Interest-generating scheme push method, device, electronic device and storage medium

Abstract

The present application provides a method, device, electronic device and storage medium for pushing an interest-bearing plan. The method first determines target customers with relatively high acceptance based on feedback information from customers who have already pushed the plan on historical interest-bearing plans, and determines in which insurance channel the target customers are mainly distributed. Then, customers subsequently obtained from this first target insurance channel will also have a high acceptance of the interest-bearing plan, thereby narrowing the scope of finding customers and being more accurate. In addition, based on the attributes of important customers in the first target insurance channel and their behaviors in the second historical time period, a customer grading model is called to perform grading to obtain first important customers and second important customers. The first interest-bearing plan is directly pushed to the first important customer with a higher grade. For the second important customer with a lower grade, the intersection of the first important customer and the important customer in the second target insurance channel is first calculated to obtain a third important customer, and then the second interest-bearing plan is pushed to the third important customer, thereby achieving targeted push, and both types of customers have low aversion and high transaction probability.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, electronic device and storage medium for pushing an interest-bearing plan. Background Art

[0002] The traditional model of expanding the credit card customer base and providing matching products for existing customers is mostly based on the business experience of bank business personnel, but the population characteristics in the experience are limited. As time goes by, the number of customers identified by traditional experience is close to saturation, which will lead to a continuous decrease in target customers. When recommending interest-bearing plans, it is easy to lead to repeated recommendations, causing customer complaints and making it difficult to achieve performance goals. In addition, when currently expanding new customers, most of them are looking for target customers without knowing it, which is easy to lead to inefficiency and customer complaints.

[0003] Therefore, the current process of recommending interest-bearing products by banks has technical problems such as difficulty in acquiring potential customers and prone to customer complaints, which needs to be improved. Summary of the invention

[0004] The embodiments of the present application provide a method, device, electronic device and storage medium for pushing interest-bearing plans, which are used to alleviate the technical problems of difficulty in acquiring potential customers and easy customer complaints in the current process of recommending interest-bearing products by banks.

[0005] To solve the above technical problems, the present application provides the following technical solutions:

[0006] The present application provides a method for pushing an interest-bearing plan, including:

[0007] Determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold value according to feedback information of customers on the historical interest-bearing scheme that has been pushed in the first historical time period;

[0008] Determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in the first historical time period in multiple insurance channels;

[0009] According to the first attribute information of the important customer in the first target insurance type channel and the first behavior information of the important customer in the second historical time period, calling the trained customer classification model to classify the important customer, and obtaining a first important customer whose grade is greater than a first preset grade and a second important customer whose grade is not greater than the first preset grade;

[0010] generating, based on the first attribute information and the first behavior information of the first important customer, and pushing a matching first interest earning plan to the first important customer;

[0011] A third important customer who overlaps with an important customer in the second target insurance type channel is determined from the second important customers, and a matching second interest earning plan is generated and pushed to the third important customer based on the first attribute information and the first behavior information of the third important customer.

[0012] At the same time, the embodiment of the present application also provides an interest-generating scheme push device, including:

[0013] A first determination module is used to determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold value based on feedback information of customers on the historical interest-bearing scheme that has been pushed in a first historical time period;

[0014] A second determination module is used to determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in the first historical time period in the multiple insurance channels;

[0015] a classification module, configured to classify the important customers by calling the trained customer classification model according to the first attribute information of the important customers in the first target insurance channel and the first behavior information of the important customers in the second historical time period, and obtain a first important customer whose grade is greater than a first preset grade and a second important customer whose grade is not greater than the first preset grade;

[0016] A first generating module, configured to generate and push a matching first interest earning plan to the first important customer according to the first attribute information and the first behavior information of the first important customer;

[0017] The second generating module is used to determine a third important customer who overlaps with an important customer in the second target insurance type channel from among the second important customers, and generate and push a matching second interest earning plan to the third important customer based on the first attribute information and the first behavior information of the third important customer.

[0018] The present application also provides an electronic device, comprising a memory and a processor; the memory stores an application, and the processor is used to run the application in the memory to execute the steps in any of the above-mentioned interest-bearing scheme push methods.

[0019] An embodiment of the present application provides a computer-readable storage medium, which stores a plurality of instructions suitable for loading by a processor to execute the steps in the above-mentioned interest-bearing scheme push method.

[0020] Beneficial effects: The present application provides a method, device, electronic device and storage medium for pushing an interest-bearing plan. The method first determines target customers with a relatively high degree of acceptance based on the feedback information of the currently pushed customers on the historical interest-bearing plan, and then starts from the insurance channel to see in which insurance channel the customers with a high degree of acceptance of the historical interest-bearing plan are mainly distributed. Then, the customers subsequently obtained from this insurance channel will also have a high degree of acceptance of the interest-bearing plan. Compared with searching for customers by investing in a large number of customers, the scope of searching for customers is narrowed, and the scope is more accurate. In addition, in the first target insurance channel, according to the attributes of important customers and the behavior in the second historical time period, the customer classification model is called for classification, and the first important customer with a higher grade can be determined from all important customers. Pushing the first interest-bearing plan to these customers can further improve the transaction probability. Finally, the second important customer is intersected with the important customers in the associated second target insurance channel. The third important customer obtained has a high degree of recognition of the company's products because he has purchased insurance in both insurance channels. Therefore, the probability of causing customer disgust when pushing the second interest-bearing plan to these customers is low, and the use of different interest-bearing plans from the first important customer also realizes targeted push of different types of customers, and the transaction probability is further improved. Based on the above points, this application achieves accurate acquisition of potential customers while minimizing customer complaints by associating interest-bearing plans with insurance channels, which is conducive to interest growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The technical solution and other beneficial effects of the present application will be made apparent by describing in detail the specific implementation methods of the present application in conjunction with the accompanying drawings.

[0022] Figure 1 It is a schematic diagram of an application scenario of the interest-bearing scheme push method provided in an embodiment of the present application.

[0023] Figure 2 A flowchart of the method for pushing an interest-bearing plan provided in an embodiment of the present application.

[0024] Figure 3 This is an overall architecture diagram of the interest-bearing scheme push method provided in an embodiment of the present application.

[0025] Figure 4 A schematic diagram of the structure of the interest-generating scheme push device provided in an embodiment of the present application.

