Pre-sales service issuing method and device, computer equipment and storage medium

Abstract

The invention discloses a pre-sales service issuing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a to-be-predicted data set corresponding to a customer; the to-be-predicted data set comprises different information data related to the customer; inputting the to-be-predicted data set into a preset customer preference prediction model to obtain a service preference corresponding to the customer; and according to a preset service preference issuing rule, on the basis of the service preference, issuing a service category product to the customer. According to the scheme, the service preference is accurately recommended through the preset customer preference prediction model, service preference recommendation can be made in combination with the to-be-predicted data set formed by combining various data of different customers, the individual requirements of the customers can be met, and the customer satisfaction and loyalty are improved; the service category products are issued according to the preset service preference issuing rule, so that the optimal allocation of resources can be realized in a limited service prediction, unnecessary waste is avoided, and the resource utilization rate is improved.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence technology and financial technology, and in particular to a pre-sales service issuance method, device, computer equipment and storage medium. Background Art

[0002] With the development of big data and artificial intelligence technology, more and more companies have begun to pay attention to the value of data, especially in the financial insurance industry. It has become a trend to use advanced data analysis technology to improve service quality and efficiency. Among them, auto insurance, as an important branch of the insurance industry, faces a large number of customer groups and diverse service needs. The traditional service distribution method often cannot meet the personalized needs of modern consumers. The traditional approach usually determines the service strategy based on manual experience or simple customer classification. This method is not only inefficient, but also easily causes uneven resource allocation, resulting in low customer satisfaction. More specifically, most companies are still at a relatively early stage in pre-sales service, relying mainly on manual judgment or simple rule engines to determine service delivery strategies. This approach has obvious limitations, including: first, it cannot effectively capture subtle differences among customers, such as the specific service preferences of customers of different age groups; second, it lacks full utilization of historical data and cannot make more accurate service recommendations based on customers’ historical behaviors and service redemptions; finally, it does not take into account the company’s overall service budget, which may lead to excessive concentration of resources on a few customers while neglecting a wider range of service groups.

[0003] Therefore, those skilled in the art are in urgent need of finding a new technical solution to solve the above problems. Summary of the invention

[0004] Based on this, a pre-sales service issuance method, apparatus, computer equipment and storage medium are provided to solve the technical problems of uneven resource allocation and low customer satisfaction caused by the service issuance method in the prior art.

[0005] A pre-sales service issuance method, comprising: Acquire a data set to be predicted corresponding to the customer; the data set to be predicted includes different information data related to the customer; Inputting the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer; According to the preset service preference issuance rules, service category products are issued to the customer based on the service preferences.

[0006] A pre-sale service issuing device, comprising: An acquisition module is used to acquire a data set to be predicted corresponding to a customer; the data set to be predicted includes different information data related to the customer; An input module, used to input the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer; The issuing module is used to issue service category products to the customer based on the service preference according to preset service preference issuing rules.

[0007] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned pre-sale service issuing method when executing the computer program.

[0008] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the pre-sales service issuance method is implemented.

[0009] The above-mentioned pre-sales service issuance method, device, computer equipment and storage medium obtain the data set to be predicted corresponding to the customer; the data set to be predicted contains different information data related to the customer; the data set to be predicted is input into the preset customer preference prediction model to obtain the service preference corresponding to the customer; and service category products are issued to the customer based on the service preference according to the preset service preference issuance rules. This solution accurately recommends service preferences through the preset customer preference prediction model, and can make service preference recommendations based on the data set to be predicted composed of multiple data of different customers, which can meet the personalized needs of customers and improve customer satisfaction and loyalty; issuing service category products according to the preset service preference issuance rules can achieve the optimal allocation of resources within the limited service budget, avoid unnecessary waste, and improve resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0011] Figure 1 This is a schematic diagram of an application environment of a pre-sales service distribution method in one embodiment of the present invention; Figure 2 is a process diagram of a pre-sale service issuing method in one embodiment of the present invention; Figure 3 It is a structural schematic diagram of a pre-sale service issuing device in one embodiment of the present invention; Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] The pre-sale service distribution method provided by the present invention can be applied in Figure 1 In an application environment, a client communicates with a server via a network. The client may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server may be implemented as an independent server or a server cluster consisting of multiple servers.

[0014] In one embodiment, if Figure 2 As shown, a pre-sales service issuance method is provided, which is applied in Figure 1 The server in the example is used as an example to illustrate the process, including the following steps: S10, obtaining a dataset to be predicted corresponding to the customer; the dataset to be predicted includes different information data related to the customer.

