Commission payment method and device, electronic equipment and storage medium

Abstract

One or more embodiments of the invention provide a commission payment method and device, electronic equipment and a storage medium, and the method comprises the steps: regularly obtaining sales data corresponding to a seller according to a preset commission settlement period; determining a sales scene corresponding to the sales data, and obtaining a commission calculation rule corresponding to the sales scene; wherein the commission calculation rule is a commission calculation rule which is obtained by performing optimization calculation on a plurality of historical commission data samples and a plurality of historical sales data samples generated in the sales scene based on a self-adaptive optimization algorithm and has the highest matching degree with the sales scene; determining commission data corresponding to the seller according to the commission calculation rule; and transferring the commission specified by the commission data from the account of the manager to the account of the seller.

Description

Technical Field

[0001] One or more embodiments of the present application relate to the field of computer application technology, and in particular, to a commission payment method, device, electronic device and storage medium. Background Art

[0002] Commission is a type of compensation paid to sales staff or other related parties based on sales performance or other performance indicators. It is widely used in many industries, especially in sales, real estate, insurance and other fields. This form of compensation links gains with work results and embodies the principle of "more work, more pay". It is usually used as part of the incentive mechanism in business activities, aiming to encourage employees to work more actively through financial rewards, thereby achieving higher performance goals.

[0003] In actual applications, for the managers and sellers of commercial activities, the managers usually hope to pay less commission, while the sellers hope to get more commission. Therefore, it is expected to improve the accuracy and rationality of the commission paid by the managers to the sellers, so that the commission paid can strike a balance in meeting the needs of the managers and sellers. Summary of the invention

[0004] One or more embodiments of the present application provide the following technical solutions:

[0005] The present application provides a commission payment method, the method comprising:

[0006] According to the preset commission settlement cycle, regularly obtain sales data corresponding to the seller;

[0007] Determine a sales scenario corresponding to the sales data, and obtain a commission calculation rule corresponding to the sales scenario; wherein the commission calculation rule is a commission calculation rule with the highest matching degree with the sales scenario obtained by optimizing and calculating a number of historical commission data samples and a number of historical sales data samples generated in the sales scenario based on an adaptive optimization algorithm;

[0008] Determining commission data corresponding to the seller according to the commission calculation rule;

[0009] The commission specified by the commission data is transferred from the account of the management party to the account of the seller.

[0010] The present application also provides a commission payment method, the method comprising:

[0011] According to a preset commission settlement cycle, regularly obtain sales data corresponding to the seller, and obtain historical commission data corresponding to the seller;

[0012] Inputting the historical commission data and the sales data into a trained commission prediction model, and having the commission prediction model predict the commission data corresponding to the seller based on the historical commission data and the sales data;

[0013] The commission specified by the commission data is transferred from the account of the management party to the account of the seller.

[0014] The present application also provides a commission payment device, the device comprising:

[0015] The acquisition module regularly acquires the sales data corresponding to the seller according to the preset commission settlement cycle;

[0016] A matching module determines a sales scenario corresponding to the sales data, and obtains a commission calculation rule corresponding to the sales scenario; wherein the commission calculation rule is a commission calculation rule with the highest matching degree with the sales scenario obtained by optimizing and calculating a number of historical commission data samples and a number of historical sales data samples generated in the sales scenario based on an adaptive optimization algorithm;

[0017] A calculation module, which determines commission data corresponding to the seller according to the commission calculation rule;

[0018] The payment module transfers the commission specified by the commission data from the account of the management party to the account of the seller.

[0019] The present application also provides a commission payment device, the device comprising:

[0020] An acquisition module, which regularly acquires sales data corresponding to the seller according to a preset commission settlement cycle, and acquires historical commission data corresponding to the seller;

[0021] A calculation module, inputting the historical commission data and the sales data into a trained commission prediction model, and having the commission prediction model predict the commission data corresponding to the seller based on the historical commission data and the sales data;

[0022] The payment module transfers the commission specified by the commission data from the account of the management party to the account of the seller.

[0023] The present application also provides an electronic device, including:

[0024] processor;

[0025] a memory for storing processor-executable instructions;

[0026] The processor implements the steps of any of the above methods by running the executable instructions.

[0027] The present application also provides a computer-readable storage medium having computer instructions stored thereon, which implement the steps of any of the above methods when executed by a processor.