[0026] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0028] The embodiments of the present application provide a method, device, electronic device and computer-readable storage medium for pushing an interest-earning scheme, wherein the interest-earning scheme pushing device can be integrated in an electronic device, which can be a server, a terminal or other device.

[0029] See also Figure 1 , Figure 1 A schematic diagram of a scenario for applying the method for pushing an interest-bearing scheme provided in an embodiment of the present application, wherein the scenario may include terminals and servers, and the terminals, servers, and terminals and servers are connected and communicated through the Internet composed of various gateways, and the application scenario includes a first user terminal 10, a server 20, a second user terminal 30, and a third user terminal 40; wherein the first user terminal 10, the second user terminal 30, and the third user terminal 40 may be devices with human-computer interaction functions; the server 20 includes a local server and / or a remote server, etc.

[0030] The first user terminal 10, the server 20, the second user terminal 30 and the third user terminal 40 are located in a wireless network or a wired network to achieve data interaction among the four parties, wherein:

[0031] During the first historical time period, the customer has received the historical interest-bearing plan through the first user terminal 10 and provided feedback information. The historical interest-bearing plan may include the credit card application plan, activation plan, loan plan, installment plan, incentive plans for various products, etc. provided by the bank.

[0032] The server 20 determines the target customers whose acceptance of the historical interest-bearing scheme is greater than the first threshold value based on the feedback information, and then obtains the insurance behavior information generated by the target customers in multiple insurance channels in the first historical time period, and determines the first target insurance channel and the second target insurance channel associated with the first target insurance channel from the multiple insurance channels based on the insurance behavior information. Then, the server 20 obtains the first attribute information of the important customers in the first target insurance channel and the first behavior information of the important customers in the second historical time period, and calls the trained customer classification model to classify the important customers based on the first attribute information and the first behavior information, and obtains the first important customer whose grade is greater than the first preset grade and the second important customer whose grade is not greater than the first preset grade.

[0033] Finally, the server 20 generates and pushes a matching first interest-bearing plan to the second user terminal 30 of the first important customer based on the first attribute information and the first behavior information of the first important customer. The server 20 determines a third important customer who overlaps with the important customer in the second target insurance channel from the second important customer, and generates and pushes a matching second interest-bearing plan to the third user terminal 40 of the third important customer based on the first attribute information and the first behavior information of the third important customer. Both the first interest-bearing plan and the second interest-bearing plan may include a credit card application plan, activation plan, loan plan, installment plan, incentive plans for various products, etc. provided by the bank.

[0034] Through the above method, first determine the target customers with relatively high acceptance based on the feedback information of the customers who have been pushed to the historical interest-bearing plan, and determine in which insurance channel the target customers are mainly distributed. Then, the customers subsequently obtained from this first target insurance channel will also have a high acceptance of the interest-bearing plan, thereby narrowing the scope of finding customers and being more accurate. In addition, according to the attributes of the important customers of the first target insurance channel and the behavior in the second historical time period, call the customer classification model to classify the first and second important customers, and directly push the first interest-bearing plan to the first important customer with a higher grade. For the second important customer with a lower grade, first find the intersection with the important customers in the second target insurance channel to obtain the third important customer, and then push the second interest-bearing plan to the third important customer, thereby achieving targeted push, and the two types of customers have low aversion and high probability of transaction.

[0035] It should be noted that Figure 1 The system scenario diagram shown is only an example. The server and scenario described in the embodiment of the present application are intended to more clearly illustrate the technical solution of the embodiment of the present application, and do not constitute a limitation on the technical solution provided in the embodiment of the present application. It is known to those skilled in the art that with the evolution of the system and the emergence of new business scenarios, the technical solution provided in the embodiment of the present application is also applicable to similar technical problems. The following are detailed descriptions. It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0036] See also Figure 2 , Figure 2 This is a flow chart of the method for pushing interest-bearing schemes provided in the embodiment of the present application. Figure 3 , Figure 3 This is the overall architecture diagram of the interest-bearing scheme push method provided in the embodiment of the present application, combined with Figure 2 and Figure 3 , the method specifically comprises:

[0037] S1: Determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold value based on the feedback information of the customers on the historical interest-bearing scheme that has been pushed in the first historical time period.

[0038] The first historical time period can be one month or several months from the current date, or it can be dozens of days from the current date, or other time periods that have passed. The length and period of the first historical time period can be selected as needed. The historical interest-bearing plan refers to the marketing plan formulated by the bank for interest-bearing products. In the embodiment of the present application, interest-bearing products refer to products related to the bank's interest-bearing assets and issued to customers in the form of credit cards. In this process, customers need to pay certain fees so that the bank can make a profit. Interest-bearing products may include the credit card itself, credit card cash advances, credit card loans, etc. After the credit card is opened, customers can withdraw cash and loans through the credit card, and customers need to pay the bank various fees such as annual fees, withdrawal fees, repayment interest, etc. in this process. Therefore, the historical interest-bearing plan may include credit card application plans, activation plans, loan plans, installment plans, incentive plans for various products, etc.

[0039] The historical interest-bearing schemes can be pushed to customers by SMS, APP, phone, email, etc. The users who receive this information are pushed customers. After receiving these historical interest-bearing schemes, the pushed customers will give certain feedback, such as clicking on the link in the SMS or email, entering the relevant page of the APP to trade, or directly closing the page and deleting the message, etc. The feedback information can be obtained by integrating the feedback operations of these pushed customers who received the historical interest-bearing schemes in the first historical time period. Different customers have different acceptance of historical interest-bearing schemes. Some customers may be more interested in historical interest-bearing schemes and successfully open cards, apply for installments and other related transactions, which means that their acceptance is high. Some users may feel disgusted with historical interest-bearing schemes and directly delete or complain, which means that their acceptance is low. The level of acceptance of each customer can be comprehensively evaluated based on their feedback information on all historical interest-bearing schemes in the entire first historical time period. In this step, based on the feedback information, target customers whose acceptance of the historical interest-bearing plan is greater than a first threshold are determined from all pushed customers. The first threshold can be set as needed. For example, if the maximum acceptance is set to 100%, the first threshold can be 70% or 80%, etc.

[0040] S2: Determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels based on the insurance behavior information generated by the target customer in the first historical time period in multiple insurance channels.

