A method, apparatus, device and medium for renewal prediction

Abstract

The present application relates to the technical field of artificial intelligence, and particularly relates to a renewal prediction method and device, equipment and medium. For any target policy, user information in the target policy is extracted, historical program operation behaviors of a user corresponding to the target policy are obtained from an insurance application program according to the user information, operation behaviors of the user corresponding to the target policy after starting the insurance application program next time are predicted according to the historical program operation behaviors, a behavior prediction result is obtained, and a renewal behavior of the target policy is predicted according to the historical program operation behaviors and the behavior prediction result, so as to obtain a renewal prediction result. The operation behaviors after starting the insurance application program next time are predicted in combination with multiple program operation behaviors of a customer, the renewal behavior is predicted according to the prediction result of the behaviors and the corresponding operation behaviors, so as to screen out customers with high renewal intention, so that the customers with high renewal intention are arranged for follow-up by a special person, and the renewal work efficiency is further improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a renewal prediction method and device, equipment and a medium. BACKGROUND

[0002] In recent years, with the development of insurance business, the number of insurance policies sold through the Internet has also increased year by year. For insurance companies, the renewal operation of the insurance policy will directly affect the premium scale. When the insurance product purchased by the customer expires, the customer will be expected to renew the insurance, and in this process, the marketing staff cannot know the customer's renewal intention, so they cannot carry out differentiated and efficient marketing.

[0003] At present, the renewal intention of the customer is generally predicted by the personal experience of the marketing staff. Due to the differences in the personal experience of different marketing staff, the accuracy of the manual judgment method is not high and is unstable. Therefore, there is an urgent need for a customer renewal prediction scheme to effectively and accurately predict the customer's renewal intention. SUMMARY

[0004] Therefore, the embodiments of the present application provide a renewal prediction method, device, equipment and medium to solve the problem of low accuracy and instability when predicting the customer's renewal intention.

[0005] In a first aspect, the embodiments of the present application provide a renewal prediction method, which comprises:

[0006] Based on a preset retrieval condition, a set of insurance policies is obtained from a database, and the set of insurance policies includes at least one target insurance policy to be renewed;

[0007] For any target insurance policy, user information in the target insurance policy is extracted, and based on the user information, historical program operation behaviors of a user corresponding to the target insurance policy are obtained from an insurance application program;

[0008] According to the historical program operation behaviors, the operation behaviors of the user corresponding to the target insurance policy after starting the insurance application program next time are predicted to obtain a behavior prediction result;

[0009] According to the historical program operation behaviors and the behavior prediction result, the renewal behavior of the target insurance policy is predicted to obtain a renewal prediction result.

[0010] In a second aspect, the embodiments of the present application provide a renewal prediction device, which comprises:

[0011] The acquisition module is configured to obtain a set of insurance policies from a database based on a preset retrieval condition, and the set of insurance policies includes at least one target insurance policy to be renewed;

[0012] extracting, for any target insurance policy, user information in the target insurance policy, and obtaining, according to the user information, historical program operation behaviors of a user corresponding to the target insurance policy from an insurance application program;

[0013] a first prediction module configured to predict, according to the historical program operation behaviors, operation behaviors of the user corresponding to the target insurance policy after the user next starts the insurance application program, to obtain a behavior prediction result;

[0014] a second prediction module configured to predict, according to the historical program operation behaviors and the behavior prediction result, a renewal behavior of the target insurance policy, to obtain a renewal prediction result.

[0015] In a third aspect, an embodiment of the present application provides a computer device, which comprises a processor, a memory, and a computer program stored in the memory and capable of running on the processor, and the computer device is configured to execute the renewal prediction method according to any one of the first aspect.

[0016] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is configured to execute the renewal prediction method according to any one of the first aspect when executed by a processor.

