Method and device for determining customer group of product and computer program product

Abstract

The invention discloses a method and a device for determining a customer group of a product, and a computer program product. Relates to the field of financial science and technology, and the method comprises the steps: obtaining the customer information of M to-be-distributed customers, carrying out the customer group division of the M to-be-distributed customers according to the M groups of customer information, and obtaining N to-be-distributed customer groups; y customer screening algorithms are obtained, customers are screened from the N to-be-distributed customer groups through the Y customer screening algorithms, N * Y groups of candidate customers are obtained, and each group of candidate customers is associated with customer information and quota data of to-be-distributed products; and calculating evaluation data of each group of candidate customers, sending the N * Y evaluation data and the N * Y groups of candidate customers to the client, and constructing an operation strategy of the to-be-distributed products according to the quota data of the to-be-distributed products associated with the N * Y groups of candidate customers. According to the invention, the technical problem of low product distribution accuracy during product distribution of different customer groups in the prior art is solved.

Description

Technical Field

[0001] This application relates to the field of financial technology, and more specifically, to a method, apparatus, and computer program product for determining the customer base of a product. Background Technology

[0002] In the current financial market environment, financial institutions face increasingly severe competitive challenges. To attract and retain customers, traditional methods such as price reductions or discounts have gradually lost their appeal, prompting financial institutions to shift towards more innovative and differentiated strategies, such as stimulating customer interest by offering financial products like "red envelopes" (cash gifts). However, given limited product availability and a diverse target customer base, how to effectively allocate limited resources to achieve optimal results during implementation has become a pressing problem for the financial industry.

[0003] Currently, most product recommendation schemes employ a fixed, single target customer group and machine learning algorithms, generating recommendation lists through modeling and training. While this approach can quickly respond to established strategies, it has significant drawbacks in practical applications: the effects between different target customer groups and the performance comparisons of different machine learning algorithms are difficult to visualize, making it hard to determine the most effective solution; once the customer group selection is determined, subsequent adjustments are challenging; and the recommendation algorithm, once deployed, may cease to be the optimal option over time, thus reducing recommendation efficiency.

[0004] There is currently no effective solution to the technical problem of low product allocation accuracy when allocating products to different customer groups in related technologies. Summary of the Invention

[0005] The main objective of this application is to provide a method, apparatus, and computer program product for determining the customer group of a product, so as to solve the technical problem of low product allocation accuracy when allocating products to different customer groups in related technologies.

[0006] To achieve the above objectives, according to one aspect of this application, a method for determining the customer base of a product is provided. The method includes: acquiring customer information for M customers to be assigned, and dividing the M customers to be assigned into N customer groups based on the M customer information, wherein each customer group to be assigned is associated with a customer group type, and M and N are positive integers; acquiring Y customer screening algorithms, and using the Y customer screening algorithms to screen customers from the N customer groups to be assigned, obtaining N*Y groups of candidate customers, wherein each group of candidate customers is associated with customer information and credit data for products to be assigned, the credit data for products to be assigned referring to the number of products allocated to each group of candidate customers; calculating evaluation data for each group of candidate customers, and sending the N*Y evaluation data and the N*Y groups of candidate customers to a client, wherein upon receiving the N*Y evaluation data and the N*Y groups of candidate customers, the client constructs an operational strategy for the products to be assigned based on the credit data for the products to be assigned associated with the N*Y groups of candidate customers.

[0007] Further, selecting customers from N unassigned customer groups using Y customer screening algorithms to obtain N*Y groups of candidate customers includes: obtaining allocation weight parameters; extracting customer group weight parameters associated with the customer group types of the N unassigned customer groups from the allocation weight parameters to obtain N customer group weight parameters; and using each customer group weight parameter to perform weighted calculations on the customer group data of each unassigned customer group to obtain N first customer groups; extracting algorithm weight parameters associated with each customer group type from the allocation weight parameters to obtain N sets of algorithm weight parameters, where each set of algorithm weight parameters includes Y sub-algorithm weight parameters corresponding to the Y customer screening algorithms; using each set of algorithm weight parameters to perform weighted calculations on the customer group data of each first customer group to obtain N groups of second customer groups, where each group of second customer groups includes Y second processing sub-customer groups; for a group of second customer groups, using each customer screening algorithm to screen customers in each second processing sub-customer group of the second customer group to obtain Y groups of candidate customers associated with the second customer group; and combining the Y groups of candidate customers associated with each group of second customer groups to obtain N*Y groups of candidate customers.

[0008] Furthermore, obtaining the allocation weight parameters includes: receiving adjustment parameters sent by the client; if the adjustment parameters indicate a first mode, determining the weight parameters sent by the client as the allocation weight parameters; if the adjustment parameters indicate a second mode, obtaining historical allocation weight parameters, and determining the allocation weight parameters based on the historical allocation weight parameters.

[0009] Further, determining the allocation weight parameters based on historical allocation weight parameters includes: obtaining historical candidate customers for historical time periods to obtain N*Y groups of historical candidate customers, and obtaining historical evaluation data for each group of historical candidate customers to obtain N*Y historical evaluation data, where each group of historical candidate customers refers to a group of historical candidate customers selected by a customer screening algorithm; calculating the sum of the N*Y historical evaluation data to obtain the total historical evaluation data, and calculating the ratio of the total historical evaluation data to N*Y to obtain the historical evaluation ratio; calculating the ratio of each historical evaluation data in the N*Y historical evaluation data to the historical evaluation ratio to obtain N*Y performance deviation data; and adjusting the historical allocation weight parameters using each performance deviation data in the N*Y performance deviation data to obtain the allocation weight parameters.

[0010] Furthermore, the historical allocation weight parameters are adjusted using each of the N*Y performance deviation data points to obtain the allocation weight parameters. This includes: acquiring learning parameters; calculating the difference between each historical performance deviation data point and a preset parameter to obtain N*Y deviation differences; multiplying each deviation value in the N*Y deviation differences by the learning parameter to obtain N*Y deviation products; multiplying each deviation product by the historical allocation weight parameters to obtain N*Y algorithm weight parameters, where the N*Y algorithm weight parameters are used to indicate the algorithm weight parameters of the Y customer screening algorithms associated with each customer group type; and determining the customer group weight parameters associated with each customer group type to be allocated based on the ratio of each algorithm weight parameter to each performance deviation data point to obtain N*Y customer group weight parameters.

[0011] Furthermore, based on the M sets of customer information, the M customers to be assigned are divided into customer groups to obtain N customer groups to be assigned. This includes: converting the format of each set of customer information to obtain M sets of initial processed customer data, where the format conversion method refers to converting the unstructured information included in each set of customer information into structured information; extracting features representing customer attributes from the M sets of initial processed customer data to obtain M customer features; and dividing the M customers to be assigned into customer groups based on the M customer features to obtain N customer groups to be assigned.

[0012] Further, the calculation of evaluation data for each group of candidate customers includes: for a group of candidate customers, obtaining customer information for K unassigned customers in each group of candidate customers, resulting in K groups of customer information, where K is a positive integer; extracting customer information representing preset attributes from each group of customer information in the K groups of customer information, resulting in K groups of customer attribute information, and normalizing the K groups of customer attribute information to obtain K groups of customer attribute data, wherein each group of customer attribute data includes at least one of the following: indirect attribute data, direct attribute data, and periodic attribute data; obtaining attribute weight coefficients, and using attribute weight coefficients to weight the customer group data of each customer attribute data in each group of customer attribute data, resulting in K groups of weighted attribute data; calculating the sum of the K groups of weighted attribute data to obtain the total attribute data, and calculating the ratio of the total attribute data K to obtain the evaluation data corresponding to a group of candidate customers.