[0028] In the above technical solution, when the sales data corresponding to the seller is obtained according to the commission settlement cycle, the sales scenario corresponding to the sales data can be determined first, and then the commission calculation rule corresponding to the sales scenario can be obtained, so as to determine the commission data corresponding to the seller according to the commission calculation rule, so that the commission specified by the commission data can be transferred from the account of the administrator to the account of the seller; wherein the commission calculation rule can be based on an adaptive optimization algorithm, and optimization calculations are performed on several historical commission data samples and several historical sales data samples generated in various sales scenarios to obtain the commission calculation rule with the highest matching degree with the sales scenario.

[0029] By adopting the above method, on the one hand, commission calculation rules corresponding to different sales scenarios are introduced, making the commission calculation mechanism more diversified and personalized, thereby providing more accurate and reasonable commission settlement services; on the other hand, the commission calculation rule corresponding to a sales scenario is the commission calculation rule with the highest matching degree with this sales scenario obtained by adaptive optimization. Therefore, for the commission corresponding to the sales data generated in this sales scenario determined according to the commission calculation rule, its accuracy and rationality can be further guaranteed, and because the adaptive optimization algorithm itself has a high computational efficiency, it can improve the generation and update efficiency of the commission calculation rules, thereby improving the response speed and flexibility of the commission calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The following is a description of the accompanying drawings required for describing the exemplary embodiments, wherein:

[0031] Figure 1 It is a schematic diagram of a commission payment system shown in an exemplary embodiment of the present application.

[0032] Figure 2 It is a flow chart of a commission payment method shown in an exemplary embodiment of the present application.

[0033] Figure 3 It is a schematic diagram of another commission payment system shown in an exemplary embodiment of the present application.

[0034] Figure 4 It is a flow chart of another commission payment method shown in an exemplary embodiment of the present application.

[0035] Figure 5 It is a structural schematic diagram of a device shown in an exemplary embodiment of the present application.

[0036] Figure 6 It is a block diagram of a commission payment device shown in an exemplary embodiment of the present application.

[0037] Figure 7 It is a block diagram of another commission payment device shown in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0038] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. When the following description relates to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of the present application. On the contrary, they are only examples consistent with some aspects of one or more embodiments of the present application.

[0039] It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this application. In some other embodiments, the steps included in the method may be more or less than those described in this application. In addition, a single step described in this application may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this application may also be combined into a single step for description in other embodiments.

[0040] For the managers and sellers of business activities, the managers usually hope to pay less commission, while the sellers hope to get more commission.

[0041] The existing commission calculation system usually adopts a fixed rate or a proportional rate method when calculating commissions. Although such a calculation method is simple and direct, it lacks flexibility and cannot meet the diverse commission calculation needs.

[0042] In practice, a fixed rate means that the same commission fee is charged regardless of the transaction amount. This approach is usually not suitable for all scenarios. Specifically, it is unfair for small transactions because the commission ratio for small transactions is too high compared to the larger transaction amount; and for large transactions, the fixed commission may not be enough to compensate the service cost provided by the seller.

[0043] A proportional fee refers to charging a commission based on a percentage of the transaction amount. This seems fair, but a uniform ratio may not be suitable for all scenarios for different products, services or market conditions. For example, the sale of certain high-value goods may require higher service input, while low-value goods do not require so many resources, but a proportional fee does not reflect this difference well.

[0044] Since the above two methods are based on simple and direct rules, they cannot adapt to more complex sales models, such as promotions, membership levels, pricing strategies for specific markets, etc. In addition, in a changing market environment, management may want to dynamically adjust the commission rate according to different time periods, customer behavior or other factors to optimize costs and benefits, but the existing commission calculation system is difficult to respond to these changes quickly.

[0045] In order to better adapt to various scenarios and improve competitiveness, some managers may introduce tiered commission structures, performance-linked commission systems, or customized commission plans for specific partners. However, existing commission calculation systems are usually not flexible enough to support these new commission calculation requirements.

[0046] Therefore, it is expected that a more diversified and personalized commission calculation mechanism can be introduced so as to provide more accurate and reasonable commission settlement services.

[0047] One or more embodiments of the present application provide a technical solution for commission payment to improve the accuracy and rationality of the commission paid by the management party to the seller. The present application specifically provides two commission calculation methods, one of which is a commission calculation method based on adaptive optimization and scenario matching, and the other is a commission calculation method based on machine learning.