[0017] Compared with the prior art, the present application has the following beneficial effects:

[0018] According to the preset retrieval condition, a set of insurance policies is obtained from a database, the set of insurance policies comprises at least one target insurance policy to be renewed, for any target insurance policy, user information in the target insurance policy is extracted, according to the user information, historical program operation behaviors of a user corresponding to the target insurance policy are obtained from an insurance application program, according to the historical program operation behaviors, operation behaviors of the user corresponding to the target insurance policy after the user next starts the insurance application program are predicted, to obtain a behavior prediction result, and according to the historical program operation behaviors and the behavior prediction result, a renewal behavior of the target insurance policy is predicted, to obtain a renewal prediction result. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 This is a schematic diagram of an application environment of a renewal prediction method provided by an embodiment of the present invention;

[0021] Figure 2 This is a flow chart of a renewal prediction method provided by one embodiment of the present invention;

[0022] Figure 3 This is a structural block diagram of a renewal prediction device provided by one embodiment of the present invention;

[0023] Figure 4 It is a structural diagram of a computer device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0025] In the following description, specific details such as particular system structures and techniques are provided for purposes of illustration, not limitation, to facilitate a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

[0026] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0027] It will also be understood that the term "and / or" used in the present description and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0028] As used in the present specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0029] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0030] References to "one embodiment" or "some embodiments" in the present specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present invention. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in yet other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0031] Embodiments of the present invention can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results.

[0032] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0033] It should be understood that the order of execution of the steps in the following embodiments does not necessarily mean the order in which they are executed. The order in which each process is executed should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0034] In order to illustrate the technical solution of the present invention, specific embodiments are provided below.

[0035] An embodiment of the present invention provides a renewal prediction method that can be applied to Figure 1 In an application environment, clients communicate with servers. Clients include, but are not limited to, PDAs, desktop computers, laptops, ultra-mobile personal computers (UMPCs), netbooks, cloud computing devices, personal digital assistants (PDAs), and other computer devices. Servers can be standalone servers or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0036] See also Figure 2 , is a flow chart of a renewal prediction method provided by an embodiment of the present invention. The renewal prediction method can be applied to Figure 1 The server in Figure 2 As shown, the renewal prediction method may include the following steps.

[0037] S201: Based on preset search conditions, obtain a policy set from a database, where the policy set includes at least one target policy to be renewed.

[0038] In step S201, based on the preset search conditions, a policy set is obtained from the database, where the policy set includes at least one target policy to be renewed, wherein the preset search condition is whether the validity period of the policy has expired, and the expired policy is used as the target policy.

[0039] In this embodiment, target insurance policies with expiration dates are retrieved from the insurance policy database. The policy expiration date is the expiration date of the user's insurance. For example, if the policy is valid for one year, the expiration date of the user's insurance is the same day of the following year. When retrieving a set of insurance policies from the insurance policy database, a search can be performed using the condition that the expiration date of the user's insurance is the current day. The policy set is then retrieved based on the search results.

[0040] In this embodiment, a policy set of the target policy is obtained so as to remind the user corresponding to the target policy so that the user can renew the policy in time and prevent the policy from being inactive due to the user forgetting to renew.

[0041] S202: For any target insurance policy, extract the user information in the target insurance policy, and obtain the historical program operation behavior of the user corresponding to the target insurance policy from the insurance application based on the user information.

[0042] In step S202, for any target insurance policy, the user information in the target insurance policy is extracted, and based on the user information, the historical program operation behavior of the user corresponding to the target insurance policy is obtained from the insurance application, wherein the user information is the identification information of the user corresponding to the target insurance policy, and the historical program operation behavior is the user's behavior during the insurance process of the target insurance policy.

[0043] In this embodiment, for any target insurance policy, user information in the target insurance policy is extracted, where the user information can be the user's identity information, name information, or phone information, etc., where the user information is unique identification information, and based on the user information, the historical program operation behavior of the user corresponding to the target insurance policy is obtained from the insurance application. The historical program operation behavior can be operation behavior obtained from different channels, such as historical program operation behavior obtained through embedded data in the application for purchasing the insurance policy, or historical program operation behavior obtained through an application that saves phone records, or historical program operation behavior obtained through business records handled by an offline counter application, etc.

[0044] Among them, the historical program operation behaviors obtained through the embedded data in the application for purchasing insurance policies may include the operation behavior of clicking on the application for purchasing insurance policies, or the operation behavior of purchasing on the application for purchasing insurance policies, or the operation behavior of transferring to manual operation on the application for purchasing insurance policies, such as the operation behavior of clicking, the operation behavior of purchasing, the operation behavior of transferring to manual operation, etc. The historical program operation behaviors obtained through the application for saving phone records may include the operation behavior of the time when the phone call was connected, such as the operation behavior of connecting the phone on the same day, etc. The historical program operation behaviors obtained through the business records handled by offline counter applications may include the operation behavior of offline business types, such as the operation behavior of handling business A.