[0015] Understandably, the dataset to be predicted contains information data that can provide services related to customers, and the information data may include gender, vehicle age, vehicle model, vehicle price, home and commercial identification, whether the customer is an online customer, customer age, customer level, whether the current vehicle is new energy, whether there are children, income level, occupation, marriage, education, and the receipt, use and write-off of various services in the auto insurance APP; This embodiment obtains a comprehensive customer portrait through data fusion technology.

[0016] S20, inputting the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer.

[0017] Understandably, the preset customer preference prediction model can be a model successfully trained based on the XGBoost algorithm, whose full name is eXtreme Gradient Boosting, which is an improvement on the gradient boosting algorithm. It uses the gradient descent algorithm to optimize the integration of decision trees, and introduces regularization terms and second-order derivative information to improve the performance and generalization ability of the model. In addition, the XGBoost algorithm is used to predict customer service preferences. Compared with traditional machine learning algorithms, XGBoost has better prediction accuracy; service preferences can be services that current customers like, such as issuing annual inspection agency, car wash coupons, gas coupons, video memberships, music memberships and other service category products to potential customers (in the prediction model, service category products can be matched by digital labels, such as marking service category products in sequence as "0", "1", "2", "3", "4"). Service preferences can assist enterprises in improving users' intention to purchase products, such as giving corresponding service category products to potential customers before they purchase the corresponding products (pre-sales).

[0018] S30, issuing service category products to the customer based on the service preference according to preset service preference issuance rules.

[0019] It is understandable that the service preferences predicted by the prediction model need to be issued to potential customers in a prescribed manner, and the preset service preference issuance rules can issue service category products to potential customers before selling corporate products based on the service issuance coefficient calculated by the service preferences.

[0020] In the embodiment of step S10 to step S30, the service preference is accurately recommended by the preset customer preference prediction model, and the service preference recommendation can be made by combining the data set to be predicted composed of multiple data of different customers, which can meet the personalized needs of customers and improve customer satisfaction and loyalty; issuing service category products according to the preset service preference issuance rules can achieve the optimal allocation of resources within the limited service budget, avoid unnecessary waste, and improve resource utilization. Furthermore, before the data set to be predicted is input into the preset customer preference prediction model to obtain the service preference corresponding to the customer, it also includes: Acquire different customer data sets from different channel systems, and classify the data in the customer data sets to obtain classified data; The classified data is subjected to feature screening using a preset screening method to obtain target feature data to be input into a preset initial customer preference prediction model.

[0021] Understandably, the channel system may include a CRM system, a website log, and a database. The customer data set obtained by each channel system may be different, and the specific data may include customer basic information, online active information, historical service write-off information, and other related data; the classified data may classify different customer data sets in advance, such as combining related data into one category, such as classifying them into gender, vehicle age, vehicle model, vehicle price, home and commercial identification, whether it is a WeChat customer, customer age, customer level, whether it is a new energy vehicle, whether it has children, income level, occupation, marriage, education, and the use, use, and write-off of various services in the auto insurance APP; the preset screening method may include one or more of a single factor analysis method, a correlation analysis method, a feature coding method, a method for selecting key features, and a feature selection method; This embodiment obtains customer data sets from multiple channel systems to ensure data diversity and comprehensiveness. Different channels may contain different information data, which together constitute a complete picture of customer behavior. Diverse data helps a cluster model to more accurately capture customer preferences. Data from different sources may reveal customer behavior patterns and preference differences in different scenarios. Classifying data in customer data sets helps to group customers with similar features or behavior patterns into one category, thereby facilitating subsequent feature screening and model training. The classified data enables the prediction model to more accurately predict the preferences of specific types of customers, and the prediction model can perform customized learning and prediction for each category.

[0022] Furthermore, the method of using a preset screening method to perform feature screening on the classified data to obtain target feature data to be input into a preset initial customer preference prediction model includes: After analyzing the association relationship between the single classified data and the classified data as variable data, and analyzing the correlation relationship between the classified data and the classified data, determine the first initial target feature data from the classified data; Encode the first initial target feature data using a preset feature encoding method to obtain second initial target feature data; Non-important feature data in the second initial target feature data are removed using a preset recursive method to obtain the target feature data to be input into a preset initial customer preference prediction model.