[0048] In a technical solution for calculating commissions based on adaptive optimization and scenario matching, when sales data corresponding to a seller is obtained according to a commission settlement period, the sales scenario corresponding to the sales data can be determined first, and then the commission calculation rule corresponding to the sales scenario can be obtained, so as to determine the commission data corresponding to the seller according to the commission calculation rule, thereby transferring the commission specified by the commission data from the account of the administrator to the account of the seller; wherein the commission calculation rule can be based on an adaptive optimization algorithm, and optimization calculations are performed on several historical commission data samples and several historical sales data samples generated in various sales scenarios to obtain the commission calculation rule with the highest matching degree with the sales scenario.

[0049] By adopting the above method, on the one hand, commission calculation rules corresponding to different sales scenarios are introduced, making the commission calculation mechanism more diversified and personalized, thereby providing more accurate and reasonable commission settlement services; on the other hand, the commission calculation rule corresponding to a sales scenario is the commission calculation rule with the highest matching degree with this sales scenario obtained by adaptive optimization. Therefore, for the commission corresponding to the sales data generated in this sales scenario determined according to the commission calculation rule, its accuracy and rationality can be further guaranteed, and because the adaptive optimization algorithm itself has a high computational efficiency, it can improve the generation and update efficiency of the commission calculation rules, thereby improving the response speed and flexibility of the commission calculation.

[0050] In a technical solution for calculating commissions based on machine learning, when the sales data corresponding to the seller is obtained according to the commission settlement cycle and the historical commission data corresponding to the seller is obtained, the historical commission data and the sales data can be input into a trained commission prediction model, and the commission prediction model predicts the commission data corresponding to the seller based on the historical commission data and the sales data, so that the commission specified by the commission data can be transferred from the manager's account to the seller's account.

[0051] By adopting the above method, a commission prediction model is constructed through machine learning, and the commission prediction model is used to determine the commission corresponding to the sales data. This can ensure the accuracy and rationality of the determined commission, thereby providing a more accurate and reasonable commission settlement service.

[0052] The two commission calculation methods mentioned above and the corresponding commission payment process are explained below.

[0053] 1. Commission calculation method based on adaptive optimization and scenario matching and its corresponding commission payment process

[0054] Please refer to Figure 1 , Figure 1 It is a schematic diagram of a commission payment system shown in an exemplary embodiment of the present application.

[0055] like Figure 1As shown, the above-mentioned commission payment system may include a user interface facing the management party, a commission calculation component and a commission payment component. Among them, the commission calculation component and the commission payment component can be deployed on the server, and the user interface can be provided to the management party through the client. The server can correspond to a server including an independent physical host, or a server cluster composed of multiple independent physical hosts; or, it can correspond to a virtual server, cloud server, etc. carried by a host cluster. The client can correspond to terminal devices such as smart phones, tablet computers, laptops, desktop computers, etc. The client can access the server through any type of wired or wireless network.

[0056] The administrator can use the commission calculation and commission payment services provided by the above-mentioned commission payment system through the above-mentioned user interface. Specifically, the administrator can execute the process of initiating commission calculation and commission payment in the user interface, input sales data corresponding to the seller, and input other influencing factors (such as promotion intensity, customer interaction intensity, market trends, etc.) that assist in the decision-making of the commission paid to the seller, so as to provide these data to the above-mentioned commission calculation component, triggering the commission calculation component to calculate the commission based on these sales data.

[0057] The above-mentioned commission calculation component can obtain the sales data for commission calculation input by the management party, and calculate the commission data based on the sales data (wherein the commission data refers to the proportion or specific amount of the commission). Alternatively, the above-mentioned commission calculation component can also collect the sales data for commission calculation by itself, and calculate the commission data based on the sales data. For example, the commission calculation component can obtain the sales data for commission calculation generated within a commission settlement cycle, and based on the sales data, calculate the commission data for specifying the commission that the management party pays to the seller after the end of this commission settlement cycle.

[0058] The above-mentioned commission payment component can complete the commission payment from the administrator to the seller through the accounts registered by the administrator and the seller in the commission payment system. Specifically, the commission payment system can transfer the corresponding commission from the administrator's account to the seller's account based on the commission data calculated by the above-mentioned commission calculation component. For example, assuming that the commission data is the proportion of the commission, the specific amount of the commission can be calculated based on the proportion and the sales amount, and the commission of this amount can be transferred from the administrator's account to the seller's account; or, assuming that the commission data is the specific amount of the commission, the commission of this amount can be directly transferred from the administrator's account to the seller's account. Among them, the form of the commission can be physical resources such as currency, stocks, bonds, bills, accounts receivable, etc. in the real world, or it can be a virtual resource with a specific value.