[0041] In the embodiment of the present application, each insurance channel may include a life insurance channel, a critical illness insurance channel, a medical insurance channel, an accident insurance channel, a property damage insurance channel, a credit insurance channel, a liability insurance channel, and other insurance channels that the company sells. Under normal circumstances, products under the same company name may include credit cards provided by banks and insurance provided by insurance companies. The target customers determined in the above steps are customers with a high acceptance of historical interest-bearing plans. Based on the insurance behavior information generated by these target customers in all the insurance channels sold by the company, it can be determined which insurance channel these target customers mainly insure in. For the customer group of this insurance channel, compared with the customer groups of other insurance channels, the proportion of customers with a high acceptance of interest-bearing plans is higher, that is, there are more potential customers. Therefore, after determining the first target insurance channel from multiple insurance channels, the interest-bearing plan is pushed to the specific customer group of the first target insurance channel. Compared with searching for customers by investing in a large number of customers, the scope of searching for customers is narrowed, and the scope is more accurate, which increases the probability of interest growth and reduces the probability of customer complaints.

[0042] In addition, the target customer's insurance behavior information is also required to determine the second target insurance channel associated with the first target insurance channel. The second target insurance channel is different from the first target insurance channel, but the two have a certain correlation. In the subsequent steps, the determination of the push object is also related to the second target insurance channel.

[0043] In one embodiment, S2 specifically includes: determining the insured types of each target customer in the first historical time period based on the insured types of all target customers; determining the number of insured products of all target customers in each insurance channel based on the insured types of all target customers; determining the insurance channel with the largest number of insured products among multiple insurance channels as the first target insurance channel, and the insurance channel with the second largest number of insured products as the second target insurance channel. In this embodiment, the insurance behavior information includes the insured products. When determining the first target insurance channel, it is possible to first determine which insurance channels of the company each first target customer has insured, and then, based on the insured types of all target customers, count how many customers have insured each insurance channel, and finally select the insurance channel with the largest number of insured products as the first target insurance channel, and select the insurance channel with the second largest number of insured products as the second target insurance channel.

[0044] In one embodiment, S2 specifically includes: determining, based on the insurance behavior information, the insured types of each target customer in the first historical time period, and the intended insurance types associated with the insured types; determining the insured quantity and intended insurance quantity of all target customers in each insurance channel based on the insured types and intended insurance types of all target customers; determining the insurance channel with the largest number of insured quantities among multiple insurance channels as the first target insurance channel, and determining the insurance channel with the largest number of intended insurance quantities as the second target insurance channel. In this embodiment, the insurance behavior information includes insured types and intended insurance types. Intended insurance types refer to insurance types that a customer is more interested in but has not yet insured for various reasons. Intended insurance types are usually recommended to customers together with insured types. The two are related to each other. For example, one or more other insurance types can be recommended on the page where the user completes the insurance purchase. The user enters the relevant link to browse but ultimately does not make a transaction. In this case, these browsed insurance types constitute the intended insurance types associated with the insured type. Therefore, the intended insurance types can be determined based on the user's browsing, consulting, trial insurance and other behaviors on various platforms such as relevant APPs, text messages, phone calls, and emails.

[0045] When determining the first target insurance channel, you can first determine which insurance channels of the company each first target customer has insured, then count how many customers have insured each insurance channel based on the insurance types that all target customers have insured, and finally select the insurance channel with the largest number of insured people as the first target insurance channel. When determining the second target insurance channel, you can first determine which insurance channels of the company each first target customer has intended to insure, then count how many customers have intended to insure each insurance channel based on the insurance types that all target customers intend to insure, and finally select the insurance channel with the largest number of intended insured people as the second target insurance channel.

[0046] It should be noted that, since the customer feedback on the historical interest-bearing plan will not remain unchanged, the first target insurance channel determined based on the target customer may also change. For example, taking the feedback information of customers who have been pushed in different time periods as a reference, the first target insurance channel obtained in a period of time is the life insurance channel, and the first target insurance channel obtained in another period of time may be the health insurance channel. For another example, taking the feedback information of customers who have been pushed in different regions as a reference, the first target insurance channel corresponding to a certain region is the life insurance channel, and the first target insurance channel corresponding to another region may be the liability insurance channel. Therefore, by changing the length and time period of the first historical time period, or changing the region where the pushed customer is located, different feedback information can be obtained, and the first target insurance channel can be dynamically determined based on the feedback information, and the dynamic push of the interest-bearing plan can be realized, rather than repeatedly pushing only one insurance channel, thereby further improving the accuracy of customer acquisition and the probability of interest growth.

[0047] S3: Based on the first attribute information of the important customers in the first target insurance channel and the first behavior information of the important customers in the second historical time period, the trained customer classification model is called to classify the important customers to obtain the first important customer whose grade is greater than the first preset grade and the second important customer whose grade is not greater than the first preset grade.

[0048] Important customers refer to important customers who have high-level qualifications and credit when applying for credit cards. These customers have shorter approval time, simpler approval process and higher efficiency when submitting various applications such as credit card application, loan application and cash withdrawal application. It is easier to reach a deal by pushing interest-bearing plans to these customers. Important customers in the first target insurance channel refer to important customers who have insured in the first target insurance channel within the second historical time period, including customers who have received historical interest-bearing plans and those who have not received historical interest-bearing plans. The second historical time period can be several months or a year from the current date, or other time periods that have passed. The length and time period of the second historical time period can be selected as needed. The second historical time period overlaps with the first historical time period, and the length can be greater than or equal to the length of the first historical time period to ensure that the feedback information of the first historical time period is of reference significance, and at the same time ensure that the customer base composed of important customers in the first target insurance channel is large enough.

[0049] The embodiment of the present application takes the life insurance channel as an example for the first target insurance channel to be explained, and the data of the life insurance channel in the company are integrated to obtain the first attribute information of the important customers of the life insurance channel and the first behavior information of the important customers in the second historical time period. The first attribute information may include the age, gender, address, work, education, income, marital status and other information of the important customers of the life insurance channel, and the first behavior information may include the credit card opening behavior information (number of cards opened, card opening channel, card opening time, etc.), card usage behavior information (self-use behavior, card usage behavior of customers under the name, etc.), operation behavior information (operation behavior of interest-bearing plans in APP, SMS, emails, etc.) of the important customers of the life insurance channel. After obtaining this information, the first attribute information and the first behavior information are input into the trained customer classification model, and the customer classification model outputs the classification result of the important customers of the life insurance channel, and obtains the first important customer whose grade is greater than the first preset grade and the second important customer whose grade is not greater than the first preset grade. Assuming that all grades can include five grades of high, relatively high, medium, low, and relatively low, the first preset grade can be medium or relatively high. The higher the level, the higher the demand for interest-bearing products and the lower the aversion to interest-bearing schemes. The lower the level, the lower the demand for interest-bearing products and the higher the aversion to interest-bearing schemes.