[0045] In this embodiment, the user's historical application operation behavior for the target policy is retrieved from the insurance application based on user information. This allows prediction of the user's operation behavior the next time the insurance application is launched based on the user's historical application operation behavior. Using historical application operation behavior to predict the user's operation behavior the next time the insurance application is launched can improve prediction accuracy.

[0046] S203: Based on the historical program operation behaviors, predict the operation behaviors of the user corresponding to the target insurance policy after starting the insurance application program next time, and obtain a behavior prediction result.

[0047] In step S203, based on historical program operation behaviors, the operation behaviors of the user corresponding to the target insurance policy after the insurance application is started next time are predicted, that is, the channel for the user to operate the target insurance policy again is predicted, where the next time the insurance application is started is the time when the corresponding user handles business next time.

[0048] In this embodiment, when predicting the operating behavior of the user corresponding to the target insurance policy after the insurance application is started next time based on the historical program operating behavior, a preset behavior prediction model can be used for prediction, wherein the input of the preset behavior prediction model can be the historical program operating behavior, and the output is the probability of the behavior prediction result.

[0049] Optionally, based on historical program operation behaviors, the operation behaviors of the user corresponding to the target insurance policy after the next launch of the insurance application are predicted, and the behavior prediction results are obtained, including:

[0050] Obtaining a preset behavior prediction model;

[0051] Input historical program operation behaviors into the preset behavior prediction model and output the behavior prediction results.

[0052] In this embodiment, a preset behavior prediction model is obtained. The preset behavior prediction model may be a MIMN (Multi-Channel User Interest Memory Network) model. Based on the corresponding model, historical program operation behavior is input into the preset behavior prediction model, and a behavior prediction result is output. The behavior prediction result includes the predicted probabilities of operation behaviors in different channels. For example, the predicted operation behavior after the next insurance application launch is 0.7 for a click on the policy purchase application, 0.2 for a purchase, 0.56 for a manual operation, etc. The probability of a phone call being connected on the same day is 0.8, and the probability of an offline transaction A is 0.8, etc.

[0053] It should be noted that to remove low-probability events, a corresponding threshold is set. When the probability value of the behavior prediction result is less than the threshold, the corresponding operation with a small probability value is removed. For example, if the probability of clicking on the insurance purchase application after launching the insurance application is 0.7, the probability of purchasing the policy is 0.2, and the probability of switching to manual operation is 0.56, then the purchase and switching to manual operation actions are removed.

[0054] S204: Based on the historical program operation behavior and the behavior prediction results, the renewal behavior of the target policy is predicted to obtain a renewal prediction result.

[0055] In step S204, the renewal behavior of the target insurance policy is predicted according to the historical program operation behavior and the behavior prediction result, that is, the possibility of the corresponding user renewing the policy is determined according to the historical program operation behavior and the behavior prediction result.

[0056] In this embodiment, the renewal behavior of the target insurance policy is predicted according to the historical program operation behavior and the behavior prediction result, and the renewal prediction result is obtained, wherein the renewal prediction result is the possibility of renewal. When predicting the renewal behavior of the target insurance policy, the behavior prediction result and the number of behaviors in the historical program operation behavior can be used for prediction. The weight value is determined according to the number of operation behaviors in the historical program operation behavior, the weight value corresponding to each operation behavior is multiplied by the prediction probability of the corresponding behavior in the behavior prediction result, and then added to obtain the prediction of the renewal behavior.

[0057] According to the behavior prediction result, the prediction behavior that meets the requirements in the behavior prediction result is determined, that is, the behavior corresponding to the probability value greater than the preset threshold value in the behavior prediction result, and the number of times of the corresponding behavior is obtained from the historical program operation behavior according to the prediction behavior that meets the requirements.

[0058] For example, the probability of the operation behavior after starting the insurance application next time is 0.7, and the preset threshold value is 0.6, and it is considered that the click operation behavior meets the requirements. The number of clicks of the click operation behavior is obtained from the historical program operation behavior. If the number of clicks of the click operation behavior on the application for purchasing the insurance policy in the historical program operation behavior is 10 times, the weight value of the click operation behavior is determined, such as the weight value of the click operation behavior is 0.2 when the number of times of the click operation behavior is between 0 and 10 including 10, the weight value of the click operation behavior is 0.5 when the number of times of the click operation behavior is between 10 and 50 including 50, and the weight value of the click operation behavior is 1 when the number of times of the click operation behavior is more than 50. The number of clicks of the click operation behavior is 10 times, and the weight value is 0.2. Then the weight value of the operation behavior is multiplied by the prediction probability of the corresponding behavior in the behavior prediction result, and 0.14 is obtained.