[0059] Furthermore, the above-mentioned commission calculation component may include a data collection module, a contract management module, a data analysis module, a rule management module, a feedback module, a rule matching module, a commission calculation module, etc.

[0060] The data collection module can be used to obtain sales data corresponding to the seller generated in a commission settlement cycle. For example, at the end of a commission settlement cycle, the management party can obtain the sales data counted in this commission settlement cycle input through the user interface, or the management party can collect the sales data generated in this commission settlement cycle by itself.

[0061] The above-mentioned contract management module can be used to manage relevant information of the contract signed by the management party and the seller, such as information of the account used by the management party to pay commissions and information of the account used by the seller to receive commissions, so as to realize transparency and automation of the commission payment process.

[0062] The above-mentioned data analysis module can be used to optimize the calculation based on an adaptive optimization algorithm, with the optimization goal of maximizing the matching degree of commission data relative to the sales data generated in the sales scenario, for several historical commission data samples and several historical sales data samples generated in a sales scenario, to obtain the commission calculation rule with the highest matching degree with this sales scenario.

[0063] The above-mentioned rule management module can be used to manage the correspondence between sales scenarios and the commission calculation rules with the highest matching degree, and the commission calculation rules can be updated regularly. For example, several historical commission data samples and several historical sales data samples that are generated in a sales scenario close to the current time can be regularly obtained, and based on the above-mentioned adaptive optimization algorithm, with maximizing the matching degree of the commission data relative to the sales data generated in the sales scenario as the optimization goal, iterative optimization calculations are performed on the historical commission data samples and the historical sales data samples to obtain the commission calculation rules with the highest matching degree with this sales scenario. In this way, the matching degree between the commission calculation rules and the latest situation of the sales scenario can be guaranteed, thereby further improving the accuracy and rationality of the commissions paid by the management direction to the seller.

[0064] The above feedback module can be used to adjust the commission calculation rules according to the feedback from the management party and the seller. Specifically, when the commission data is determined according to the sales data and the commission calculation rules, the satisfaction of the management party and the seller with the commission data can be obtained in real time, and the commission data and the sales data can be used as new samples. Based on the above adaptive optimization algorithm, with maximizing the matching degree of the commission data relative to the sales data generated in the sales scenario as the optimization goal, iterative optimization calculations are performed on the new samples to achieve the adjustment of the commission calculation rules. In this way, it can be ensured that the commission calculation rules always meet the actual needs of the management party and the seller.

[0065] The rule matching module can be used to match commission calculation rules, that is, to determine the sales scenario corresponding to the sales data and obtain the commission calculation rule corresponding to the sales scenario.

[0066] The commission calculation module may be used to calculate the commission data corresponding to the seller based on the matched commission calculation rules and the sales data corresponding to the seller.

[0067] Please refer to Figure 2 , Figure 2 It is a flow chart of a commission payment method shown in an exemplary embodiment of the present application.

[0068] In this embodiment, the above commission payment method can be applied to Figure 1 The server shown in Figure 1 is as follows. Figure 2 As shown, the commission payment method may include the following steps:

[0069] Step 202: According to the preset commission settlement cycle, regularly obtain sales data corresponding to the seller.

[0070] In this embodiment, the sales data corresponding to the seller can be obtained regularly according to the preset commission settlement cycle, so that the proportion or specific amount of the commission corresponding to the seller can be calculated based on the sales data, that is, the proportion or specific amount of the commission paid by the management party to the seller. For example, at the end of a commission settlement cycle, the sales data counted in this commission settlement cycle input by the management party through the user interface can be obtained, or the sales data generated in this commission settlement cycle can be collected by the management party itself.

[0071] It should be noted that the above commission settlement cycle can be set and adjusted according to the actual situation, the needs of the management party and the seller; that is, the commission settlement cycle can be fixed or change according to the actual situation and needs. For example, the commission settlement cycle can be shortened to one month after entering the peak sales period, and extended to one quarter after entering the trough sales period, and so on.

[0072] In some embodiments, the sales data may include one or more of the following: sales amount; market activity data; competitor data; seasonal data. Among them, the market activity data may further include numerical data such as promotion intensity and customer activity intensity, the competitor data may further include numerical data such as competitor sales amount and competitor promotion intensity, and the seasonal data may further include statistical data such as customer active time distribution and purchase intention time distribution.

[0073] Step 204: Determine the sales scenario corresponding to the sales data, and obtain the commission calculation rule corresponding to the sales scenario; wherein the commission calculation rule is based on an adaptive optimization algorithm, and optimizes and calculates a number of historical commission data samples and a number of historical sales data samples generated in the sales scenario to obtain the commission calculation rule with the highest degree of match with the sales scenario.