[0023] Understandably, the distribution characteristics, value range, and abnormal value data of a single categorical data are analyzed, and based on the analysis results, the first initial target feature data that is associated with the variable is screened out, such as determining the first initial target feature data with a relationship based on the data type stored in a preset data table, or using the Pearson correlation coefficient or the Spearman correlation coefficient to measure the linear correlation between the feature and the variable; the preset feature encoding method can use One-Hot Encoding. For numerical features (such as customer age, online stay time, click-through rate, etc.), scaling and standardization can be performed to eliminate the influence of different feature dimensions. For category features (such as gender, car model, etc.), One-Hot Encoding can be used. Encoding is performed, and One-Hot Encoding creates a new set of binary features, each of which represents a specific category in the original features. In the new binary feature vector, only one of the feature values ​​is allowed to be 1 (indicating that the corresponding category exists), and all other feature values ​​are 0 (indicating that the corresponding category does not exist); using the preset recursive method includes using all available second initial target features to train a model (such as logistic regression, support vector machine, etc.), and calculating the weight or coefficient of each second initial target feature, sorting the features according to the weight or coefficient of the second initial target feature, usually the feature with the smallest weight or coefficient is considered to be unimportant feature data, deleting one or more features with the smallest weight or coefficient in the sorting, and retraining the model with the features of the second initial target feature, repeating the above steps (in a recursive manner) until the required number of second initial target features is reached or the features cannot be further eliminated; the preset initial customer preference prediction model is a model structure that has not been successfully trained yet; The processing process of this embodiment can gradually screen out key features - target feature data, and perform feature encoding and feature selection, thereby reducing the complexity of the prediction model and improving the prediction model performance and prediction accuracy.

[0024] Furthermore, before obtaining target feature data to be input into a preset initial customer preference prediction model after feature screening of the classified data using a preset screening method, the method further includes: The classified data is divided into a plurality of training sets and at least one validation set according to the relationship between the current time point and the historical time point; the number of the training sets is greater than the number of the validation sets.

[0025] Understandably, the current time point may be the time point for predicting the service preference of the user, and the historical time point may be the relevant time point in the historical time period. For example, the training set and the prediction set are divided according to the expiration month of the auto insurance sales. The current time may be July 1, and the lists of the three historical months, i.e., the expiration months of April, May, and June, may be used as the training set, and the list of the expiration months of July may be used as the prediction set. This embodiment uses training data from recent historical time points, and this data usually contains more information related to the current market environment, such as policy changes, market activities, consumer behavior, etc., which is conducive to the training of the prediction model; using training data close to the prediction time can reduce model deviations caused by outdated data and improve the accuracy of the prediction model.

[0026] Furthermore, after the target feature data to be input into the preset initial customer preference prediction model is obtained by performing feature screening on the classified data using a preset screening method, the method further includes: Presetting target model parameters corresponding to the preset customer preference prediction model; Adjusting the model parameters in the preset initial customer preference prediction model by using the model function corresponding to the preset initial customer preference prediction model, so as to adjust the model parameters to the target model parameters; After the classification data as a training set is input into a preset initial customer preference prediction model adjusted to target model parameters, it is determined that the model indicators meet the target expected indicators, and the preset customer preference model is obtained.

[0027] Understandably, the target model parameters can be parameters that define a parameter range, and then each parameter combination is trained and cross-validation is used to evaluate its performance. Common target model parameters include learning rate, maximum depth, minimum sub-node weight and, sub-sample ratio, etc.; the model function corresponding to the initial customer preference prediction model can be defined in advance, such as predicting based on model parameters such as multi-classification problems, number of categories, multi-classification logarithmic loss, and returning the predicted service preference category index; on each parameter combination of the tuning grid, cross-validation (such as K-fold cross-validation) is used to evaluate whether the model model indicators meet the target expected indicators. Cross-validation evaluates the generalization ability of the prediction model by dividing the training data into K subsets, and then taking turns to use one of the subsets as the validation set and the remaining subsets as the training set. In addition, appropriate evaluation indicators (such as accuracy, F1 score, AUC, etc.) are used to measure the performance of the prediction model on the validation set; This embodiment can use the preset initial customer preference prediction model - XGBoost model for multi-classification training, and obtain the preset service preference prediction model through grid optimization of the corresponding model function, which can significantly improve the prediction accuracy and generalization ability of the model and provide the company with more accurate customer service.