[0013] To achieve the above objectives, according to another aspect of this application, an apparatus for determining the customer group of a product is provided. The apparatus includes: a first acquisition unit, configured to acquire customer information of M customers to be allocated, and to divide the M customers to be allocated into N customer groups based on the M customer information, wherein each customer group to be allocated is associated with a customer group type, and M and N are positive integers; a second acquisition unit, configured to acquire Y customer screening algorithms, and to screen customers from the N customer groups to be allocated using the Y customer screening algorithms to obtain N*Y groups of candidate customers, wherein each group of candidate customers is associated with customer information and credit data of products to be allocated, the credit data of products to be allocated referring to the number of products allocated to each group of candidate customers; and a calculation unit, configured to calculate evaluation data for each group of candidate customers, and to send the N*Y evaluation data and the N*Y groups of candidate customers to a client, wherein upon receiving the N*Y evaluation data and the N*Y groups of candidate customers, the client constructs an operational strategy for the products to be allocated based on the credit data of the products to be allocated associated with the N*Y groups of candidate customers.

[0014] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein a method for determining a customer group that controls the device where the computer-readable storage medium is located to execute any of the above-described products is provided when the executable program is running.

[0015] According to another aspect of the present invention, an electronic device is also provided, including one or more processors and a memory, the memory storing an executable program, and the processor for running the program, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement a method for determining a customer group of any of the above-described products.

[0016] According to another aspect of the present invention, a computer program product is also provided, the computer program product including a computer program, wherein when the computer program is executed by a processor, a method for determining a customer group that implements any of the above-described products is provided.

[0017] In this embodiment, the customer groups for the product are determined by acquiring customer information for M potential customers and dividing them into N potential customer groups based on the M customer information groups. Each potential customer group is associated with a customer group type, and M and N are positive integers. Y customer filtering algorithms are then acquired and used to filter customers from the N potential customer groups, resulting in N*Y candidate customer groups. Each candidate customer group is associated with customer information and the credit limit data for the products to be allocated. The credit limit data for the products to be allocated refers to the number of products allocated to each candidate customer group. Evaluation data for each candidate customer group is then calculated. The system sends N*Y evaluation data and N*Y groups of candidate customers to the client. Upon receiving the N*Y evaluation data and N*Y groups of candidate customers, the client constructs an operational strategy for the products to be allocated based on the credit limit data of the products associated with the N*Y groups of candidate customers. This solves the technical problem of low product allocation accuracy when allocating products to different customer groups in related technologies. By dividing the customers to be allocated into groups based on customer information, multiple customer groups are obtained. A customer screening algorithm is then used to screen customers from each customer group to obtain multiple groups of candidate customers, thereby improving the technical effect of product allocation accuracy for different customer groups. Attached Figure Description

[0018] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0019] Figure 1 It is a hardware structure block diagram of a computer terminal (or mobile device) for determining the customer group of a product.

[0020] Figure 2 This is a flowchart of a method for determining the customer group of a product according to an embodiment of this application;

[0021] Figure 3 This is a schematic diagram of a system for determining the customer base of a product according to an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of a product dispensing device provided according to an embodiment of this application;

[0023] Figure 5 This is a schematic diagram of a product allocation processing method provided according to an embodiment of this application;

[0024] Figure 6 This is a schematic diagram of a device for determining the customer group of a product according to an embodiment of this application;

[0025] Figure 7 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties. For example, this system has interfaces with relevant users or organizations to provide users with corresponding operation data for them to choose to agree to or refuse automated decision results. Before obtaining relevant information, a request for obtaining the information needs to be sent to the aforementioned user or organization through the interface, and the relevant information is obtained after receiving consent from the aforementioned user or organization; if the user chooses to refuse, the expert decision-making process is initiated. Users can view the purpose of data use in real time through authorization decoding and have the right to withdraw authorization or delete data at any time. After the authorization is withdrawn, the system will terminate the relevant data processing within 24 hours.

[0029] It should be noted that the information collected in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation access points for users to choose to authorize use or refuse use.

[0030] Example 1

[0031] According to an embodiment of this application, a method embodiment for determining a customer group of a product is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0032] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 This is a hardware structure block diagram of a computer terminal (or mobile device) used for determining the customer group for implementing a product, such as... Figure 1 As shown, computer terminal 10 (or mobile device) may include one or more ( Figure 1 (Illustrated as 102a, 102b, ..., 102n) Processor 102 (processor 102 may include, but is not limited to, a microprocessor MCU (Microcontroller Unit) or a programmable gate array (FPGA), etc.), memory 104 for storing data, and transmission device 106 for communication functions. In addition, it may include: a display, input / output interface (I / O interface), Universal Serial Bus (USB) port (which may be included as one of the ports of a BUS bus), network interface, keyboard, cursor control device, power supply and / or camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0033] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0034] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for determining the customer group of the product in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the above-mentioned method for determining the customer group of the product. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, financial institution intranets, local area networks, mobile communication networks, and combinations thereof.

[0035] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network interface controller (NIC) and a network interface, which can be connected to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a radio frequency (RF) module, used for wireless communication with the Internet.

[0036] The display can be, for example, a touchscreen liquid crystal display (LCD), which allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0037] Under the aforementioned operating environment, this application provides the following: Figure 2 The method for determining the customer base for the products shown. Figure 2 This is a flowchart of a method for determining the customer group of a product according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0038] Step S201: Obtain customer information for M customers to be assigned, and divide the M customers to be assigned into customer groups according to the M groups of customer information to obtain N customer groups to be assigned. Each customer group to be assigned is associated with a customer group type, and M and N are positive integers.

[0039] Specifically, customer information refers to a data set that describes customer characteristics and behaviors, and may include quantitative or qualitative information related to behavioral patterns, such as personal customer information and historical transaction records. Customer segmentation refers to classifying customers to be allocated according to different customer group types. At this point, based on specific characteristics and indicators in the customer information, data analysis and machine learning techniques can be used to identify customer groups with similar behavioral patterns or value potential, thus obtaining multiple customer groups to be allocated. These customer group types may include high-contribution customers, new customers, long-tail customers, etc. Through customer segmentation, different customer groups can be identified and defined, thereby determining different product strategies. For example, "high-contribution customers" require a larger allocation of product funds for maintenance.

[0040] Step S202: Obtain Y customer screening algorithms, and use the Y customer screening algorithms to screen customers from N customer groups to be allocated, to obtain N*Y groups of candidate customers. Each group of candidate customers is associated with customer information and credit data of products to be allocated. The credit data of products to be allocated refers to the number of products allocated to each group of candidate customers.

[0041] It should be noted that customer screening algorithms refer to a set of algorithms used to select customers from each customer group who respond positively to products. These algorithms may include random forests, logistic regression, gradient boosting trees, etc., and can make predictions based on the customer characteristics and behavioral patterns of each customer in the pending customer group to identify potential target customers. The credit limit data for pending product allocation refers to the specific number of products that financial institutions allocate to candidate customers.

[0042] After obtaining the customer groups to be allocated, corresponding customer screening algorithms can be used to conduct in-depth analysis of each group, identifying those most likely to benefit from specific products or most likely to respond to promotions, thus obtaining multiple candidate customer groups. By selecting candidate customers and allocating product quotas, data-driven decision-making can be made, making strategies more precise and efficient, reducing blind spending and wasted opportunities.