[0059] If the behavior prediction result also includes the probability of answering the phone on that day, which is 0.8, in the historical program operation behavior, the operation behavior of answering the phone on that day is considered to meet the requirements, and the number of operation behaviors of answering the phone on that day is obtained from the historical program operation behavior. If the number of operation behaviors of answering the phone on that day in the historical program operation behavior is 15 times, the weight value of the operation behavior of answering the phone on that day is determined. For example, the weight value of the number of operation behaviors of answering the phone on that day is between 0-10, including 10, is 0.2, the weight value of the number of operation behaviors of answering the phone on that day is between 10-50, including 50, is 0.5, the weight value of the number of operation behaviors of answering the phone on that day is above 50, and so on. The number of operation behaviors of answering the phone on that day in the historical program operation behavior is 15 times, and the corresponding weight value is 0.5. The weight value corresponding to the operation behavior is multiplied by the predicted probability of the corresponding behavior in the behavior prediction result to obtain 0.4.

[0060] If other behaviors in the behavior prediction result do not meet the corresponding requirements, 0.14 is added to 0.4 to obtain the renewal prediction result. The larger the renewal prediction result, the more likely the user is to renew.

[0061] Optionally, obtaining a scorecard for scoring the target policy;

[0062] Based on historical program operation behavior and behavior prediction results, combined with the score card, the target policy is scored to obtain the scoring result, which is used as the renewal prediction result.

[0063] In this embodiment, a scoring card for scoring the target insurance policy is obtained, and the target insurance policy is scored based on the historical program operation behavior and the behavior prediction results, combined with the scoring card. Among them, the scoring card includes different scoring dimensions, such as the willingness to pay dimension, the satisfaction dimension, the activity dimension, etc. Different scoring dimensions can include a single evaluation standard or multiple evaluation standards. For example, the willingness to pay dimension can be a single evaluation standard, and the evaluation standard is the number of payments in the past two years. The satisfaction dimension can be multiple evaluation standards, and the multiple evaluation standards are the number of transfers to manual labor in the past three months and the probability of the next transfer to manual labor within 30 days. The activity dimension can be multiple evaluation standards, and the multiple evaluation standards are the number of clicks on the application for purchasing the insurance policy in the past month and the number of business handling behaviors within 30 days.

[0064] Scoring is performed on the scoring criteria within each scoring dimension. Based on the scoring results, the renewal behavior of the target policy is predicted to obtain a renewal prediction result. When scoring the scoring criteria within each scoring dimension, different scoring criteria are assigned levels, with each level corresponding to a certain score. For example, if the scoring criteria for the number of payments in the past two years are set to the first, second, third, and fourth levels, the first level is the number of payments with a range of (0,1], the second level is the number of payments with a range of (1,5], the third level is the number of payments with a range of (5,10], and the fourth level is the number of payments with a range of (10, more than 10 times], then the first level is set to 10 points, the second level is set to 20 points, the third level is set to 30 points, and the fourth level is set to 40 points. For example, if the scoring criteria for the number of transfers to manual labor in the past three months are set to the first, second, and third levels, the first level is the number of transfers to manual labor with a range of (0,1], the second level is the number of transfers to manual labor with a range of (1,2], and the third level is the number of transfers to manual labor with a range of (2, more than 2 times], then the first level is set to 20 points, the second level is set to 10 points, and the third level is set to 0 points. For example, if the scoring criteria for the probability of the next transfer to manual labor within 30 days are set to the first, second, and third levels, then the probability range of the first level is (0,0.3] , the second-level probability interval is (0.3,0.6], the third-level probability interval is (0.6,1], the first level is set to 20 points, the second level is set to 10 points, and the third level is set to 0 points. For example, in the scoring criteria for the number of clicks on the application for purchasing insurance policies in the past month, the first-level click interval is (0,1], the second-level click interval is (1,5], the third-level click interval is (5,5 times or more], the first level is set to 0 points, the second level is set to 10 points, and the third level is set to 20 points. For example, in the scoring criteria for the number of business transactions within 30 days, the first-level business transaction interval is (0,2], the second-level business transaction interval is (2,5], the third-level business transaction interval is (5,10], the first level is set to 0 points, the second level is set to 10 points, and the third level is set to 20 points.