[0074] In this embodiment, in order to introduce a more diversified and personalized commission calculation mechanism, so as to improve the accuracy and rationality of the commission paid by the management to the seller, different sales scenarios can be divided, and different commission calculation rules can be set for different sales scenarios. For example, the spring sales scenario, summer sales scenario, autumn sales scenario and winter sales scenario can be divided according to the season, where the commission calculation rule of the winter sales scenario is a high incentive mode, that is, when the sales data is roughly the same, the commission ratio or amount calculated according to the commission calculation rules corresponding to these four sales scenarios is the largest. For another example, high-input sales scenarios and low-input sales scenarios can be divided according to the promotion intensity and customer interaction intensity, where the commission calculation rule of the high-input sales scenario is a high incentive mode, which is intended to encourage the seller to hold more promotional activities and strengthen interaction with customers.

[0075] When the above sales data is obtained, the sales scenario corresponding to the sales data may be determined first, and then the commission calculation rule corresponding to the sales scenario may be obtained.

[0076] Specifically, on the one hand, the sales scenario corresponding to the sales data means that the sales data is generated in the sales scenario, so the sales scenario corresponding to the sales data can be determined based on the specific content of the sales data. For example, assuming that the sales data includes the transaction date of each transaction that occurred in the current commission settlement cycle, and these transaction dates are all dates within December, the winter sales scenario can be determined as the sales scenario corresponding to the sales data. For another example, assuming that the sales data includes the number of promotional activities that occurred in the current commission settlement cycle, and the number of promotional activities falls into the high-intensity promotional activity frequency range, the high-investment sales scenario can be determined as the sales scenario corresponding to the sales data.

[0077] In actual applications, when determining the sales scenarios corresponding to the sales data, different priorities can be set for different types of sales scenarios. When multiple corresponding sales scenarios can be determined based on the specific content of the sales data, the sales scenario with the highest priority among these sales scenarios can be determined as the sales scenario corresponding to the sales data. For example, a higher priority can be set for sales scenarios divided by promotion intensity and customer interaction intensity, while a lower priority can be set for sales scenarios divided by season. Assuming that the sales data includes both the transaction date of each transaction occurring in the current commission settlement cycle, and these transaction dates are all dates within December, and the number of promotional activities occurring in the current commission settlement cycle, and the number of promotional activities falls within the high-intensity promotional activity number range, then according to the priority level, the high-investment sales scenario can be determined as the sales scenario corresponding to the sales data.

[0078] On the other hand, the correspondence between the sales scenario and the commission calculation rule can be maintained, and the correspondence includes each sales scenario and its corresponding commission calculation rule. In this case, after determining the sales scenario corresponding to the sales data, the commission calculation rule corresponding to the sales scenario can be determined by querying the correspondence.

[0079] For a sales scenario, the commission calculation rule corresponding to this sales scenario can be based on an adaptive optimization algorithm, and optimized and calculated for several historical commission data samples (where commission data refers to the proportion or specific amount of commission) and several historical sales data samples generated in this sales scenario, to obtain the commission calculation rule with the highest matching degree with this sales scenario. The matching degree of the commission calculation rule with this sales scenario can be converted into the matching degree of the commission data in the commission calculation rule relative to the sales data generated in this sales scenario.

[0080] Adaptive optimization algorithms are a type of algorithm that can dynamically adjust its own parameters or strategies during the optimization process to improve convergence speed, improve solution quality, or better cope with non-stationary environments. Such algorithms are usually used to solve complex optimization problems, especially those with unknown, discontinuous, multi-modal or multi-objective objective functions. Common adaptive optimization algorithms include adaptive gradient algorithms (Adaptive Gradient Algorithm), evolutionary algorithms (Evolutionary Algorithms, EAs), adaptive random search (Adaptive Random Search, ARS), Bayesian Optimization (Bayesian Optimization, BO), etc., among which evolutionary algorithms further include genetic algorithms (Genetic Algorithm, GA), differential evolution (Differential Evolution, DE), and particle swarm optimization (Particle Swarm Optimization, PSO).

[0081] When optimizing and calculating the historical commission data samples and the historical sales data samples based on the adaptive optimization algorithm to obtain the commission calculation rules, in some embodiments, for a sales scenario, several historical commission data samples and several historical sales data samples generated in the sales scenario may be first obtained. The historical commission data samples may be a batch of commission data selected from the previously generated commission data, and the historical sales data samples may be a batch of sales data selected from the previously generated sales data.