[0043] Step S203: Calculate the evaluation data for each group of candidate customers, and send N*Y evaluation data and N*Y groups of candidate customers to the client. When the client receives N*Y evaluation data and N*Y groups of candidate customers, it constructs the operation strategy for the products to be allocated based on the credit limit data of the products to be allocated associated with the N*Y groups of candidate customers.

[0044] Specifically, the evaluation data is a set of value indicators derived from in-depth analysis of candidate customers. It quantifies the potential contribution and value of each group of candidate customers. The evaluation data can include multiple dimensions, such as direct value (net revenue generated by customer use of the product), indirect value (the increase in customer activity multiplied by the potential value coefficient), and long-term value (the predicted customer lifetime value multiplied by the loyalty factor). After obtaining candidate customers, the evaluation data for each candidate customer can be sent to the client. The client then constructs an operational strategy based on the evaluation data and the available credit data for the products to be allocated. This operational strategy can cover elements such as product allocation rules, timing, target customers, and expected results, maximizing product benefits while ensuring effective resource utilization and avoiding over-investment or waste.

[0045] The method for determining the customer group of a product provided in this application embodiment involves obtaining customer information for M customers to be allocated, and dividing the M customers to be allocated into N customer groups based on the M groups of customer information. Each customer group to be allocated is associated with a customer group type, and M and N are positive integers. Y customer screening algorithms are obtained, and customers are screened from the N customer groups to be allocated using the Y customer screening algorithms to obtain N*Y groups of candidate customers. Each group of candidate customers is associated with customer information and the credit limit data for products to be allocated. The credit limit data for products to be allocated refers to the number of products allocated to each group of candidate customers. Evaluation data for each group of candidate customers is calculated. N*Y evaluation data points and N*Y groups of candidate customers are sent to the client. Upon receiving the N*Y evaluation data points and N*Y groups of candidate customers, the client constructs an operational strategy for the products to be allocated based on the credit limit data of the products to be allocated associated with the N*Y groups of candidate customers. This solves the technical problem of low product allocation accuracy when allocating products to different customer groups in related technologies. By dividing the customers to be allocated into customer groups based on customer information, multiple customer groups to be allocated are obtained. Customer screening algorithms are then used to screen customers from each customer group to obtain multiple groups of candidate customers, thereby achieving the technical effect of improving the accuracy of product allocation to different customer groups.

[0046] Optionally, in the method for determining the customer group of the product provided in this application embodiment, selecting customers from N unassigned customer groups using Y customer screening algorithms to obtain N*Y groups of candidate customers includes: obtaining allocation weight parameters; extracting customer group weight parameters associated with the customer group types of the N unassigned customer groups from the allocation weight parameters to obtain N customer group weight parameters; and using each customer group weight parameter to perform weighted calculation on the customer group data of each unassigned customer group to obtain N first customer groups; and extracting algorithm weight parameters of the customer screening algorithm associated with each customer group type from the allocation weight parameters to obtain... N sets of algorithm weight parameters are provided, where each set of algorithm weight parameters includes Y sub-algorithm weight parameters corresponding to Y customer screening algorithms. The customer data of each first customer group is weighted using each set of algorithm weight parameters to obtain N sets of second customer groups, where each second customer group includes Y second processing sub-customer groups. For each set of second customer groups, each customer screening algorithm is used to screen customers in each second processing sub-customer group within the second customer group, resulting in Y sets of candidate customers associated with the second customer group. The Y sets of candidate customers associated with each set of second customer groups are combined to obtain N*Y sets of candidate customers.

[0047] It should be noted that the weighting parameters are key parameters for controlling the relative importance and resource allocation of different customer groups and algorithms in operational activities. These include customer group weighting parameters and algorithm weighting parameters, which guide the optimized allocation of resources and the priority ranking of model paths. Specifically, the customer group weighting parameters are related to the customer group type to be assigned and are used to quantify the importance of each customer group in the overall strategy. These parameters determine how resources are allocated among target customer groups. For example, "high-contribution customers" receive higher weights due to their higher value. The algorithm weighting parameters are related to the customer screening algorithm associated with each customer group type. These parameters are used to adjust the processing priority and allocation ratio of different algorithms for the same customer group sample to achieve resource optimization at the algorithm level.

[0048] First, pre-set or dynamically adjusted allocation weight parameters are obtained. Then, customer group weight parameters associated with the customer group types to be allocated are extracted from these parameters, forming customer group weight parameters. These parameters quantify the priority of different customer groups in product allocation; for example, the weights of high-contribution customers, new customers, and long-tail customers differ according to business needs. Next, the customer group data for each customer group is weighted using these weight parameters to generate multiple primary customer groups. This ensures that high-weight customer groups occupy a more significant position in subsequent data processing and model calculations, improving the targeting and efficiency of resources.

[0049] Furthermore, algorithm weight parameters corresponding to each customer group type are extracted from the assigned weight parameters to form algorithm weight parameters. Each set of algorithm weight parameters includes sub-algorithm weight parameters corresponding to each customer screening algorithm. These parameters are used to adjust the influence of different algorithms when processing the same customer group data to adapt to the characteristics of different customer groups. At this point, the algorithm weight parameters can be used to perform a secondary weighted calculation on the customer group data of the first customer group to obtain the corresponding second customer group. Each second customer group is further subdivided into a second processing sub-customer group, further refining the granularity of customer screening and ensuring the applicability and effectiveness of the algorithm in specific customer groups.

[0050] For each second customer group, a separate customer screening algorithm is used to filter the second customer subgroup, resulting in candidate customers associated with the second customer group. Then, the candidate customers associated with each second customer group are combined to generate a final candidate customer list, providing an accurate customer list for subsequent activities.

[0051] This embodiment achieves dynamic optimization and combination of resources among different customer groups and algorithms by adjusting both customer group weight parameters and algorithm weight parameters. This ensures that limited product quotas are allocated to the most promising customers and the most effective algorithm paths, enabling activities to more accurately target various customer groups and improve customer experience and satisfaction. The "horse race" mechanism automatically adjusts weight parameters based on feedback business results through dynamic competition among different paths, avoiding the rigidity of a single strategy and promoting model self-optimization.

[0052] Optionally, in the method for determining the customer group of the product provided in the embodiments of this application, obtaining the allocation weight parameter includes: receiving the adjustment parameter sent by the client; if the adjustment parameter indicates a first mode, determining the weight parameter sent by the client as the allocation weight parameter; if the adjustment parameter indicates a second mode, obtaining the historical allocation weight parameter, and determining the allocation weight parameter based on the historical allocation weight parameter.

[0053] It should be noted that in order to determine the allocation weight parameters, the obtained adjustment parameters must first be identified. If the adjustment parameters indicate the first mode, the adjustment parameters indicate the manual adjustment mode. That is, when the adjustment parameters sent by the client indicate that manual adjustment is required, the subsequent customer screening and product allocation decisions need to be made based on the weight parameters (i.e., allocation weight parameters) provided by the client. In other words, the latest weight parameters provided by the client are directly received and used as allocation weight parameters to adjust the current customer group allocation ratio and algorithm path weight. In this mode, the weight parameters are completely determined by manual input and are mainly used to deal with business needs or market changes in a specific period.

[0054] When the adjustment parameter indicates the second mode, that is, when the condition parameter is in automatic adjustment mode, it is necessary to obtain the historical allocation weight parameters, and then adjust and determine the new allocation weight parameters based on the latest business data and market feedback (such as customer response, business indicators, transaction volume data, etc.) to ensure that the resource allocation between different customer groups and algorithms is always in the best state and adapts to market changes.

[0055] This embodiment receives adjustment parameters and adjusts the weighting parameters according to their instructions, enabling flexible switching between manual and automatic adjustment modes. This satisfies the need for manual intervention during specific periods while ensuring automated optimization capabilities in daily operations, avoiding resource waste and improving overall efficiency.