[0050] In one embodiment, before S3, it also includes: obtaining a training sample set, the training sample set including the second attribute information of each training customer, and the second behavior information of each training customer in a third historical time period; scoring each attribute factor in the second attribute information and each behavior factor in the second behavior information, and normalizing each score to obtain normalized data of each training customer; obtaining a total score of each training customer based on the normalized data of each training customer, and determining the grade information of each training customer based on the total score of each training customer; training a customer grading model based on the second attribute information, second behavior information and grade information of each training customer to obtain a trained customer grading model.

[0051] Before classifying the repeat customers of the life insurance channel, the customer classification model is trained first. The training sample set includes the second attribute information of the training customers and the second behavior information of the training customers in the third historical time period. The training customers are a part of the repeat customers randomly selected from all the repeat customers. The length and time period of the third historical time period can be set as needed. Similarly, the second attribute information can include the training customers' age, gender, address, job, education, income, marital status and other information, which respectively constitute the attribute factors in the second attribute information; the second behavior information can include the training customers' credit card opening behavior information, card usage behavior information, operation behavior information and other information, which respectively constitute the behavior factors in the second behavior information.

[0052] First, score each attribute factor and each behavior factor. For example, for the attribute factor age, you can set the age of 18-30 years old to 2 points, 31-50 years old to 1 point, and 61 years old and above to 0 points; for the attribute factor gender, you can set the gender to be 2 points for male and 1 point for female; for the attribute factor education, you can set it to be 3 points for those below undergraduate, 1 point for undergraduate, and 2 points for masters and doctors; for the attribute factor income, you can set it to be 1 point for annual salary below a, 2 points for the range of a to b, 3 points for the range of b to c, 4 points for the range of c to d, and 5 points for those above d. And so on, the other attribute factors are scored according to the preset scoring rules, and the interval settings and score settings in each attribute factor can be adjusted as needed. Similarly, for the behavioral factor card opening quantity, you can set the quantity to 0 to get 1 point, and the quantity to 1 to get 2 points; for the behavioral factor APP operation behavior, you can set the operation to be more than 3 times in a week to get 2 points, the operation to be more than 1 time but less than 3 times in a week to get 1 point, and the operation to be 0 times in a week to get 0 points. By analogy, the other behavioral factors are also scored according to the preset scoring rules, and the interval settings and score settings in each behavioral factor can also be adjusted as needed.

[0053] After scoring each attribute factor and each behavior factor, each score is normalized, that is, each score is normalized to the interval [0, 1], and each normalized data of each training customer is obtained. The total score of each training customer is calculated, and the level information of each training customer is obtained according to the interval in which the total score is located. For example, it can be divided into five levels: high, relatively high, medium, low, and relatively low. Then the total score is compared with the preset score. If it is greater than the preset score, the second attribute information, second behavior information, and level information of the training customer are used as positive samples. If it is not greater than the preset score, it is used as a negative sample. The pre-constructed customer grading model is trained using positive samples and negative samples respectively, and the weights of each node in the customer grading model are adjusted according to the deviation between the output value and the expected value, and finally the trained customer grading model is obtained.

[0054] When it is necessary to classify important customers in the first target insurance channel, the first attribute information and first behavior information of these important customers are input into the trained customer classification model. When inputting, the relevant information of each attribute factor and each behavior factor is input in the preset format and order. The model will first score and normalize according to the preset scoring conditions and normalization conditions, and calculate the total score. Then, according to the total score, the grade information is output to realize the classification of the first important customer and the second important customer.

[0055] S4: Based on the first attribute information and the first behavior information of the first important customer, generate and push a matching first interest earning plan to the first important customer.

[0056] For the first important customer, since their level is higher, it means that they are more receptive to the interest-bearing plan. After obtaining the first attribute information and the first behavior information of the first important customer, these attributes and behaviors can be analyzed to determine the first important customer's demand points for interest-bearing products, and then the first interest-bearing plan suitable for the first important customer is generated and pushed to the first important customer. For example, the first important customer is mainly a female customer who has opened a card, has a large number of card consumption times and amounts, uses the APP frequently, and is between 30 and 40 years old. Then, interest-bearing plans about consumption incentives and new customer invitation incentives can be mainly pushed to them.

[0057] S5: Determine a third important customer from the second important customers who overlaps with the important customers in the second target insurance channel, and generate and push a matching second interest-earning plan to the third important customer based on the first attribute information and the first behavior information of the third important customer.

[0058] For the second most important customer, due to its low level, directly pushing the second interest-bearing plan to it may cause customer disgust, and further screening is required. At this time, the intersection of the second most important customer and the important customers in the second target insurance channel can be calculated, and the important customers in the intersection can be regarded as the third most important customers. Since the second target insurance channel is the insurance channel with the second largest number of insured products or the highest number of intended insurance products for target customers, the important customers in the second target insurance channel have a higher degree of recognition of the company's products compared with other insurance channels, and since the third most important customer has relevant insurance behaviors in both insurance channels, it means that he has a high degree of recognition of the company's products, so compared with important customers who are not in the intersection, his acceptance of the interest-bearing plan will be higher. After obtaining the first attribute information and the first behavior information of the third most important customer, these attributes and behaviors can be analyzed to determine the demand points of the third most important customer for interest-bearing products, and then a second interest-bearing plan suitable for the third most important customer is generated and pushed to the third most important customer. For example, the third-level customers are low-level customers who have no opened cards or have opened cards but have not activated them. Therefore, interest-earning plans regarding card opening and activation incentives can be mainly pushed to them.

[0059] Through the above process, we can achieve the targeted generation and push of interest-earning plans for the first and third important customers, which will minimize customer complaints and maximize the chance of closing a deal.

[0060] In one embodiment, S5 specifically includes: calling the trained customer classification model to classify the important customers according to the fourth attribute information of the important customers in the second target insurance channel and the fourth behavior information of the important customers in the second historical time period, and obtaining a fourth important customer whose grade is greater than the second preset grade; and determining a third important customer overlapping with the fourth important customer from the second important customers. For important customers in the second target insurance channel, before intersecting them with the second qualified important customers in the first target insurance channel, they can also be pre-screened to select the fourth important customer with a higher acceptance of the interest-bearing plan, and then intersecting the fourth important customer with the second important customer to further improve the accuracy of identifying potential customers.

