] In the embodiment of the present application, one or more of the training results are compared with the preset transaction rate, and a renewal prediction model is determined based on the comparison result.
[0084] Optionally, the preset transaction rate is obtained by:
[0085] Determine the actual premiums collected, cancellation data, and surrender data for user policies within a preset time period based on historical transaction data and preset renewal rate rules;
[0086] Based on the actual premiums collected, cancellation data, surrender data and total premiums, the preset renewal rate, preset cancellation rate and preset surrender rate are determined.
[0087] In the embodiment of the present application, the renewal rate, surrender rate and other related indicators after the insurance policy is completed for different users are generated based on the sales chain, sales product characteristics and user-related attributes, namely the preset renewal rate, preset cancellation rate and preset surrender rate;
[0088] The sales chain at least includes communication via telephone, WeChat, and text messages, and payment chains include WeChat, Alipay, bank cards, and other product payment chains. User-related attributes at least include the number of outbound calls, the number of connected calls, user name, age, region, gender, and user operation behaviors, such as entering an activation code, submitting an order, etc.
[0089] Some embodiments of the present application process historical transaction data and preset renewal rate rules to obtain preset transaction rates, namely, preset renewal rates, preset cancellation rates and preset surrender rates, which are used to compare the training results in model training and improve the accuracy of model training.
[0090] like Figure 2 As shown, the insurance renewal prediction method provided in the embodiment of the present application includes:
[0091] Step 1: The terminal platform stores outbound call records, tags, sales records, and user attributes in a data warehouse, i.e., a database; the Party A platform enters the activation code, submits the order, makes the payment, and sends the upgrade to the database.
[0092] Step 2: The terminal platform performs statistical analysis on the data received in the database to obtain the renewal rate, actual premium collected, and cost;
[0093] Step 3: The terminal platform determines whether it is profitable based on the analysis results. If it is profitable, it provides a profit forecast and then makes a judgment on the executed business to determine whether the business objectives need to be adjusted.
[0094] Step 4: If there is no profit, give the key indicators that can be adjusted in the middle according to the actual situation, that is, the intermediate indicators:
[0095] For example: 1. Connection rate, impact on call charges;
[0096] 2. Connection transaction rate and completed policies;
[0097] 3. User age group, impact on average transaction volume, annualized premium conversion rate, impact on online agent availability, the number of concurrent outbound calls required to adjust labor costs, and impact on technical service servers;
[0098] The terminal platform can modify intermediate indicators according to different business conditions, and then produce operational change strategies to achieve gross profit targets.
[0099] It should be noted that each implementable method in this embodiment can be implemented separately, or can be implemented in combination in any combination without conflict, and this application does not limit it.
[0100] Another embodiment of the present application provides an insurance renewal prediction device for executing the insurance renewal prediction method provided in the above embodiment.
[0101] like Figure 3 FIG. 3 is a schematic diagram of the structure of the insurance renewal prediction device provided in an embodiment of the present application. The insurance renewal prediction device includes an acquisition module 301, a determination module 302, and a prediction module 303, wherein:
[0102] The acquisition module 301 is used to obtain user feature data;
[0103] The determination module 302 is used to determine a prediction result corresponding to the user characteristic data based on a pre-trained renewal prediction model; wherein the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data;
[0104] The prediction module 303 is used to determine an adjustment strategy corresponding to the prediction result based on the prediction result.
[0105] Some embodiments of the present application use code to automatically calculate the renewal rate, surrender rate and other indicators of the policies of relevant transaction users with unique attributes such as different age groups, genders, and regions based on different products, sales links and transaction user characteristics. Based on the actual statistical results, a large model is used to predict future effects to provide the business with better direction strategies and improve the accuracy of insurance renewal predictions.
[0106] Regarding the device in this embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method and will not be elaborated here.
[0107] Another embodiment of the present application further supplements the insurance renewal prediction device provided in the above embodiment.
[0108] Optionally, a prediction module is used to:
[0109] If the forecast result does not meet the preset profit requirements, the intermediate indicators will be adjusted, where the intermediate indicators include at least the connection rate, the connection transaction rate and the product prices for each age group.
[0110] Some embodiments of the present application compare the prediction results with the preset profit requirements. If the prediction results do not meet the profit requirements, the intermediate indicators, namely the connection rate, the connection transaction rate and the product prices for each age group, are adjusted, and then predictions are made based on the adjusted indicators until the profit requirements are met.
[0111] Optionally, a pre-trained renewal prediction model is obtained as follows:
[0112] Obtain historical transaction data, which includes at least user operation data and profit data. User operation data includes at least sales records and user order data, and profit data includes at least user payment record data.
[0113] Perform feature extraction on historical transaction data to obtain feature vectors corresponding to the historical transaction data;
[0114] Training the neural network model based on the feature vector to obtain training results, where the training results include at least renewal rate, cancellation rate, and surrender rate;
[0115] A renewal prediction model is determined based on the training results and the preset transaction rate, wherein the preset transaction rate includes at least a preset renewal rate, a preset cancellation rate and a preset surrender rate.
[0116] Some embodiments of the present application use historical transaction data to train a neural network model to obtain a renewal prediction model, and use a large model to predict future effects to provide better direction strategies for the business.
[0117] Optionally, based on the training results and the preset transaction rate, a renewal prediction model is determined, including:
[0118] If the difference between the renewal rate and the preset renewal rate is less than the first preset value, the difference between the cancellation rate and the preset cancellation rate is less than the second preset value, and the difference between the surrender rate and the preset surrender rate is less than the third preset value, the neural network model corresponding to the training result will be determined as the renewal prediction model.