[0056] Optionally, in the method for determining the customer group of the product provided in this application embodiment, determining the allocation weight parameter based on the historical allocation weight parameter includes: obtaining historical candidate customers for a historical time period to obtain N*Y groups of historical candidate customers, and obtaining historical evaluation data for each group of historical candidate customers to obtain N*Y historical evaluation data, wherein each group of historical candidate customers refers to a group of historical candidate customers selected by a customer screening algorithm; calculating the sum of the N*Y historical evaluation data to obtain the total historical evaluation data, and calculating the ratio of the total historical evaluation data to N*Y to obtain the historical evaluation ratio; calculating the ratio of each historical evaluation data in the N*Y historical evaluation data to the historical evaluation ratio to obtain N*Y performance deviation data; and adjusting the historical allocation weight parameter using each performance deviation data in the N*Y performance deviation data to obtain the allocation weight parameter.

[0057] Specifically, when the adjustment parameter indicates the second parameter, the first step is to obtain the set of potential target customers identified through various customer screening algorithms (such as random forest, logistic regression, etc.) during historical activities, resulting in multiple groups of historical candidate customers. Simultaneously, quantitative evaluation indicators generated based on the performance of these historical candidate customers during activities are obtained, i.e., historical evaluation data. Then, the historical evaluation data of these candidate customers are summarized to obtain the sum of historical evaluation data. Finally, the historical evaluation ratio is calculated by dividing the sum of historical evaluation data by the product of N*Y (i.e., the number of groups).

[0058] Furthermore, to quantify the relative effectiveness of different customer groups and algorithm paths in historical performance and identify which paths contribute more than the average, the ratio of each historical evaluation data point to the historical evaluation ratio can be calculated to obtain the corresponding performance deviation data. It should be noted that the performance deviation data R can be calculated using the following formula. n×y :

[0059]

[0060] Among them, E ny This represents the contribution value of a unit sample generated by the nth customer group using the yth algorithm (i.e., historical evaluation data).

[0061] Furthermore, the historical allocation weight parameters are adjusted based on each performance deviation data point to obtain the allocation weight parameters corresponding to each performance deviation data point. For example, the weight of high-performing paths is increased, and the weight of low-performing paths is decreased, thereby obtaining new allocation weight parameters.

[0062] This embodiment determines the allocation weight parameters based on historical allocation weight parameters, which can more accurately adjust the allocation weights and concentrate limited resources on strategies most likely to bring high returns. It can manage its product allocation recommendation strategy more intelligently and efficiently, which not only improves customer satisfaction and the success rate of activities, but also achieves optimal resource allocation.

[0063] Optionally, in the method for determining the customer group of a product provided in this application embodiment, adjusting the historical allocation weight parameters using each performance deviation data point in N*Y performance deviation data to obtain the allocation weight parameters includes: obtaining learning parameters; calculating the difference between each historical performance deviation data point and a preset parameter to obtain N*Y deviation differences; calculating the product of each deviation value in the N*Y deviation differences and the learning parameter to obtain N*Y deviation products; calculating the product of each deviation product and the historical allocation weight parameters to obtain N*Y algorithm weight parameters, wherein the N*Y algorithm weight parameters are used to indicate the algorithm weight parameters of the Y customer screening algorithms associated with each customer group type; and determining the customer group weight parameters associated with the customer group type of each customer group to be allocated based on the ratio of each algorithm weight parameter to each performance deviation data point to obtain N*Y customer group weight parameters.

[0064] Specifically, the learning parameters guide the updating of weight parameters based on historical performance deviation data. A larger learning parameter results in a more significant adjustment of the weight parameters; conversely, a smaller learning parameter leads to a more conservative adjustment. The learning parameter balances the model's adaptability and stability, avoiding instability caused by over-adjustment. Historical performance deviation data is obtained by comparing the performance (such as conversion rate, revenue contribution, etc.) of each group of historical candidate customers with historical evaluation ratios over a historical period. This data quantifies the relative performance of different customer selection algorithms and customer groups in the past.

[0065] To adjust the historical weighting parameters, we can first calculate the difference between the historical performance deviation data and the preset parameters to obtain the deviation difference. Then, we can multiply each deviation difference by the learning parameters to obtain the deviation product. A large deviation product indicates that a larger adjustment is needed, while a small deviation product indicates that it can remain stable or be fine-tuned.

[0066] Furthermore, the product of each deviation product and the historical allocation weight parameter is calculated to obtain the algorithm weight parameter. This means adjusting the relative importance of each customer screening algorithm based on historical performance and learning parameters to ensure the accuracy and efficiency of the screening process. Then, the customer group weight parameter associated with each customer group type is determined based on the ratio of each algorithm weight parameter to each performance deviation data point. This yields the customer group weight parameter, which optimizes the resource allocation strategy for different customer groups based on the algorithm's performance across them. Adjusting the algorithm weight parameter changes the influence of different algorithms in the customer screening process, while adjusting the customer group weight parameter affects the resource allocation ratio among different customer groups. It should be noted that the customer group weight parameter can be calculated using the following formula:

[0067]

[0068] The algorithm weight parameters can be calculated using the following formula:

[0069]

[0070] in, R represents the historical weighting parameters, α represents the learning parameters, and weight oscillations are avoided through gradual adjustment. ny E indicates that historical performance deviates from the data. ny This indicates historical assessment data.

[0071] This embodiment achieves dynamic adjustment of algorithm weight parameters and customer group weight parameters by combining historical performance deviation data and learning parameters, ensuring continuous optimization and adaptability of the model and avoiding resource waste.

[0072] Optionally, in the method for determining the customer group of the product provided in this application embodiment, dividing the M customers to be assigned into N customer groups based on the M groups of customer information includes: converting the format of each group of customer information to obtain M groups of initial processed customer data, wherein the format conversion method refers to converting the unstructured information included in each group of customer information into structured information; extracting features representing customer attributes from the M groups of initial processed customer data to obtain M customer features; and dividing the M customers to be assigned into N customer groups based on the M customer features.

[0073] When segmenting customers to be assigned, customer information includes unstructured information such as text descriptions, images, and emails. This information needs to be processed, that is, unstructured information without a fixed format or data structure needs to be converted and parsed into usable structured data. For example, text parsing, image recognition, and speech-to-text conversion can be performed to obtain the initial customer data for processing.

[0074] Then, customer attribute features, such as income level and consumption preferences, are extracted from the above data to describe customer characteristics, resulting in multiple customer features. Based on these features, the customers to be assigned are then segmented into customer groups, that is, the customer group is subdivided according to different attributes and needs, ultimately forming sets from different customer groups. Each set represents a customer group with similar attributes and needs, thus obtaining the customer groups to be assigned.

[0075] This embodiment significantly improves the speed and consistency of data processing by standardizing customer information and effectively segmenting customer groups, reducing processing difficulties and errors caused by inconsistent data formats. By extracting customer attribute characteristics, it is possible to gain a deeper understanding of each customer's characteristics, providing rich data support for subsequent personalized services and precision, and achieving efficient allocation and maximum utilization of resources.