[0028] Furthermore, before issuing the service category product to the customer based on the service preference according to the preset service preference issuance rule, the method further includes: The write-off rate of each type of service corresponding to the to-be-predicted data set corresponding to the customer is calculated using a preset first calculation formula; wherein the first calculation formula is calculated as follows: , is the write-off rate of the i-th type of service, is the usage cost of the service corresponding to the i-th service, is the actual delivery cost of the i-th type of service; The preset second calculation formula is used to calculate the renewal rate of each type of service customer actually issued last month corresponding to the to-be-predicted data set corresponding to the customer, and the renewal rate of each type of service customer is calculated with the write-off rate of each type of service to obtain the service issuance coefficient; wherein the second calculation formula is: , It is the renewal rate of each type of service customers actually issued last month.

[0029] Understandably, the write-off rate, also known as the write-off rate, refers to the ratio of the amount of bad debts reduced by write-off operations to the actual total amount of accounts receivable in financial management. It reflects the ability of the company to successfully write off bad debts and reduce losses when dealing with accounts receivable; the service issuance coefficient is used in the model to evaluate the issuance order of service category products; the renewal rate refers to the situation of maintaining the purchase of a product between the customer and the company. In the insurance field, the renewal rate can be the renewal rate; the corresponding service usage cost and actual issuance cost can be calculated through the channel system where the company is located; The service issuance coefficient calculated in this embodiment can be used for the issuance order of service category products. During the calculation process, actual calculations can be performed based on the actual service issuance and verification records of the previous month and the corresponding budget data, thereby obtaining an accurate and appropriate service issuance coefficient.

[0030] Furthermore, the issuing of service category products to the customer based on the service preferences according to the preset service preference issuing rules includes: After sorting according to the size of the service issuance coefficient, a permutation matrix is ​​obtained; Determine the corresponding target customer group according to the service preferences; Service category products are distributed to the target customer group in sequence according to the distribution coefficients in the arrangement matrix until the budget for a preset time period reaches the target budget or all customers in the target customer group are distributed.

[0031] Understandably, the arrangement matrix can be a matrix generated by sorting the sizes of service issuance coefficients, such as the service issuance coefficient of annual inspection agency is 5, the service issuance coefficient of car wash coupons is 4, the service issuance coefficient of gas coupons is 3, the service issuance coefficient of video members is 2, and the service issuance coefficient of music members is 1; the target customer group can be the intended group in the customer group, specifically, the customers who match the service preferences are determined by service preferences, and the determined customers are used as a target customer group; the size of the issuance coefficients in the arrangement matrix is ​​used to issue service category products to the target customer group in turn, and after the current service category product is issued, it is postponed to the next size of service category products until the company's service budget for the month is used up or all customers in the target customer group are issued, wherein the budget for the preset time period can be the service budget for the month, each service issuance coefficient corresponds to a service category, and each service category corresponds to a service category product; Through sorting and screening, this embodiment can give priority to issuing service category products to customers who are most likely to accept the service, thereby improving the efficiency and effectiveness of marketing activities; sorting according to the size of the service issuance coefficient can more reasonably allocate limited marketing resources to ensure that resources are used most effectively; by issuing products that meet the service preferences of target customers, customer satisfaction and loyalty can be increased, thereby promoting sustained business growth; budget limits for preset time periods can ensure cost control in the company's marketing activities and avoid waste of resources caused by excessive investment, that is, it can optimize the service issuance strategy within the budget range to ensure efficient use of resources; through precision marketing and high-quality services, companies can stand out in the fierce market competition and enhance their market position and brand influence.

[0032] In summary, the above provides a pre-sales service issuance method, which obtains a set of data to be predicted corresponding to a customer; the set of data to be predicted contains different information data related to the customer; the set of data to be predicted is input into a preset customer preference prediction model to obtain a service preference corresponding to the customer; and service category products are issued to the customer based on the service preference according to the preset service preference issuance rules. The technical solution accurately recommends service preferences through a preset customer preference prediction model, and can make service preference recommendations based on a set of data to be predicted composed of multiple data of different customers, which can meet the personalized needs of customers and improve customer satisfaction and loyalty; issuing service category products according to the preset service preference issuance rules can achieve the optimal allocation of resources within a limited service budget, avoid unnecessary waste, and improve resource utilization; providing targeted services to customers, that is, personalized service experience can increase the company's market competitiveness and attract more potential customers; the automated service category product issuance process reduces the need for manual operation, reduces operating costs, and improves work efficiency; by comprehensively analyzing different information data of customers, including multi-dimensional data such as basic customer information, online active information, and historical service write-off information, accurately predict the service preferences of each customer, and combine the company's service budget to automatically calculate the optimal service issuance strategy using a planning algorithm, so as to maximize customer satisfaction; maximized customer satisfaction and lower operating costs help promote business growth and development, and bring more profit space to the enterprise. It should be understood that the size of the sequence number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0033] In one embodiment, a pre-sale service issuing device is provided, and the pre-sale service issuing device corresponds one-to-one with the pre-sale service issuing method in the above embodiment. Figure 3 As shown, the pre-sale service issuing device includes an acquisition module 11, an input module 12 and an issuing module 13. The detailed description of each functional module is as follows: An acquisition module 11 is used to acquire a dataset to be predicted corresponding to a customer; the dataset to be predicted includes different information data related to the customer; An input module 12, used to input the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer; The issuing module 13 is used to issue service category products to the customer based on the service preference according to preset service preference issuing rules.