[0061] Specifically, the fourth attribute information of the important customers in the second target insurance channel and the fourth behavior information of the important customers in the second historical time period are obtained. The fourth attribute information may include the age, gender, address, work, education, income, marital status and other information of the important customers in the second target insurance channel. The fourth behavior information may include the credit card opening behavior information, card use behavior information, operation behavior information and other information of the important customers in the second target insurance channel. After obtaining this information, the fourth attribute information and the fourth behavior information are input into the trained customer classification model, and the customer classification model outputs the classification result of the important customers in the second target insurance channel, and obtains the fourth important customer whose grade is greater than the second preset grade. The second preset grade may be the same as the first preset grade. The higher the level, the higher the demand for interest-bearing products and the lower the aversion to interest-bearing schemes. The lower the level, the lower the demand for interest-bearing products and the higher the aversion to interest-bearing schemes. Through the above method, the heavy customers with low demand for interest-bearing products in the second target insurance channel are first removed, and then the fourth heavy customers with relatively high demand for interest-bearing products are intersected with the second heavy customers. The third heavy customers finally obtained will have a higher acceptance of interest-bearing schemes.

[0062] In one embodiment, before S3, it also includes: determining, based on feedback information, disgusted customers whose acceptance of historical interest-bearing plans is less than a second threshold; constructing a disgusted customer portrait based on the third attribute information of the disgusted customer and the third behavior information of the disgusted customer in the first historical time period; constructing a heavy customer portrait of the heavy customers in the first target insurance type channel based on the first attribute information and the first behavior information; and removing potential disgusted heavy customers from the heavy customers of the first target insurance type channel based on the disgusted customer portrait and the heavy customer portrait.

[0063] After obtaining the feedback information of the pushed customers on the historical interest-bearing scheme in S1, not only the target customers whose acceptance of the historical interest-bearing scheme is greater than the first threshold can be determined, but also the customers who dislike the historical interest-bearing scheme whose acceptance is less than the second threshold can be determined. The second threshold can be set as needed. For example, if the highest acceptance is set to 100%, the second threshold can be 40% or 30%. The third attribute information of the disliked customers includes the age, gender, address, work, education, income, marital status and other information of the disliked customers, and the third behavior information includes the credit card opening behavior information, card use behavior information, operation behavior information and other information of the disliked customers. According to the third attribute information and the third behavior information, a disliked customer portrait can be constructed. At the same time, a heavy customer portrait is constructed according to the first attribute information and the first behavior information of the heavy customers in the first target insurance channel, and then the disliked customer portrait is used as a reference to determine the heavy customer portrait with a higher degree of consistency with the disliked customer portrait from the heavy customer portrait, and then these heavy customers are determined as potential disliked heavy customers.

[0064] After the first target insurance channel heavy customers are classified by the trained customer classification model, the first heavy customer and the second heavy customer are obtained, and the intersection of the second heavy customer and the second target insurance channel heavy customer is obtained to obtain the third heavy customer. However, since the second heavy customer has a lower grade, there may be a part of potential disgusted heavy customers. If the intersection is directly obtained with the second target insurance channel heavy customer to obtain the third heavy customer, there may still be potential disgusted heavy customers. If the second interest-bearing plan is directly pushed to them, there is still a risk of causing customer disgust. In this embodiment, disgusted customers are first determined based on the feedback information of the pushed customers, and disgusted customer portraits are constructed. At the same time, the heavy customer portraits of the first target insurance channel heavy customers are constructed. According to the two types of portraits, potential disgusted heavy customers are determined from the first target insurance channel heavy customers, first removed, and then classified to obtain the first heavy customer and the second heavy customer. This method can reduce the proportion of heavy customers who are disgusted with the interest-bearing plan among the second heavy customers, so that the risk of the third heavy customer who is disgusted with the interest-bearing plan is reduced, thereby further improving the accuracy of identifying potential customers, and reducing the risk of customer complaints, which is conducive to interest growth.

[0065] In one embodiment, after S5, it also includes: obtaining an interest-bearing target and breaking down the interest-bearing target into multiple process indicators; monitoring each process indicator, and improving the first interest-bearing plan and the second interest-bearing plan according to the monitoring data. The purpose of pushing the interest-bearing plan to the customer is to achieve the interest-bearing target, which is also the interest-bearing turnover of the comprehensive development. Breaking down the interest-bearing target into multiple process indicators can quantify the main influencing factors of the comprehensive development interest-bearing turnover, and then monitor each process indicator. When the indicator does not meet expectations, analyze the existing problems and make targeted improvements to the interest-bearing plan to achieve the completion of the interest-bearing target and the growth of the interest-bearing target.

[0066] Specifically, the Zongtuo interest-bearing transaction volume is broken down to obtain the following formula:

[0067] Zongtuo interest-bearing transaction volume = ∑ (number of qualified key customers * key customer activity * key customer penetration rate) * (number of new orders per capita + number of existing orders per capita) * average transaction amount

[0068] Among them, the active rate of important customers = the number of visitors to the important customer sharing page / the number of qualified important customers, the penetration rate of important customers for placing orders = the number of important customers for placing interest-bearing orders / the number of visitors to the important customer sharing page, the number of new account orders per capita = the number of new account orders opened by Zongtuo sales / the number of qualified important customers, the number of existing account orders per capita = the number of existing account orders opened by Zongtuo / the number of qualified important customers. The number of qualified important customers refers to the number of important customers who are qualified to receive the push of interest-bearing plans, that is, the total number of first-time important customers and third-time important customers, the number of visitors to the important customer sharing page refers to the total number of visits to these important customers on the sharing page corresponding to the interest-bearing product (multiple visits by the same important customer are counted only once), the number of important customers for placing interest-bearing orders refers to the total number of important customers who have traded interest-bearing products on the sharing page, the number of new account orders opened by Zongtuo sales refers to the total number of interest-bearing product transactions made by important customers who have not previously applied for the bank's credit card, and the number of existing account orders opened by Zongtuo refers to the total number of interest-bearing product transactions made by important customers who have previously applied for the bank's credit card. The number of qualified and important customers, the activity of important customers, the penetration rate of important customers who place orders, the average number of new orders placed per capita, the average number of existing orders placed per capita, and the average amount per order all constitute process indicators of the interest-bearing target.