[0076] Optionally, in the method for determining the customer group of the product provided in this application embodiment, calculating the evaluation data for each group of candidate customers includes: for a group of candidate customers, obtaining customer information of K customers to be assigned in each group of candidate customers to obtain K groups of customer information, where K is a positive integer; extracting customer information representing preset attributes from each group of customer information in the K groups of customer information to obtain K groups of customer attribute information, and normalizing the K groups of customer attribute information to obtain K groups of customer attribute data, wherein each group of customer attribute data includes at least one of the following: indirect attribute data, direct attribute data, and periodic attribute data; obtaining attribute weight coefficients, and using attribute weight coefficients to perform weighted calculations on the customer group data of each customer attribute data in each group of customer attribute data to obtain K groups of weighted attribute data; calculating the sum of K groups of weighted attribute data to obtain the total attribute data, and calculating the ratio of the total attribute data K to obtain the evaluation data corresponding to a group of candidate customers.

[0077] Specifically, when calculating the evaluation data, it is first necessary to obtain the customer information of these customers to be assigned, and extract information representing the customer's economic attributes in operational activities, thus obtaining customer attribute information. This includes direct attribute data (such as transaction volume, deposit balance, etc.), indirect attribute data (such as potential value, customer activity, etc.), and periodic attribute data (such as quarterly revenue, annual contribution, etc.). Then, this data is normalized to convert customer attribute information of different ranges and units into data on the same scale, avoiding differences in numerical magnitude.

[0078] Furthermore, attribute weight coefficients are obtained to quantify the contribution of different preset attributes to the total customer value. By weighting the customer attribute information using these coefficients, weighted attribute data is obtained. This data integrates the influence of direct, indirect, and periodically preset attributes, providing a more comprehensive reflection of customer value. Finally, the sum of the weighted attribute data is divided by the total number of customers to be assigned, thus obtaining the evaluation data. It should be noted that the evaluation data can be calculated using the following formula:

[0079]

[0080] Among them, K ny Let Vk be the number of customers selected along the path, and Vk be the value component of the k-th customer. as well as These represent direct value (the net benefit generated by the customer using the product, i.e.) Indirect value (the increase in customer activity multiplied by the potential value coefficient, i.e.) And long-term value (customer lifetime value multiplied by the loyalty factor, i.e.) The weighting coefficients of ).

[0081] This embodiment calculates and evaluates data, which ensures that resource and product allocation is effectively focused on the customer groups that generate the most benefits, making more scientific and reasonable product allocation decisions and improving the efficiency and success rate of the activities.

[0082] This application also provides a system for determining the customer base of a product. Figure 3 This is a schematic diagram of a system for determining the customer group of a product according to an embodiment of this application, such as... Figure 3 As shown, the system includes: a data input device, a product allocation device, and a data output device. The data input device 1 is used to collect product data of products to be allocated and customer information of customers to be allocated, and pushes the product data and customer information to the product allocation device for processing. The product allocation device is used to perform product allocation processing for multiple customer groups, output the result data of candidate customers, and push it to the downstream data output device. The data output device is used to receive the result data of candidate customers and output it to external systems for use.

[0083] Figure 4 This is a schematic diagram of a product dispensing device provided according to an embodiment of this application, such as... Figure 4 As shown, the product distribution device includes a data receiving module, an adjustment switch switching module, a strategy parsing module, a customer group distribution module, an algorithm execution module, a data output module, an effect evaluation module, a parameter adjustment module, and a parameter storage module.

[0084] The system comprises several modules: a data receiving module for receiving user-inputted target customer group categories, discount product information, product limits, and customer information; an adjustment switch module for switching between manual and automatic adjustment modes, which typically operates in automatic mode, where the machine learning model adjusts and stores customer group weight parameters based on performance evaluations; and a manual adjustment mode for specific tasks, using custom customer group weight parameters for model computation. After the specific task is completed, the system switches back to automatic adjustment mode. A strategy parsing module handles data processing (such as customer profile data, business indicator data, and transaction volume data), feature extraction (such as customer attribute features, deposit and loan features, fee income features, and contribution features), and customer group weight parameter parsing and calculation. A customer group allocation module is used to classify customers into different customer groups. The target customer group is identified, and the sample ratio of the corresponding target customer group is adjusted according to the customer group weight parameter. The algorithm execution module is used to perform model calculations by combining multiple machine learning algorithms, and adjusts the sample ratio of the corresponding "customer group + algorithm" path according to the algorithm weight parameter. Different targets and different algorithm combinations form multiple paths of the model, and multiple paths are "competed" simultaneously. The data output module is used to output the recommended customer list and product allocation quota data (i.e., candidate customer information). The effect evaluation module evaluates the actual business effect generated by different "customer group + algorithm" paths based on the evaluation data of each group of candidate customers. The parameter adjustment module increases / decreases the algorithm weight parameter and customer group weight parameter according to the "competition" result of the effect evaluation to optimize the model. The parameter storage module is used to store the algorithm weight parameter and customer group weight parameter.

[0085] The above system is applied to the processing method of optional product allocation. Figure 5 This is a schematic diagram of a product allocation processing method provided according to an embodiment of this application, such as... Figure 5 As shown, the method includes:

[0086] Data processing: Responsible for data cleaning, data filtering, data format conversion, and other processing.

[0087] Feature extraction: Read the processed data and perform feature extraction according to the rules, including important features such as basic customer attributes, main business indicators, recent transaction data, and overall customer contribution.

[0088] Parameter selection: Read the customer group weight parameters and algorithm weight parameters from the parameter storage module, and select the relevant parameters according to the adjustment switch mode. For example, in manual adjustment mode, the weight parameters entered by the user are selected, and in automatic adjustment mode, the weight parameters automatically optimized by the model are selected.

[0089] The customer group allocation module has N customer group nodes, which are divided into different customer group nodes such as high-contribution customer group 241, new customer group, and long-tail customer group according to preset rules. Each customer group node filters samples according to the customer group weight parameter. Assuming that the number of high-contribution customer groups is Cn and the weight parameter of high-contribution customer groups is Pn, then the number of high-contribution customer groups to be selected for model training is Cn*Pn.

[0090] The algorithm execution module has M algorithm nodes, each receiving sample data generated from N customer group nodes, forming a total of M*N product allocation recommendation paths. Each "customer group + algorithm" path has a corresponding weight parameter. For example, when the weight parameter of "high-contribution customer group - random forest algorithm" is Pnm, the number of customer samples for the "high-contribution customer group - random forest algorithm" path is Cn*Pn*Pnm. Therefore, the number of customer samples used for model training using the random forest algorithm is C1*P1*P11+C2*P2*P21+C3*P3*P31+...+Cn*Pn*Pn1. The M algorithm nodes simultaneously train the model, outputting candidate customers, and then constructing an operational strategy for the products to be allocated based on the credit limit data of the products associated with the candidate customers.

[0091] This embodiment divides the customers to be assigned into multiple customer groups based on customer information. Then, a customer screening algorithm is used to screen customers from each customer group to obtain multiple candidate customers. This achieves the technical effect of improving the accuracy of product allocation to different customer groups.

[0092] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0093] Example 2

[0094] This application also provides a device for determining the customer group of a product. It should be noted that this device can be used to execute the method for determining the customer group of a product provided in this application. The following describes the device for determining the customer group of a product provided in this application.

[0095] According to an embodiment of this application, an apparatus for determining a customer group for implementing the above-described product is also provided. Figure 6 This is a schematic diagram of a device for determining the customer group of a product according to an embodiment of this application, such as... Figure 6 As shown, the device includes: a first acquisition unit 60, a second acquisition unit 61, and a calculation unit 62.

[0096] The first acquisition unit 60 is used to acquire customer information of M customers to be assigned, and to divide the M customers to be assigned into customer groups according to the M groups of customer information to obtain N customer groups to be assigned. Each customer group to be assigned is associated with a customer group type, and M and N are positive integers.