[0034] Furthermore, the pre-sale service issuing device further comprises: A classification module, used to obtain different customer data sets from different channel systems, and classify the data in the customer data sets to obtain classified data; The feature screening module is used to perform feature screening on the classification data using a preset screening method to obtain target feature data to be input into a preset initial customer preference prediction model.

[0035] Furthermore, the feature screening module includes: A first determining unit, configured to determine first initial target feature data from the classification data after analyzing the association relationship between the single classification data and the classification data as variable data, and analyzing the correlation relationship between the classification data and the classification data; an encoding unit, configured to encode the first initial target feature data using a preset feature encoding method to obtain second initial target feature data; A removing unit is used to remove non-important feature data in the second initial target feature data using a preset recursive method to obtain the target feature data to be input into a preset initial customer preference prediction model.

[0036] Furthermore, the feature screening module includes: A division unit is used to divide the classification data into multiple training sets and at least one verification set according to the relationship between the current time point and the historical time point; the number of the training sets is greater than the number of the verification sets.

[0037] Furthermore, the pre-sale service issuing device further comprises: A setting module, used to pre-set target model parameters corresponding to the preset customer preference prediction model; An adjustment module, configured to adjust the model parameters in the preset initial customer preference prediction model through a model function corresponding to the preset initial customer preference prediction model, so as to adjust the model parameters to the target model parameters; The determination module is used to determine whether the model indicators meet the target expected indicators after the classification data as a training set is input into the preset initial customer preference prediction model adjusted to the target model parameters, so as to obtain the preset customer preference model.

[0038] Furthermore, the pre-sale service issuing device further comprises: The first calculation module is used to calculate the write-off rate of each type of service corresponding to the to-be-predicted data set corresponding to the customer using a preset first calculation formula; wherein the first calculation formula is calculated as follows: , is the write-off rate of the i-th type of service, is the usage cost of the service corresponding to the i-th service, is the actual delivery cost of the i-th type of service; The second calculation module is used to calculate the renewal rate of each type of service customer actually issued last month corresponding to the to-be-predicted data set corresponding to the customer using a preset second calculation formula, and calculate the renewal rate of each type of service customer and the write-off rate of each type of service to obtain the service issuance coefficient; wherein the second calculation formula is , It is the renewal rate of each type of service customers actually issued last month.

[0039] Furthermore, the issuing module includes: A sorting unit, used to sort the services according to the service issuance coefficients to obtain a permutation matrix; A second determining unit, configured to determine a corresponding target customer group according to the service preference; The distribution unit is used to distribute service category products to the target customer group in sequence according to the distribution coefficients in the arrangement matrix until the budget of the preset time period reaches the target budget or all customers in the target customer group are distributed.

[0040] For the specific definition of the pre-sale service issuance device, please refer to the definition of the pre-sale service issuance method above, which will not be repeated here. Each module in the above-mentioned pre-sale service issuance device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0041] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data involved in the pre-sales service issuance method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a pre-sales service issuance method is implemented.

[0042] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the pre-sales service issuing method in the above embodiment are implemented, such as Figure 2 Steps S10 to S30 shown include the following steps: Acquire a data set to be predicted corresponding to the customer; the data set to be predicted includes different information data related to the customer; Inputting the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer; According to the preset service preference issuance rules, service category products are issued to the customer based on the service preferences.

[0043] Alternatively, when the processor executes the computer program, the functions of each module / unit of the pre-sales service issuing device in the above embodiment are realized, for example Figure 3 The functions of modules 11 to 14 are shown in Figure 1. To avoid repetition, they will not be described here.

[0044] In one embodiment, a computer-readable storage medium is provided on which a computer program is stored. When the computer program is executed by a processor, the steps of the pre-sales service issuance method in the above embodiment are implemented, such as Figure 2 Steps S10 to S30 shown include the following steps: Acquire a data set to be predicted corresponding to the customer; the data set to be predicted includes different information data related to the customer; Inputting the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer; According to the preset service preference issuance rules, service category products are issued to the customer based on the service preferences.