[0069] Monitor each process indicator. If it is found that the value of a certain indicator does not reach the preset target value, the first and second interest-bearing plans will be improved for this indicator. For example, the interest-bearing target for the month is 400,000,000. Referring to the order situation of the previous month, the average number of orders per capita (average number of new account orders + average number of existing orders per capita) is 1.26, the average amount is 16,200, and the number of qualified re-customers is 355,267. If the interest-bearing target is to be achieved this month, the re-customer activity * order opening re-customer penetration rate needs to reach 5.5%, that is, 5.5% of re-customers need to complete the order opening to achieve the interest-bearing target. If the interest-bearing growth is to be achieved this month, the required re-customer ratio will be even higher.

[0070] Taking the values ​​of the process indicators of the previous month as the benchmark, if it is monitored that other indicators of the month are up to standard, but the activity of important customers*the penetration rate of placing orders for important customers does not reach 5.5%, it is necessary to increase the number of visitors to the important customer sharing page and the number of important customers who place interest-bearing orders. At this time, the content and incentives related to the number of transactions in the first interest-bearing plan and the second interest-bearing plan can be highlighted and adjusted to increase the number of transactions of important customers. If other indicators of the month are up to standard, but the average amount of each transaction does not reach 16,200, it is necessary to highlight and adjust the content and incentives related to the amount in the first interest-bearing plan and the second interest-bearing plan to increase the amount of each transaction of important customers. In this way, the process leakage before placing an order can be identified and the business strategy can be improved in a timely manner, so as to achieve interest-bearing growth under the condition of good user satisfaction.

[0071] Through the above embodiments, it can be seen that the method for pushing the interest-bearing scheme of the present application first determines the target customers with relatively high acceptance according to the feedback information of the currently pushed customers on the historical interest-bearing scheme, and then starts from the insurance channel to see in which insurance channel the customers with high acceptance of the historical interest-bearing scheme are mainly distributed. Then, the customers subsequently obtained from this insurance channel will also have a high acceptance of the interest-bearing scheme. Compared with searching for customers by investing in many companies, the scope of searching for customers is narrowed, and the scope is more accurate. In addition, in the first target insurance channel, according to the attributes of important customers and the behavior in the second historical time period, the customer classification model is called for classification, and the first important customer with a higher grade can be determined from all important customers. Pushing the first interest-bearing scheme to these customers can further improve the transaction probability. Finally, the second important customer is intersected with the important customers in the associated second target insurance channel. The third important customer obtained has a high degree of recognition of the company's products because he has been insured in both insurance channels. Therefore, the probability of pushing the second interest-bearing scheme to these customers causing customer disgust is low, and the use of different interest-bearing schemes from the first important customer also realizes targeted push of different types of customers, and the transaction probability is further improved. Based on the above points, this application achieves accurate acquisition of potential customers while minimizing customer complaints by associating interest-bearing plans with insurance channels, which is conducive to interest growth.

[0072] Based on the method described in the above embodiment, this embodiment will be further described from the perspective of the interest-generating scheme push device. Figure 4 The interest-generating scheme push device may include:

[0073] The first determination module 110 is used to determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold according to the feedback information of the customers on the historical interest-bearing scheme that has been pushed in the first historical time period;

[0074] A second determination module 120 is configured to determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in the multiple insurance channels during the first historical period;

[0075] A classification module 130 is used to call a trained customer classification model to classify the important customers according to the first attribute information of the important customers in the first target insurance channel and the first behavior information of the important customers in the second historical time period, and obtain a first important customer whose grade is greater than a first preset grade and a second important customer whose grade is not greater than the first preset grade;

[0076] A first generating module 140 is used to generate and push a matching first interest earning plan to the first important customer according to the first attribute information and the first behavior information of the first important customer;

[0077] The second generating module 150 is used to determine a third important customer from the second important customers who overlaps with the important customer in the second target insurance type channel, and generate and push a matching second interest earning plan to the third important customer based on the first attribute information and the first behavior information of the third important customer.

[0078] In one embodiment, the second determining module 120 includes:

[0079] A first determination submodule is used to determine the insurance type purchased by each target customer in the first historical time period according to the insurance behavior information;

[0080] The second determination submodule is used to determine the number of insurance policies purchased by all target customers in each insurance policy channel according to the insurance policies purchased by all target customers;

[0081] The third determination submodule is used to determine the insurance type channel with the largest number of insured policies among the multiple insurance type channels as the first target insurance type channel, and determine the insurance type channel with the second largest number of insured policies as the second target insurance type channel.

[0082] In one embodiment, the second determining module 120 includes:

[0083] A fourth determination submodule is used to determine, based on the insurance behavior information, the insurance types purchased by each target customer in the first historical time period, and the intended insurance types associated with the purchased insurance types;

[0084] The fifth determination submodule is used to determine the insured quantity and intended insured quantity of all target customers in each insurance channel according to the insured types and intended insured types of all target customers;

[0085] The sixth determination submodule is used to determine the insurance channel with the largest number of insured persons among the multiple insurance channels as the first target insurance channel, and determine the insurance channel with the largest number of intended insured persons as the second target insurance channel.

[0086] In one embodiment, the interest-generating scheme push device further includes:

[0087] An acquisition module, used for acquiring a training sample set, wherein the training sample set includes second attribute information of each training customer and second behavior information of each training customer in a third historical time period;

[0088] A scoring module, used for scoring each attribute factor in the second attribute information and each behavior factor in the second behavior information, and normalizing each score to obtain each normalized data of each training customer;

[0089] A third determination module is used to obtain a total score of each training client according to each normalized data of each training client, and determine the level information of each training client according to the total score of each training client;

[0090] The training module is used to train the customer grading model based on the second attribute information, the second behavior information and the grade information of each training customer to obtain the trained customer grading model.

[0091] In one embodiment, the interest-generating scheme push device further includes:

[0092] A fourth determination module is used to determine, based on the feedback information, customers who dislike the historical interest-bearing scheme and whose acceptance degree is less than a second threshold;

[0093] A first construction module is used to construct a profile of the disliked customer based on the third attribute information of the disliked customer and the third behavior information of the disliked customer in the first historical time period;

[0094] A second construction module is used to construct a customer profile of a repeat customer in the first target insurance type channel according to the first attribute information and the first behavior information;

[0095] A removal module is used to remove potential disgusting heavy customers from heavy customers of the first target insurance type channel according to the disgusting customer portrait and the heavy customer portrait.