[0097] The second acquisition unit 61 is used to acquire Y customer screening algorithms, and to screen customers from N customer groups to be allocated through the Y customer screening algorithms to obtain N*Y groups of candidate customers. Each group of candidate customers is associated with customer information and quota data of products to be allocated. The quota data of products to be allocated refers to the number of products allocated to each group of candidate customers.

[0098] The calculation unit 62 is used to calculate the evaluation data of each group of candidate customers and send N*Y evaluation data and N*Y groups of candidate customers to the client. When the client receives N*Y evaluation data and N*Y groups of candidate customers, it constructs the operation strategy of the product to be allocated based on the quota data of the product to be allocated associated with the N*Y groups of candidate customers.

[0099] The customer group determination device for the product provided in this application embodiment acquires customer information of M customers to be allocated by a first acquisition unit 60, and divides the M customers to be allocated into N customer groups according to the M groups of customer information, wherein each customer group to be allocated is associated with a customer group type, and M and N are positive integers; the second acquisition unit 61 acquires Y customer screening algorithms, and filters customers from the N customer groups to be allocated by the Y customer screening algorithms respectively, to obtain N*Y groups of candidate customers, wherein each group of candidate customers is associated with customer information and quota data of products to be allocated, and the quota data of products to be allocated refers to the number of products allocated to each group of candidate customers; the calculation unit 62 calculates the number of products allocated to each group of candidate customers. The evaluation data of candidate customers is sent to the client along with N*Y evaluation data and N*Y groups of candidate customers. Upon receiving the N*Y evaluation data and N*Y groups of candidate customers, the client constructs an operational strategy for the products to be allocated based on the credit limit data of the products associated with the N*Y groups of candidate customers. This solves the technical problem of low product allocation accuracy when allocating products to different customer groups in related technologies. By dividing the customers to be allocated into multiple groups based on customer information, a customer screening algorithm is used to filter customers from each group to obtain multiple groups of candidate customers, thereby improving the technical effect of improving the accuracy of product allocation to different customer groups.

[0100] Optionally, in the customer group determination device for the product provided in this application embodiment, the second acquisition unit 61 includes: a first acquisition module, used to acquire allocation weight parameters, extract customer group weight parameters associated with customer group types of N customer groups to be allocated from the allocation weight parameters to obtain N customer group weight parameters, and use each customer group weight parameter to perform weighted calculation on the customer group data of each customer group to be allocated to obtain N first customer groups; a first extraction module, used to extract algorithm weight parameters of customer screening algorithms associated with each customer group type from the allocation weight parameters to obtain N sets of algorithm weight parameters, wherein each set of algorithms The weight parameters include Y sub-algorithm weight parameters corresponding to Y customer screening algorithms; the first calculation module is used to perform weighted calculations on the customer data of each first customer group using the weight parameters of each algorithm to obtain N second customer groups, where each second customer group includes Y second processing sub-customer groups; the screening module is used to screen customers in each second processing sub-customer group of a second customer group using each customer screening algorithm to obtain Y candidate customers associated with the second customer group; the combination module is used to combine the Y candidate customers associated with each second customer group to obtain N*Y candidate customers.

[0101] Optionally, in the device for determining the customer group of the product provided in the embodiments of this application, the second acquisition unit 61 includes: a receiving module, used to receive adjustment parameters sent by the client, and when the adjustment parameters indicate a first mode, to determine the weight parameters sent by the client as allocation weight parameters; and a second acquisition module, used to acquire historical allocation weight parameters when the adjustment parameters indicate a second mode, and to determine allocation weight parameters based on the historical allocation weight parameters.

[0102] Optionally, in the customer group determination device for the product provided in this application embodiment, the second acquisition unit 61 includes: a third acquisition module, used to acquire historical candidate customers for a historical time period, obtain N*Y groups of historical candidate customers, and acquire historical evaluation data for each group of historical candidate customers, obtaining N*Y historical evaluation data, wherein each group of historical candidate customers refers to a group of historical candidate customers selected by a customer screening algorithm; a second calculation module, used to calculate the sum of N*Y historical evaluation data, obtain the total historical evaluation data, and calculate the ratio of the total historical evaluation data to N*Y, obtaining the historical evaluation ratio; a third calculation module, used to calculate the ratio of each historical evaluation data in the N*Y historical evaluation data to the historical evaluation ratio, obtaining N*Y performance deviation data; and an adjustment module, used to adjust the historical allocation weight parameters through each performance deviation data in the N*Y performance deviation data, obtaining allocation weight parameters.

[0103] Optionally, in the customer group determination device for the product provided in this application embodiment, the second acquisition unit 61 includes: a fourth acquisition module, used to acquire learning parameters, calculate the difference between each historical performance deviation data and a preset parameter to obtain N*Y deviation differences, and calculate the product of each deviation value in the N*Y deviation differences and the learning parameter to obtain N*Y deviation products; a fourth calculation module, used to calculate the product of each deviation product and the historical allocation weight parameter to obtain N*Y algorithm weight parameters, wherein the N*Y algorithm weight parameters are used to indicate the algorithm weight parameters of the Y customer screening algorithms associated with each customer group type; and a determination module, used to determine the customer group weight parameters associated with the customer group type of each customer group to be allocated based on the ratio of each algorithm weight parameter to each performance deviation data, to obtain N*Y customer group weight parameters.

[0104] Optionally, in the customer group determination device for the product provided in the embodiments of this application, the first acquisition unit 60 includes: a conversion module, used to convert the format of each group of customer information to obtain M groups of initial processed customer data, wherein the format conversion method refers to converting the unstructured information included in each group of customer information into structured information; and a second extraction module, used to extract features representing customer attributes from the M groups of initial processed customer data to obtain M customer features, and to divide the M customers to be assigned into customer groups based on the M customer features to obtain N customer groups to be assigned.

[0105] Optionally, in the customer group determination device for the product provided in this application embodiment, the calculation unit 62 includes: a fifth acquisition module, used to acquire customer information of K unassigned customers in each group of candidate customers to obtain K groups of customer information, where K is a positive integer; a third extraction module, used to extract customer information representing preset attributes from each group of customer information in the K groups of customer information to obtain K groups of customer attribute information, and to normalize the K groups of customer attribute information to obtain K groups of customer attribute data, wherein each group of customer attribute data includes at least one of the following: indirect attribute data, direct attribute data, and periodic attribute data; a sixth acquisition module, used to acquire attribute weight coefficients, and to use the attribute weight coefficients to perform weighted calculations on the customer group data of each customer attribute data in each group of customer attribute data to obtain K groups of weighted attribute data; and a fifth calculation module, used to calculate the sum of the K groups of weighted attribute data to obtain the total attribute data, and to calculate the ratio of the total attribute data K to obtain the evaluation data corresponding to a group of candidate customers.

[0106] It should be noted that the first acquisition unit 60, the second acquisition unit 61, and the calculation unit 62 mentioned above correspond to steps S201 to S203 in Embodiment 1. The instances and application scenarios implemented by the above units and the corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above units can also be part of a device and can run in the computer terminal 10 provided in Embodiment 1.

[0107] Example 3

[0108] Embodiments of this application may provide a computer terminal, which may be any computer terminal device in a group of computer terminals. Optionally, in this embodiment, the aforementioned computer terminal may also be replaced with a mobile terminal or an electronic device, etc.

[0109] Optionally, in this embodiment, the computer terminal may be located in at least one of a plurality of network devices in a computer network.