[0045] Alternatively, when the computer program is executed by the processor, the functions of each module / unit of the pre-sales service issuing device in the above embodiment are realized, for example Figure 3 The functions of modules 11 to 14 are shown in Figure 1. To avoid repetition, they will not be described here.

[0046] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0047] 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. 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.

[0048] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A pre-sale service issuance method, characterized in that: include: Acquire a data set to be predicted corresponding to the customer; the data set to be predicted includes different information data related to the customer; Inputting the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer; According to the preset service preference issuance rules, service category products are issued to the customer based on the service preferences.

2. The pre-sale service issuance method according to claim 1, characterized in that: Before inputting the to-be-predicted data set into a preset customer preference prediction model to obtain a service preference corresponding to the customer, the method further includes: Obtain different customer data sets from different channel systems, and classify the data in the customer data sets to obtain classified data; The classified data is subjected to feature screening using a preset screening method to obtain target feature data to be input into a preset initial customer preference prediction model.

3. The pre-sale service issuance method according to claim 2, characterized in that: The method of using a preset screening method to perform feature screening on the classified data to obtain target feature data to be input into a preset initial customer preference prediction model includes: After analyzing the association relationship between the single classified data and the classified data as variable data, and analyzing the correlation relationship between the classified data and the classified data, determine the first initial target feature data from the classified data; Encode the first initial target feature data using a preset feature encoding method to obtain second initial target feature data; Non-important feature data in the second initial target feature data are removed using a preset recursive method to obtain the target feature data to be input into a preset initial customer preference prediction model.

4. The pre-sale service issuance method according to claim 2, characterized in that: Before the method uses a preset screening method to perform feature screening on the classified data to obtain target feature data to be input into a preset initial customer preference prediction model, it also includes: The classified data is divided into a plurality of training sets and at least one validation set according to the relationship between the current time point and the historical time point; the number of the training sets is greater than the number of the validation sets.

5. The pre-sale service issuance method according to claim 2, characterized in that: After the method uses a preset screening method to perform feature screening on the classified data to obtain target feature data to be input into a preset initial customer preference prediction model, the method further includes: Presetting target model parameters corresponding to the preset customer preference prediction model; Adjusting the model parameters in the preset initial customer preference prediction model by using the model function corresponding to the preset initial customer preference prediction model, so as to adjust the model parameters to the target model parameters; After the classification data as a training set is input into a preset initial customer preference prediction model adjusted to target model parameters, it is determined that the model indicators meet the target expected indicators, and the preset customer preference model is obtained.

6. The pre-sale service issuance method according to claim 1, characterized in that: Before issuing the service category product to the customer based on the service preference according to the preset service preference issuance rule, the method further includes: The write-off rate of each type of service corresponding to the to-be-predicted data set corresponding to the customer is calculated using a preset first calculation formula; wherein the first calculation formula is calculated as follows: , is the write-off rate of the i-th type of service, is the usage cost of the service corresponding to the i-th service, is the actual delivery cost of the i-th type of service; The preset second calculation formula is used to calculate the renewal rate of each type of service customer actually issued last month corresponding to the to-be-predicted data set corresponding to the customer, and the renewal rate of each type of service customer is calculated with the write-off rate of each type of service to obtain the service issuance coefficient; wherein the second calculation formula is: , It is the renewal rate of each type of service customers actually issued last month.

7. The pre-sale service issuance method according to claim 1 or 6, characterized in that: The issuing of service category products to the customer based on the service preferences according to the preset service preference issuing rules includes: After sorting according to the size of the service issuance coefficient, a permutation matrix is ​​obtained; Determine the corresponding target customer group according to the service preferences; Service category products are distributed to the target customer group in sequence according to the distribution coefficients in the arrangement matrix until the budget for a preset time period reaches the target budget or all customers in the target customer group are distributed.

8. A pre-sale service distribution device, characterized in that: include: An acquisition module is used to acquire a data set to be predicted corresponding to a customer; the data set to be predicted includes different information data related to the customer; An input module, used to input the data set to be predicted into a preset customer preference prediction model to obtain a service preference corresponding to the customer; The issuing module is used to issue service category products to the customer based on the service preference according to preset service preference issuing rules.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the pre-sales service issuance method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the pre-sales service issuance method according to any one of claims 1 to 7 is implemented.

Images