[0096] In one embodiment, the second generation module 150 includes:

[0097] a grading submodule, configured to call the trained customer grading model to grade the important customers according to the fourth attribute information of the important customers in the second target insurance channel and the fourth behavior information of the important customers in the second historical time period, and obtain a fourth important customer whose grade is greater than the second preset grade;

[0098] The seventh determining submodule is used to determine a third guest overlapping with the fourth guest from the second guests.

[0099] In one embodiment, the interest-generating scheme push device further includes:

[0100] A disassembly module is used to obtain an interest-generating target and disassemble the interest-generating target into multiple process indicators;

[0101] The monitoring module is used to monitor various process indicators and improve the first interest-generating scheme and the second interest-generating scheme according to the monitoring data.

[0102] Different from the prior art, the interest-bearing scheme push device provided in the present application first determines the target customers with relatively high acceptance based on the feedback information of the currently pushed customers on the historical interest-bearing schemes, and then starts from the insurance channel to see in which insurance channel the customers with relatively high acceptance of the historical interest-bearing schemes are mainly distributed. Then, the customers subsequently obtained from this insurance channel will also have a high acceptance of the interest-bearing scheme. Compared with searching for customers by investing in a large number of customers, the scope of searching for customers is narrowed, and the scope is more accurate. In addition, in the first target insurance channel, according to the attributes of important customers and the behavior in the second historical time period, the customer classification model is called for classification, and the first important customer with a higher grade can be determined from all important customers. Pushing the first interest-bearing scheme to these customers can further improve the transaction probability. Finally, the second important customer is intersected with the important customers in the associated second target insurance channel. The third important customer obtained has a high degree of recognition of the company's products because he has purchased insurance in both insurance channels. Therefore, the probability of causing customer disgust when the second interest-bearing scheme is pushed to these customers is low, and the use of different interest-bearing schemes from the first important customer also realizes targeted push of different types of customers, and the transaction probability is further improved. Based on the above points, this application achieves accurate acquisition of potential customers while minimizing customer complaints by associating interest-bearing plans with insurance channels, which is conducive to interest growth.

[0103] Accordingly, the present application also provides an electronic device, such as Figure 5 As shown, the electronic device may include a radio frequency (RF) circuit 501, a memory 502 including one or more computer-readable storage media, an input unit 503, a display unit 504, a sensor 505, an audio circuit 506, a WiFi module 507, a processor 508 including one or more processing cores, and a power supply 509. Those skilled in the art will appreciate that Figure 5 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0104] The radio frequency circuit 501 can be used for receiving and sending signals during information transmission or calls. In particular, after receiving the downlink information of the base station, it is handed over to one or more processors 508 for processing; in addition, the data related to the uplink is sent to the base station. The memory 502 can be used to store software programs and modules. The processor 508 executes various functional applications and pushes interest plans by running the software programs and modules stored in the memory 502. The input unit 503 can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal input related to customer settings and function control.

[0105] The display unit 504 may be used to display information input by a client or information provided to a client and various graphical client interfaces of the server. These graphical client interfaces may be composed of graphics, texts, icons, videos and any combination thereof.

[0106] The electronic device may also include at least one sensor 505, such as a light sensor, a motion sensor, and other sensors. The audio circuit 506 includes a speaker, which may provide an audio interface between the user and the electronic device.

[0107] WiFi is a short-range wireless transmission technology. Electronic devices can help customers send and receive emails, browse web pages, and follow streaming media through WiFi module 507. It provides customers with wireless broadband Internet follow-up. Figure 5 A WiFi module 507 is shown, but it is understandable that it is not an essential component of the electronic device and can be omitted as needed without changing the essence of the application.

[0108] Processor 508 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire mobile phone. It executes various functions of the electronic device and processes data by running or executing software programs and / or modules stored in memory 502 and calling data stored in memory 502, thereby monitoring the mobile phone as a whole.

[0109] The electronic device also includes a power source 509 (such as a battery) for supplying power to various components. Preferably, the power source can be logically connected to the processor 508 through a power management system, so that the power management system can manage charging, discharging, power consumption and other functions.

[0110] Although not shown, the electronic device may also include a camera, a Bluetooth module, etc., which will not be described in detail here. Specifically in this embodiment, the processor 508 in the server will load the executable files corresponding to the processes of one or more applications into the memory 502 according to the following instructions, and the processor 508 will run the application stored in the memory 502, thereby realizing the following functions:

[0111] Determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold value according to feedback information of customers on the historical interest-bearing scheme that has been pushed in the first historical time period;

[0112] Determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in the first historical time period in multiple insurance channels;

[0113] According to the first attribute information of the important customer in the first target insurance type channel and the first behavior information of the important customer in the second historical time period, calling the trained customer classification model to classify the important customer, and obtaining a first important customer whose grade is greater than a first preset grade and a second important customer whose grade is not greater than the first preset grade;

[0114] generating, based on the first attribute information and the first behavior information of the first important customer, and pushing a matching first interest earning plan to the first important customer;

[0115] A third important customer who overlaps with an important customer in the second target insurance type channel is determined from the second important customers, and a matching second interest earning plan is generated and pushed to the third important customer based on the first attribute information and the first behavior information of the third important customer.

[0116] In the above embodiments, the description of each embodiment has its own emphasis. For the part that is not described in detail in a certain embodiment, please refer to the detailed description above and will not be repeated here.

[0117] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0118] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to implement the following functions:

[0119] Determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold value according to feedback information of customers on the historical interest-bearing scheme that has been pushed in the first historical time period;

[0120] Determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in the first historical time period in multiple insurance channels;

[0121] According to the first attribute information of the important customer in the first target insurance type channel and the first behavior information of the important customer in the second historical time period, calling the trained customer classification model to classify the important customer, and obtaining a first important customer whose grade is greater than a first preset grade and a second important customer whose grade is not greater than the first preset grade;

[0122] generating, based on the first attribute information and the first behavior information of the first important customer, and pushing a matching first interest earning plan to the first important customer;

[0123] A third important customer who overlaps with an important customer in the second target insurance type channel is determined from the second important customers, and a matching second interest earning plan is generated and pushed to the third important customer based on the first attribute information and the first behavior information of the third important customer.