[0110] In this embodiment, the computer terminal described above can execute the program code for the following steps in the method for determining the customer group of a product: obtaining customer information for M customers to be assigned, and dividing the M customers to be assigned into customer groups based on the M groups of customer information to obtain N customer groups to be assigned, wherein each customer group to be assigned is associated with a customer group type, and M and N are positive integers; obtaining Y customer screening algorithms, and screening customers from the N customer groups to be assigned using the Y customer screening algorithms to obtain N*Y groups of candidate customers, wherein each group of candidate customers is associated with customer information and the credit data of the product to be assigned, and the credit data of the product to be assigned refers to the number of products allocated to each group of candidate customers; calculating the evaluation data of each group of candidate customers, and sending the N*Y evaluation data and the N*Y groups of candidate customers to the client, wherein, when the client receives the N*Y evaluation data and the N*Y groups of candidate customers, it constructs an operation strategy for the product to be assigned based on the credit data of the product to be assigned associated with the N*Y groups of candidate customers.

[0111] Optionally, the computer terminal described above can execute the following steps in the method for determining the customer groups of the product: obtaining allocation weight parameters; extracting customer group weight parameters associated with the customer group types of N customer groups to be allocated from the allocation weight parameters to obtain N customer group weight parameters; and using each customer group weight parameter to perform weighted calculation on the customer group data of each customer group to be allocated to obtain N first customer groups; extracting algorithm weight parameters of the customer screening algorithm associated with each customer group type from the allocation weight parameters to obtain N sets of algorithm weight parameters, wherein each set of algorithm weight parameters includes Y sub-algorithm weight parameters corresponding to Y customer screening algorithms; using each set of algorithm weight parameters to perform weighted calculation on the customer group data of each first customer group to obtain N second customer groups, wherein each second customer group includes Y second processing sub-customer groups; for a set of second customer groups, using each customer screening algorithm to perform customer screening on each second processing sub-customer group in the second customer group to obtain Y sets of candidate customers associated with the second customer group; and combining the Y sets of candidate customers associated with each set of second customer groups to obtain N*Y sets of candidate customers.

[0112] Optionally, the computer terminal described above can execute the program code for the following steps in the method for determining the customer group of the product: receiving adjustment parameters sent by the client; if the adjustment parameters indicate a first mode, determining the weight parameters sent by the client as the allocation weight parameters; if the adjustment parameters indicate a second mode, obtaining historical allocation weight parameters, and determining the allocation weight parameters based on the historical allocation weight parameters.

[0113] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the method for determining the customer group of the product: obtaining historical candidate customers for a historical time period, obtaining N*Y groups of historical candidate customers, and obtaining historical evaluation data for each group of historical candidate customers, obtaining N*Y historical evaluation data, wherein each group of historical candidate customers refers to a group of historical candidate customers selected by a customer screening algorithm; calculating the sum of the N*Y historical evaluation data, obtaining the total historical evaluation data, and calculating the ratio of the total historical evaluation data to N*Y, obtaining the historical evaluation ratio; calculating the ratio of each historical evaluation data in the N*Y historical evaluation data to the historical evaluation ratio, obtaining N*Y performance deviation data; adjusting the historical allocation weight parameters using each performance deviation data in the N*Y performance deviation data, obtaining the allocation weight parameters.

[0114] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the method for determining the customer group of the product: obtaining learning parameters, calculating the difference between each historical performance deviation data and a preset parameter to obtain N*Y deviation differences, and calculating the product of each deviation value in the N*Y deviation differences with the learning parameter to obtain N*Y deviation products; calculating the product of each deviation product with the historical allocation weight parameter to obtain N*Y algorithm weight parameters, wherein the N*Y algorithm weight parameters are used to indicate the algorithm weight parameters of the Y customer screening algorithms associated with each customer group type; determining the customer group weight parameters associated with the customer group type of each customer group to be allocated based on the ratio of each algorithm weight parameter to each performance deviation data to obtain N*Y customer group weight parameters.

[0115] Optionally, the computer terminal described above can execute the program code for the following steps in the method for determining the customer group of the product: converting the format of each group of customer information to obtain M groups of initial processed customer data, wherein the format conversion method refers to converting the unstructured information included in each group of customer information into structured information; extracting features representing customer attributes from the M groups of initial processed customer data to obtain M customer features; and dividing the M customers to be assigned into customer groups based on the M customer features to obtain N customer groups to be assigned.

[0116] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the method for determining the customer group of the product: For a group of candidate customers, obtain customer information for K unassigned customers in each group of candidate customers to obtain K groups of customer information, where K is a positive integer; extract customer information representing preset attributes from each group of customer information in the K groups of customer information to obtain K groups of customer attribute information, and normalize the K groups of customer attribute information to obtain K groups of customer attribute data, wherein each group of customer attribute data includes at least one of the following: indirect attribute data, direct attribute data, and periodic attribute data; obtain attribute weight coefficients, and use the attribute weight coefficients to perform weighted calculations on the customer group data of each customer attribute data in each group of customer attribute data to obtain K groups of weighted attribute data; calculate the sum of the K groups of weighted attribute data to obtain the total attribute data, calculate the ratio of the total attribute data K to obtain the evaluation data corresponding to a group of candidate customers.

[0117] Optionally, Figure 7 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 7 As shown, the electronic device may include: one or more ( Figure 7 Only one of the following is shown: processor 702, memory 704, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module, and display.

[0118] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the method and apparatus for determining the customer group of the product in this embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the aforementioned method for determining the customer group of the product. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, financial institution intranets, local area networks, mobile communication networks, and combinations thereof.

[0119] The processor can access the information and application programs stored in the memory via a transmission device to execute the steps described above in the method for determining the customer base of the aforementioned product.

[0120] This application provides a scheme for determining the customer group of a product. It involves obtaining customer information for M potential customers and dividing them into N potential customer groups based on the M groups of customer information. Each potential customer group is associated with a customer group type, and M and N are positive integers. Y customer screening algorithms are then obtained, and these algorithms are used to screen customers from the N potential customer groups, resulting in N*Y groups of candidate customers. Each candidate customer group is associated with customer information and the amount of product to be allocated, where the amount of product to be allocated refers to the number of products allocated to each candidate customer group. Evaluation data for each candidate customer group is calculated, and the N*Y evaluation data and N*... Y groups of candidate customers are sent to the client. When the client receives N*Y evaluation data and N*Y groups of candidate customers, it constructs an operational strategy for the products to be allocated based on the credit limit data of the products to be allocated associated with the N*Y groups of candidate customers. This solves the technical problem of low product allocation accuracy when allocating products to different customer groups in related technologies. By dividing the customers to be allocated into customer groups based on customer information, multiple customer groups to be allocated are obtained. Customer screening algorithms are used to screen customers from each customer group to obtain multiple groups of candidate customers, thereby achieving the technical effect of improving the accuracy of product allocation to different customer groups.

[0121] Those skilled in the art will understand that Figure 7 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 7 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 7The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 7 The different configurations shown.

[0122] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0123] Example 4

[0124] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the method for determining the customer group of the product provided in Embodiment 1.

[0125] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0126] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: obtaining customer information for M customers to be assigned, and dividing the M customers to be assigned into customer groups based on the M groups of customer information to obtain N customer groups to be assigned, wherein each customer group to be assigned is associated with a customer group type, and M and N are positive integers; obtaining Y customer screening algorithms, and screening customers from the N customer groups to be assigned using the Y customer screening algorithms to obtain N*Y groups of candidate customers, wherein each group of candidate customers is associated with customer information and credit data of products to be assigned, and the credit data of products to be assigned refers to the number of products allocated to each group of candidate customers; calculating the evaluation data of each group of candidate customers, and sending the N*Y evaluation data and the N*Y groups of candidate customers to the client, wherein, upon receiving the N*Y evaluation data and the N*Y groups of candidate customers, the client constructs an operation strategy for the products to be assigned based on the credit data of the products to be assigned associated with the N*Y groups of candidate customers.