[0065] It should be noted that the scoring is performed based on historical program operation behavior and behavior prediction results combined with the score card, and the scoring results are used as renewal prediction results.

[0066] For example, when scoring user A for a target insurance policy, if user A has made 2 payments in the past 2 years, has transferred to manual service 0 times in the past 3 months, has a probability of transferring to manual service within 30 days of 0.3, and has clicked 7 times on the insurance policy purchase application in the past month, then the target insurance policy is scored 80 points, and the renewal prediction result is 80 points.

[0067] In this embodiment, the target insurance policy is scored according to different scoring dimensions, and the behavior prediction results are combined with the historical program operation behavior, which fully utilizes the historical program operation behavior and improves the accuracy of predicting the renewal behavior of the target insurance policy.

[0068] Optionally, the target policy is scored based on historical program operation behaviors and behavior prediction results, combined with a scorecard, to obtain a scoring result, including:

[0069] Set weight values ​​for each scoring dimension;

[0070] Calculate the dimension score results for each scoring dimension based on historical program operation behavior and behavior prediction results, combined with the scorecard;

[0071] According to the weight value set for each scoring dimension and the dimension scoring result of each scoring dimension, the target insurance policy is scored to obtain the scoring result.

[0072] In this embodiment, a weight value is set for each scoring dimension. When setting the weight value, it can be set according to the number of evaluation criteria in each scoring dimension. For example, the willingness to pay dimension includes one evaluation criteria, the satisfaction dimension includes 2 evaluation criteria, and the activity dimension includes 2 evaluation criteria. A weight value of 0.2 can be set for the willingness to pay dimension, a weight value of 0.4 can be set for the satisfaction dimension, and a weight value of 0.4 can be set for the activity dimension.

[0073] In this embodiment, the scoring results for each scoring dimension are calculated based on historical program operation behavior and behavior prediction results, combined with a scorecard. For example, when scoring user A for a target insurance policy, if user A has made two payments in the past two years, has transferred to manual service at no time in the past three months, has a probability of another transfer to manual service within 30 days of 0.3, has clicked seven times on the policy purchase application in the past month, and has conducted three transactions within 30 days, then the score for willingness to pay is 20 points, the score for satisfaction is 40 points, and the score for activity is 30 points. Based on the weights set for each scoring dimension and the score results for each scoring dimension, the target insurance policy is scored, resulting in a score of 32.

[0074] Optionally, based on historical program operation behaviors and behavior prediction results, the renewal behavior of the target policy is predicted. After obtaining the renewal prediction results, the following steps are also included:

[0075] According to the renewal prediction results, each target policy is sorted according to the size of the renewal prediction results to obtain the sorting results;

[0076] Based on the sorting results, the renewal reminder method for each target policy is determined, and based on the reminder method for each target policy, the user corresponding to each target policy is reminded.

[0077] In this embodiment, based on the renewal prediction results, each target policy is sorted by the size of its renewal prediction results to obtain a sorting result. Specifically, each target policy is sorted from largest to smallest based on its score result to obtain a sorting result. Based on the sorting result, a renewal reminder method is determined for each target policy. Specifically, different renewal reminder methods may vary for different score results. Based on the reminder method for each target policy, a reminder is then sent to the user corresponding to each target policy.

[0078] In this embodiment, based on the sorting results, the renewal reminder method for each target policy is determined, so that the target policies can be screened according to the sorting results, and the policies that can be reminded for renewal by manual customer service are screened out. Manual customer service reminders are given to the screened out policies that can be reminded for renewal by manual customer service, and no manual customer service reminders are given to the remaining policies, thereby improving the work efficiency of manual customer service.

[0079] Optionally, based on the ranking results, determine a renewal reminder method for each target policy, including:

[0080] When the prediction result is greater than a preset threshold, determining the first reminder method as the reminder method for the insurance policy whose prediction result is greater than the preset threshold;

[0081] When the prediction result is not greater than the preset threshold, the second reminder method is determined to be the reminder method for the insurance policy whose prediction result is not greater than the preset threshold.

[0082] In this embodiment, the reminder method includes a first reminder method and a second reminder method, wherein the first reminder method is a manual customer service reminder, and the second reminder method is a text message reminder. When the prediction result is greater than a preset threshold, the first reminder method is determined to be the reminder method for policies with prediction results greater than the preset threshold, that is, the manual customer service reminder is used. When the prediction result is not greater than the preset threshold, the second reminder method is determined to be the reminder method for policies with prediction results not greater than the preset threshold, that is, the text message reminder is used.