[0124] The above is a detailed introduction to a method, device, electronic device and computer-readable storage medium for pushing an interest-bearing scheme provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the technical scheme and its core idea of ​​the present application; ordinary technicians in this field should understand that they can still modify the technical schemes recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents; and these modifications or replacements do not make the essence of the corresponding technical scheme deviate from the scope of the technical schemes of the embodiments of the present application.

Claims

1. A method for pushing interest-bearing schemes, characterized in that: include: Determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold value according to feedback information of customers on the historical interest-bearing scheme that has been pushed in the first historical time period; Determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in the first historical time period in multiple insurance channels; According to the feedback information, determining the averse customers whose acceptance of the historical interest-bearing scheme is less than a second threshold; Constructing a profile of the disliked customer based on the third attribute information of the disliked customer and the third behavior information of the disliked customer in the first historical time period; Based on the first attribute information and the first behavior information, construct a profile of a key customer in the first target insurance channel, wherein a key customer refers to an important customer with high-level qualifications and credit when applying for a credit card; According to the disgusting customer portrait and the disgusting customer portrait, removing potential disgusting heavy customers from the heavy customers of the first target insurance channel; According to the first attribute information of the important customer in the first target insurance type channel and the first behavior information of the important customer in the second historical time period, calling the trained customer classification model to classify the important customer, and obtaining a first important customer whose grade is greater than a first preset grade and a second important customer whose grade is not greater than the first preset grade; generating, based on the first attribute information and the first behavior information of the first important customer, and pushing a matching first interest earning plan to the first important customer; According to the fourth attribute information of the important customer in the second target insurance channel and the fourth behavior information of the important customer in the second historical time period, calling the trained customer classification model to classify the important customer, and obtaining a fourth important customer whose grade is greater than the second preset grade; Determine a third guest from the second guest that overlaps with the fourth guest; Generate and push a matching second interest earning plan to the third important customer based on the first attribute information and the first behavior information of the third important customer; Obtaining an interest-generating target, and breaking down the interest-generating target into a plurality of process indicators; Monitor each process indicator, and improve the first interest-generating scheme and the second interest-generating scheme according to the monitoring data.

2. The method for pushing interest-bearing scheme according to claim 1, characterized in that: The step of determining a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in the first historical time period in multiple insurance channels includes: Determine, based on the insurance behavior information, the types of insurance purchased by each target customer during the first historical time period; Determine the number of insurance policies purchased by all target customers in each insurance channel based on the insurance policies purchased by all target customers; The insurance type channel with the largest number of insured policies among the multiple insurance type channels is determined as the first target insurance type channel, and the insurance type channel with the second largest number of insured policies is determined as the second target insurance type channel.

3. The method for pushing interest-bearing schemes according to claim 1, wherein the step of determining a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information of the target customer in the first historical time period in the multiple insurance channels comprises: Determine, based on the insurance behavior information, the insurance types purchased by each target customer in the first historical time period, and the intended insurance types associated with the purchased insurance types; Determine the number of insurance policies that all target customers have purchased and the number of insurance policies that they intend to purchase in each insurance channel based on the types of insurance policies that all target customers have purchased and the types of insurance policies that they intend to purchase; The insurance channel with the largest number of insured persons among the multiple insurance channels is determined as the first target insurance channel, and the insurance channel with the largest number of intended insured persons is determined as the second target insurance channel.

4. The method for pushing interest-bearing scheme according to claim 1, characterized in that: Before the step of calling the trained customer classification model to classify the important customers according to the first attribute information of the important customers in the first target insurance channel and the first behavior information of the important customers in the second historical time period to obtain the first important customers whose grades are greater than the first preset grade and the second important customers whose grades are not greater than the first preset grade, the method further includes: Acquire a training sample set, wherein the training sample set includes second attribute information of each training customer and second behavior information of each training customer in a third historical time period; Scoring each attribute factor in the second attribute information and each behavior factor in the second behavior information, and normalizing each score to obtain normalized data for each training customer; According to each normalized data of each training client, a total score of each training client is obtained, and the level information of each training client is determined according to the total score of each training client; The customer classification model is trained based on the second attribute information, the second behavior information and the grade information of each training customer to obtain a trained customer classification model.

5. An interest-generating scheme push device, characterized in that: include: A first determination module is used to determine target customers whose acceptance of the historical interest-bearing scheme is greater than a first threshold value based on feedback information of customers on the historical interest-bearing scheme that has been pushed in a first historical time period; The second determination module is used to determine a first target insurance channel and a second target insurance channel associated with the first target insurance channel from the multiple insurance channels according to the insurance behavior information generated by the target customer in multiple insurance channels during the first historical time period; determine the disgusting customers whose acceptance of the historical interest-bearing plan is less than a second threshold according to the feedback information; construct a disgusting customer portrait according to the third attribute information of the disgusting customer and the third behavior information of the disgusting customer during the first historical time period; construct a heavy customer portrait of the heavy customer in the first target insurance channel according to the first attribute information and the first behavior information, wherein the heavy customer refers to an important customer with high-level qualifications and credit when applying for a credit card; remove potential disgusting heavy customers from the heavy customers of the first target insurance channel according to the disgusting customer portrait and the heavy customer portrait; a classification module, configured to classify the important customers by calling the trained customer classification model according to the first attribute information of the important customers in the first target insurance channel and the first behavior information of the important customers in the second historical time period, and obtain a first important customer whose grade is greater than a first preset grade and a second important customer whose grade is not greater than the first preset grade; A first generating module, configured to generate and push a matching first interest earning plan to the first important customer according to the first attribute information and the first behavior information of the first important customer; The second generating module is used to call the trained customer grading model to grade the important customers according to the fourth attribute information of the important customers in the second target insurance type channel and the fourth behavior information of the important customers in the second historical time period, so as to obtain a fourth important customer whose grade is greater than the second preset grade; determine a third important customer overlapping with the fourth important customer from the second important customers; generate and push a matching second interest-earning plan to the third important customer according to the first attribute information and the first behavior information of the third important customer; obtain an interest-earning target and decompose the interest-earning target into multiple process indicators; monitor each process indicator, and improve the first interest-earning plan and the second interest-earning plan according to the monitoring data.

6. An electronic device, characterized in that: It comprises a memory and a processor; the memory stores an application, and the processor is used to run the application in the memory to execute the steps in the interest-bearing scheme push method described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in the interest-bearing scheme push method according to any one of claims 1 to 4.

Images