[0127] This application also provides a computer program product, which, when executed on a data processing device, is a program suitable for performing the steps of determining a customer group for the product.

[0128] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0129] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0130] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of units or modules may be electrical or other forms.

[0131] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0132] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0133] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0134] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of determining a customer base of a product, characterized by, The method comprises the following steps: obtaining customer information of M to-be-allocated customers, and performing customer group division on the M to-be-allocated customers according to the M sets of customer information to obtain N to-be-allocated customer groups, wherein each to-be-allocated customer group is associated with a customer group type, and M and N are positive integers; obtaining Y customer screening algorithms, screening customers from the N to-be-allocated customer groups respectively through the Y customer screening algorithms to obtain N*Y sets of candidate customers, wherein each set of candidate customers is associated with customer information and quota data of a to-be-allocated product, and the quota data of the to-be-allocated product refers to the number of products allocated to each set of candidate customers; calculating evaluation data of each set of candidate customers, and sending N*Y evaluation data and the N*Y sets of candidate customers to a client, wherein in a case where the N*Y evaluation data and the N*Y sets of candidate customers are received by the client, an operation strategy of the to-be-allocated product is constructed according to the quota data of the to-be-allocated product associated with the N*Y sets of candidate customers.

2. The method of claim 1, wherein, The method of screening customers from the N to-be-allocated customer groups respectively through the Y customer screening algorithms to obtain N*Y sets of candidate customers comprises the following steps: obtaining an allocation weight parameter, extracting a customer group weight parameter associated with the customer group type of each to-be-allocated customer group from the allocation weight parameter to obtain N customer group weight parameters, and performing weighted calculation on customer group data of each to-be-allocated customer group by using each customer group weight parameter to obtain N first customer groups; extracting an algorithm weight parameter of a customer screening algorithm associated with each customer group type from the allocation weight parameter to obtain N sets of algorithm weight parameters, wherein each set of algorithm weight parameters comprises Y sub-algorithm weight parameters corresponding to the Y customer screening algorithms; performing weighted calculation on customer group data of each first customer group by using each set of algorithm weight parameters to obtain N sets of second customer groups, wherein each set of second customer groups comprises Y second processing sub-customer groups; for a set of second customer groups, performing customer screening on each second processing sub-customer group in the second customer groups by using each customer screening algorithm to obtain Y sets of candidate customers associated with the second customer groups; combining the Y sets of candidate customers associated with each set of second customer groups to obtain the N*Y sets of candidate customers.

3. The method of claim 2, wherein, The method of obtaining an allocation weight parameter comprises the following steps: receiving an adjustment parameter sent by the client, and in a case where the adjustment parameter indicates a first mode, determining the weight parameter sent by the client as the allocation weight parameter; in a case where the adjustment parameter indicates a second mode, obtaining a historical allocation weight parameter, and determining the allocation weight parameter according to the historical allocation weight parameter.

4. The method of claim 3, wherein, The method of determining the allocation weight parameter according to the historical allocation weight parameter comprises the following steps: obtaining historical candidate customers of a historical time period to obtain N*Y sets of historical candidate customers, and obtaining historical evaluation data of each set of historical candidate customers to obtain N*Y historical evaluation data, wherein each set of historical candidate customers refers to a set of historical candidate customers screened by one customer screening algorithm; calculating a sum of the N*Y historical evaluation data to obtain a historical evaluation data sum, and calculating a ratio of the historical evaluation data sum to the N*Y to obtain a historical evaluation ratio; calculate a ratio of each of the N*Y historical evaluation data and the historical evaluation ratio to obtain N*Y performance deviation data; adjust the historical allocation weight parameter through each of the N*Y performance deviation data to obtain the allocation weight parameter.

5. The method of claim 4, wherein, adjust the historical allocation weight parameter through each of the N*Y performance deviation data to obtain the allocation weight parameter includes: obtain a learning parameter, calculate a difference between each historical performance deviation data and a preset parameter to obtain N*Y deviation difference values, and calculate a product of each deviation difference value in the N*Y deviation difference values and the learning parameter respectively to obtain N*Y deviation products; calculate a product of each deviation product and the historical allocation weight parameter respectively to obtain N*Y algorithm weight parameters, wherein the N*Y algorithm weight parameters are used to indicate an algorithm weight parameter of the Y customer screening algorithms associated with each customer group type; determine a customer group weight parameter of the customer group type associated with each to-be-allocated customer group according to a ratio of each algorithm weight parameter and each performance deviation data to obtain N*Y customer group weight parameters.

6. The method of claim 1, wherein, perform customer group division on the M to-be-allocated customers according to M groups of customer information to obtain N to-be-allocated customer groups includes: perform format conversion on each group of customer information to obtain M groups of initial processed customer data, wherein the format conversion manner refers to converting unstructured information included in each group of customer information into structured information; extract features representing customer attributes from M groups of initial processed customer data respectively to obtain M customer features, and perform customer group division on the M to-be-allocated customers based on the M customer features to obtain the N to-be-allocated customer groups.

7. The method of claim 1, wherein, calculating evaluation data of each group of candidate customers includes: for a group of candidate customers, obtain customer information of K to-be-allocated customers in each group of candidate customers to obtain K groups of customer information, wherein K is a positive integer; extract customer information representing a preset attribute from each group of customer information in the K groups of customer information respectively to obtain K groups of customer attribute information, and perform normalization processing on the K groups of customer attribute information to obtain K groups of customer attribute data, wherein each group of customer attribute data at least includes one of the following: indirect attribute data, direct attribute data, and periodic attribute data; obtain an attribute weight coefficient, and perform weighted calculation on customer group data in each customer attribute data in the K groups of customer attribute data by using the attribute weight coefficient to obtain K groups of weighted attribute data; calculate a sum of the K groups of weighted attribute data to obtain an attribute data sum, and calculate a ratio of the attribute data sum to the K to obtain evaluation data corresponding to the group of candidate customers.

8. An apparatus for determining a customer group of a product, characterized by includes: a first obtaining unit, configured to obtain customer information of M to-be-allocated customers, and perform customer group division on the M to-be-allocated customers according to M groups of customer information to obtain N to-be-allocated customer groups, wherein each to-be-allocated customer group is associated with a customer group type, and M and N are positive integers; A second acquisition unit is configured to acquire Y customer screening algorithms, and screen customers from the N groups of to-be-assigned customer groups by using the Y customer screening algorithms to obtain N*Y groups of candidate customers, wherein each group of candidate customers is associated with customer information and quota data of a to-be-assigned product, and the quota data of the to-be-assigned product refers to a number of products assigned to each group of candidate customers; A calculation unit is configured to calculate evaluation data of each group of candidate customers, and send N*Y evaluation data and the N*Y groups of candidate customers to a client, wherein when the N*Y evaluation data and the N*Y groups of candidate customers are received by the client, an operation strategy of the to-be-assigned product is constructed according to the quota data of the to-be-assigned product associated with the N*Y groups of candidate customers.

9. An electronic device, comprising: Comprise: A memory storing an executable program; A processor configured to run the program, wherein the program performs the method for determining a customer group of a product according to any one of claims 1 to 7 when the program is running.

10. A computer program product comprising computer instructions, characterized in that, The computer instructions are executed by the processor to implement the steps of the method for determining a customer group of a product according to any one of claims 1 to 7. The computer instructions are executed by the processor to implement the steps of the method for determining a customer group of a product according to any one of claims 1 to 7.

Images