[0083] Optionally, according to the reminder method of each target policy, the user corresponding to each target policy is reminded, including:

[0084] Generate reminder words that match the reminder method;

[0085] Remind each target policy user based on the reminder script.

[0086] In this embodiment, when a renewal reminder is sent to a user, reminder language matching the reminder method is generated. For example, for the first reminder method, i.e., a manual customer service reminder, manual customer service outbound call language is generated. For the second reminder method, i.e., a text message reminder, text message content is generated. The manual customer service outbound call language and text message content can be generated based on corresponding models, such as the chatGLM model.

[0087] For example, when generating outbound call scripts, the chatGLM model input could be, "I'm an insurance customer service representative, preparing to call customer A to remind him that his policy is about to expire. The model predicts that he'll likely answer the phone between 8 and 10 a.m. daily, and his immediate concern is how to claim his survival benefit. Please generate an outbound call script for me to guide this customer to pay his premium as soon as possible to avoid the risks associated with policy expiration. I suggest starting with: Hello, Mr. / Ms. A! I'm XXX. Then ask if his issue has been resolved. Finally, remind him to pay his premium as soon as possible." The output of the outbound call script could be, "Hello! I'm XXX. Thank you very much for answering our call. Has the XXX issue you inquired about last time been resolved? This call is mainly to remind you that your policy is about to expire on X / X / X. Please renew your premium as soon as possible to avoid other risks associated with policy expiration. Thank you again for your patience and wish you a happy life!"

[0088] When generating SMS content, the chatGLM model input might be, "I'm an insurance customer service representative, preparing a SMS to remind customer A that their policy is about to expire. Please help me generate the SMS content. If they need my assistance, please reply 1." The output SMS content might be, "Dear Mr. / Ms. A, hello! Thank you for choosing our insurance service. This message reminds you that your policy will expire on X / X. To ensure your protection, please renew your premium on time. If you need assistance from a customer service specialist, please reply 1. I wish you a happy life!"

[0089] It should be noted that customer service personnel can modify the outbound call scripts / SMS content generated by chatGLM and upload them to the work platform for record. The system will record each customer service personnel's input, output, and modified content, and will prioritize the modified outbound call scripts / SMS content when it is output next time.

[0090] Based on preset search conditions, a policy set is obtained from the database, and the policy set includes at least one target policy to be renewed. For any target policy, the user information in the target policy is extracted. Based on the user information, the historical program operation behavior of the user corresponding to the target policy is obtained from the insurance application. Based on the historical program operation behavior, the operation behavior of the user corresponding to the target policy after the insurance application is started next time is predicted to obtain a behavior prediction result. Based on the historical program operation behavior and the behavior prediction result, the renewal behavior of the target policy is predicted to obtain a renewal prediction result. In this application, the operation behavior after the next startup of the insurance application is predicted based on multiple program operation behaviors of the customer. The renewal behavior is predicted based on the behavior prediction result and the corresponding operation behavior, so as to screen out customers with a high willingness to renew, and arrange for dedicated personnel to follow up on customers with a high willingness to renew, so as to further improve the efficiency of the renewal work.

[0091] See also Figure 3 , Figure 3 This is a structural block diagram of a renewal prediction device provided by an embodiment of the present invention, which is applied to the above-mentioned server. For ease of explanation, only the parts related to the embodiment of the present application are shown. Figure 3 The renewal prediction device 30 includes an acquisition module 31 , an extraction module 32 , a first prediction module 33 , and a second prediction module 34 .

[0092] The acquisition module 31 is used to acquire a policy set from the database based on a preset search condition, where the policy set includes at least one target policy to be renewed.

[0093] The extraction module 32 is used to extract user information from any target insurance policy and obtain historical program operation behaviors of the user corresponding to the target insurance policy from the insurance application based on the user information.

[0094] The first prediction module 33 is used to predict the operation behavior of the user corresponding to the target insurance policy after starting the insurance application program next time based on the historical program operation behavior, and obtain a behavior prediction result.

[0095] The second prediction module 34 is used to predict the renewal behavior of the target policy based on the historical program operation behavior and the behavior prediction result to obtain the renewal prediction result.