[0119] Optionally, the preset transaction rate is obtained by:
[0120] Determine the actual premiums collected, cancellation data, and surrender data for user policies within a preset time period based on historical transaction data and preset renewal rate rules;
[0121] Based on the actual premiums collected, cancellation data, surrender data and total premiums, the preset renewal rate, preset cancellation rate and preset surrender rate are determined.
[0122] Some embodiments of the present application process historical transaction data and preset renewal rate rules to obtain preset transaction rates, namely, preset renewal rates, preset cancellation rates and preset surrender rates, which are used to compare the training results in model training and improve the accuracy of model training.
[0123] Regarding the device in this embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method and will not be elaborated here.
[0124] It should be noted that each implementable method in this embodiment can be implemented separately, or can be implemented in combination in any combination without conflict, and this application does not limit it.
[0125] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the operation of the method corresponding to any embodiment of the insurance renewal prediction method provided in the above embodiments can be implemented.
[0126] An embodiment of the present application also provides a computer program product, which includes a computer program, wherein when the computer program is executed by a processor, it can implement the operations corresponding to any embodiment of the insurance renewal prediction method provided in the above embodiments.
[0127] like Figure 4 As shown, some embodiments of the present application provide an electronic device 400, which includes: a memory 410, a processor 420, and a computer program stored in the memory 410 and executable on the processor 420, wherein the processor 420 reads the program from the memory 410 through the bus 430 and executes the program to implement a method of any embodiment included in the above-mentioned insurance renewal prediction method.
[0128] Processor 420 can process digital signals and can include various computing architectures, such as a complex instruction set computer architecture, a reduced instruction set computer architecture, or an architecture that implements a combination of multiple instruction sets. In some examples, processor 420 can be a microprocessor.
[0129] The memory 410 can be used to store instructions executed by the processor 420 or data related to the execution of instructions. These instructions and / or data may include code for implementing some or all functions of one or more modules described in the embodiments of this application. The processor 420 of the embodiment of the present disclosure can be used to execute the instructions in the memory 410 to implement the method shown above. The memory 410 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memory known to those skilled in the art.
[0130] The above are merely examples of the present application and are not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application. It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined or explained in subsequent figures.
[0131] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.
[0132] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.
Claims
1. A method for predicting insurance renewal, characterized in that: The method comprises: Obtain user feature data; Determine a prediction result corresponding to the user feature data according to a pre-trained renewal prediction model; wherein the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data; According to the prediction result, an adjustment strategy corresponding to the prediction result is determined.
2. The insurance renewal prediction method according to claim 1, characterized in that: The step of determining an adjustment strategy corresponding to the prediction result according to the prediction result includes: If the prediction result does not meet the preset profit requirement, the intermediate indicators are adjusted, wherein the intermediate indicators at least include the connection rate, the connection transaction rate and the product price for each age group.
3. The insurance renewal prediction method according to claim 1, characterized in that: The pre-trained renewal prediction model is obtained in the following way: Acquire the historical transaction data, wherein the historical transaction data at least includes user operation data and profit data, wherein the user operation data at least includes sales records and user order data, and the profit data at least includes user payment record data; Performing feature extraction on the historical transaction data to obtain a feature vector corresponding to the historical transaction data; According to the feature vector, the neural network model is trained to obtain training results, wherein the training results at least include renewal rate, cancellation rate and surrender rate; The renewal prediction model is determined based on the training results and the preset transaction rate, wherein the preset transaction rate at least includes a preset renewal rate, a preset cancellation rate and a preset surrender rate.
4. The insurance renewal prediction method according to claim 3, characterized in that: Determining the renewal prediction model according to the training result and the preset transaction rate includes: If the difference between the renewal rate and the preset renewal rate is less than the first preset value, the difference between the termination rate and the preset termination rate is less than the second preset value, and the difference between the surrender rate and the preset surrender rate is less than the third preset value, the neural network model corresponding to the training result is determined as the renewal prediction model.
5. The insurance renewal prediction method according to claim 3, characterized in that: The preset transaction rate is obtained in the following manner: Determine the actual premiums collected, cancellation data, and policy surrender data of the user's policy within a preset time period based on the historical transaction data and the preset renewal rate rules; The preset renewal rate, the preset cancellation rate and the preset surrender rate are determined based on the actually collected premiums, the termination data, the policy surrender data and the total premiums.
6. An insurance renewal prediction device, characterized in that: The device comprises: An acquisition module, used to acquire user feature data; A determination module, used to determine a prediction result corresponding to the user characteristic data according to a pre-trained renewal prediction model; wherein the pre-trained renewal prediction model is obtained by training a neural network model using historical transaction data; The prediction module is used to determine an adjustment strategy corresponding to the prediction result according to the prediction result.
7. The insurance renewal prediction device according to claim 6, characterized in that: The prediction module is used to: If the prediction result does not meet the preset profit requirement, the intermediate indicators are adjusted, wherein the intermediate indicators at least include the connection rate, the connection transaction rate and the product price for each age group.
8. The insurance renewal prediction device according to claim 6, characterized in that: The pre-trained renewal prediction model is obtained in the following way: Acquire the historical transaction data, wherein the historical transaction data at least includes user operation data and profit data, wherein the user operation data at least includes sales records and user order data, and the profit data at least includes user payment record data; Performing feature extraction on the historical transaction data to obtain a feature vector corresponding to the historical transaction data; According to the feature vector, the neural network model is trained to obtain training results, wherein the training results at least include renewal rate, cancellation rate and surrender rate; The renewal prediction model is determined based on the training results and the preset transaction rate, wherein the preset transaction rate at least includes a preset renewal rate, a preset cancellation rate and a preset surrender rate.
9. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor can implement the insurance renewal prediction method as claimed in any one of claims 1 to 5 when executing the program.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the program is executed by a processor, the insurance renewal prediction method described in any one of claims 1 to 5 can be implemented.