[0096] Optionally, the first prediction module 33 includes:

[0097] The first acquisition unit is used to acquire a preset behavior prediction model.

[0098] The output unit is used to input historical program operation behaviors into a preset behavior prediction model and output behavior prediction results.

[0099] Optionally, the second prediction module 34 includes:

[0100] The second acquisition unit is used to obtain a scoring card for scoring the target insurance policy.

[0101] The scoring unit is used to score the target policy based on historical program operation behavior and behavior prediction results, combined with the scoring card, to obtain the scoring result, and use the scoring result as the renewal prediction result.

[0102] Optionally, the scoring unit includes:

[0103] The setting subunit is used to set the weight value for each scoring dimension.

[0104] The calculation subunit is used to calculate the dimension scoring result of each scoring dimension based on historical program operation behavior and behavior prediction results in combination with the score card.

[0105] The scoring subunit is used to score the target policy based on the weight value set for each scoring dimension and the dimension scoring result of each scoring dimension to obtain the scoring result.

[0106] Optionally, the insurance renewal prediction device 30 further includes:

[0107] The sorting module is used to sort each target policy according to the renewal prediction result to obtain the sorting result.

[0108] The determination module is used to determine the renewal reminder method for each target insurance policy based on the sorting results, and to remind the user corresponding to each target insurance policy according to the reminder method of each target insurance policy.

[0109] Optionally, the determination module includes:

[0110] The first determining unit is used to determine, when the prediction result is greater than a preset threshold, that the first reminder method is a reminder method for the insurance policy whose prediction result is greater than the preset threshold.

[0111] The second determining unit is used to determine, when the prediction result is not greater than the preset threshold, that the second reminder method is a reminder method for the insurance policy whose prediction result is not greater than the preset threshold.

[0112] Optionally, the determination module further includes:

[0113] The generation unit is used to generate a reminder wording that matches the reminder method according to the reminder method.

[0114] The reminder unit is used to remind users of each target insurance policy according to the reminder words.

[0115] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0116] Figure 4 FIG. 1 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Figure 4 As shown, the computer device of this embodiment includes: at least one processor ( Figure 4 Only one is shown), a memory, and a computer program stored in the memory and executable on at least one processor, wherein when the processor executes the computer program, the steps of any of the above-mentioned renewal prediction method embodiments are implemented.

[0117] The computer device may include, but is not limited to, a processor and a memory. It will be understood by those skilled in the art that Figure 4 The above is merely an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include a network interface, a display screen, and an input device.

[0118] The processor may be a CPU, other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0119] Memory includes readable storage media, internal memory, and the like. Internal memory can be the internal memory of a computer device, providing an environment for the operation of the operating system and computer-readable instructions stored in the readable storage medium. The readable storage medium can be the computer device's hard drive. In other embodiments, it can also be an external storage device, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, or a flash memory card. Furthermore, memory can include both the computer device's internal storage unit and external storage devices. Memory is used to store the operating system, application programs, boot loaders, data, and other programs, such as the program code of computer programs. Memory can also be used to temporarily store data that has been output or is about to be output.

[0120] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process steps in the above-described method embodiments by instructing the relevant hardware through a computer program. The computer program may be stored in a computer-readable storage medium. When executed by a processor, the computer program implements the steps of the above-described method embodiments. The computer program includes computer program code, which may be in source code form, object code form, executable file, or some intermediate form. Computer-readable media may include at least: any entity or device capable of carrying computer program code, recording media, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunications signals, and software distribution media. Examples include USB flash drives, removable hard drives, magnetic disks, or optical disks. In some jurisdictions, based on legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunications signals.

[0121] The present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed through a computer program product. When the computer program product runs on a computer device, the computer device can implement the steps in the above-mentioned method embodiment when executing it.

[0122] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0123] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0124] In the embodiments provided in this application, it should be understood that the disclosed apparatus / computer equipment and methods can be implemented in other ways. For example, the apparatus / computer equipment embodiments described above are merely schematic. For example, the division of modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components 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 through some interfaces, indirect coupling or communication connection of the apparatus or unit, which can be electrical, mechanical or other forms.

[0125] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0126] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. 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 make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A renewal prediction method, characterized in that: The renewal prediction method includes: Based on a preset search condition, obtaining a policy set from a database, wherein the policy set includes at least one target policy to be renewed; For any target policy, extract the user information in the target policy, and obtain the historical program operation behavior of the user corresponding to the target policy from the insurance application based on the user information; Based on the historical program operation behavior, predict the operation behavior of the user corresponding to the target insurance policy after launching the insurance application program next time, and obtain a behavior prediction result; Predicting the renewal behavior of the target policy based on the historical program operation behavior and the behavior prediction result to obtain a renewal prediction result; The step of predicting the renewal behavior of the target policy based on the historical program operation behavior and the behavior prediction result to obtain a renewal prediction result includes: Obtaining a scorecard for scoring the target insurance policy; Scoring the target policy based on the historical program operation behavior and the behavior prediction result in combination with the scoring card to obtain a scoring result, and using the scoring result as a renewal prediction result; The scorecard includes N scoring dimensions, where N is an integer greater than zero; Scoring the target policy based on the historical program operation behavior and the behavior prediction result in combination with the scoring card to obtain a scoring result includes: Set weight values ​​for each scoring dimension; Calculate the dimension scoring result of each scoring dimension based on the historical program operation behavior and the behavior prediction result in combination with the scoring card; According to the weight value set for each scoring dimension and the dimension scoring result of each scoring dimension, the target insurance policy is scored to obtain a scoring result.

2. The renewal prediction method according to claim 1, wherein: The step of predicting the operation behavior of the user corresponding to the target insurance policy after the insurance application is started next time based on the historical program operation behavior to obtain a behavior prediction result includes: Obtaining a preset behavior prediction model; The historical program operation behavior is input into the preset behavior prediction model, and the behavior prediction result is output.

3. The renewal prediction method according to claim 1, wherein: After predicting the renewal behavior of the target policy based on the historical program operation behavior and the behavior prediction result and obtaining the renewal prediction result, the method further includes: According to the renewal prediction result, each target policy is sorted according to the size of the renewal prediction result to obtain a sorting result; Based on the ranking results, a reminder method for renewal of each target policy is determined, and based on the reminder method of each target policy, a reminder is given to the user corresponding to each target policy.

4. The renewal prediction method according to claim 3, wherein: The reminder mode includes a first reminder mode and a second reminder mode; Determining a renewal reminder method for each target policy based on the ranking result includes: When the prediction result is greater than a preset threshold, determining the first reminder method as a reminder method for the insurance policy whose prediction result is greater than the preset threshold; When the prediction result is not greater than the preset threshold, the second reminder method is determined to be the reminder method for the insurance policy whose prediction result is not greater than the preset threshold.

5. The renewal prediction method according to claim 4, wherein: The reminding of each target policy's corresponding user according to the reminder method of each target policy includes: According to the reminder method, generating a reminder speech that matches the reminder method; According to the reminder words, remind the users of each target insurance policy.

6. A renewal prediction device, characterized in that: The insurance renewal prediction device comprises: An acquisition module, configured to acquire a policy set from a database based on a preset search condition, wherein the policy set includes at least one target policy to be renewed; An extraction module is used to extract user information from any target insurance policy and, based on the user information, obtain historical program operation behaviors of the user corresponding to the target insurance policy from the insurance application; A first prediction module is configured to predict, based on the historical program operation behavior, the operation behavior of the user corresponding to the target insurance policy after the insurance application is launched next time, and obtain a behavior prediction result; A second prediction module is configured to predict the renewal behavior of the target policy based on the historical program operation behavior and the behavior prediction result to obtain a renewal prediction result; The step of predicting the renewal behavior of the target policy based on the historical program operation behavior and the behavior prediction result to obtain a renewal prediction result includes: Obtaining a scorecard for scoring the target insurance policy; Scoring the target policy based on the historical program operation behavior and the behavior prediction result in combination with the scoring card to obtain a scoring result, and using the scoring result as a renewal prediction result; The scorecard includes N scoring dimensions, where N is an integer greater than zero; Scoring the target policy based on the historical program operation behavior and the behavior prediction result in combination with the scoring card to obtain a scoring result includes: Set weight values ​​for each scoring dimension; Calculate the dimension scoring result of each scoring dimension based on the historical program operation behavior and the behavior prediction result in combination with the scoring card; According to the weight value set for each scoring dimension and the dimension scoring result of each scoring dimension, the target insurance policy is scored to obtain a scoring result.

7. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and the computer device is configured to execute the renewal prediction method according to any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the renewal prediction method according to any one of claims 1 to 5 is implemented.

Images