[0082] It should be noted that there is a corresponding relationship between the above historical commission data sample and the above historical sales data sample. A historical commission data sample corresponds to a historical sales data sample, which means that the commission data obtained by calculating the commission based on the historical sales data sample is the historical commission data sample.

[0083] Subsequently, based on the adaptive optimization algorithm, the optimization goal is to maximize the matching degree of the commission data relative to the sales data generated in the sales scenario, and perform iterative optimization calculations on the above historical commission data samples and the above historical sales data samples, so that the commission calculation rules with the highest matching degree with the above sales scenario can be obtained. Among them, the matching degree of the commission data relative to the sales data generated in the sales scenario can be an indicator for quantifying the satisfaction of the management party and the satisfaction of the seller. In practical applications, the seller's satisfaction can specifically be the seller's satisfaction with the commission data after the sales data is generated in the sales scenario, and similarly, the management party's satisfaction can specifically be the management party's satisfaction with the commission data. For example, the seller's satisfaction can be the seller's satisfaction with the management party paying a commission of 5,000 yuan for the sales amount of 100,000 yuan after the sales amount of 100,000 yuan is generated in the sales scenario. Similarly, the management party's satisfaction can be the management party's satisfaction with paying a commission of 5,000 yuan for the sales amount of 100,000 yuan.

[0084] Since the management party hopes to pay less commission to control costs in most cases, while the seller hopes to get more commission, the ratio of commission amount to sales amount can also be used to replace the quantitative value of management party satisfaction or seller party satisfaction, where the ratio of commission amount to sales amount is inversely proportional to the quantitative value of management party satisfaction and directly proportional to the quantitative value of seller party satisfaction.

[0085] It should be noted that the calculation method of the matching degree of the commission data relative to the sales data generated in the sales scenario can be set according to the actual situation and needs, and this application does not impose any special restrictions on this. For example, an indicator for quantifying market reaction can also be used as part of the matching degree; market reaction refers to the willingness of each seller to continue selling after the sales data is generated in the sales scenario and the commission indicated by the commission data is paid accordingly.

[0086] In order to further improve the accuracy and rationality of the commission paid by the management to the seller, in some embodiments, for any sales scenario, the specific content of the sales data generated in the sales scenario can be divided into different intervals, and different commission data can be set for different intervals to achieve tiered commissions. Specifically, various types of data in the sales data can be converted into sales indicators, and these sales indicators can include multiple sales indicator intervals. For example, the sales indicator may be the sales amount (unit: 10,000 yuan), and this sales indicator may include four sales indicator intervals of 0-10, 10-50, 50-100, and above 100; the sales indicator may also be an indicator used to indicate the season, and the value of this sales indicator is 1 for spring, 2 for summer, 3 for autumn, and 4 for winter. In this case, this indicator may include two sales indicator intervals of [1,2] and [3,4]; or, the sales indicator may include the sales amount (unit: 10,000 yuan) and the promotion intensity (unit: level). These two sales indicators may include 16 sales indicator intervals in the form of combinations of (0-10, 0-2), (10-50, 0-2), (50-100, 0-2), (above 100, 0-2), (0-10, 2-4), (10-50, 2-4), (50-100, 2-4), and (above 100, 2-4).

[0087] In the above case, the commission calculation rules corresponding to the sales scenario may include the corresponding relationship between each sales index interval in the multiple sales index intervals and the commission data. For example, the commission calculation rules may be as shown in Table 1 below:

[0088] Table 1

[0089]

[0090] In this commission calculation rule, if the sales amount is 80,000 yuan and the promotion intensity is level 1, the commission amount can be determined to be 5,000 yuan.

[0091] It should be noted that when the sales indicators generated in a sales scenario are divided into multiple sales indicator intervals, the matching degree between the commission calculation rule and the sales scenario can be converted into the matching degree of the commission data in the commission calculation rule relative to the corresponding sales indicator interval. For example, assuming that the sales indicators generated in the sales scenario are divided into sales indicator interval 1, sales indicator interval 2 and sales indicator interval 3, in the commission calculation rule, sales indicator interval 1 corresponds to commission data 1, sales indicator interval 2 corresponds to commission data 2, and sales indicator interval 3 corresponds to commission data 3, then the following formula can be obtained: the matching degree of the commission calculation rule and the sales scenario = (the matching degree of commission data 1 relative to sales indicator interval 1 + the matching degree of commission data 2 relative to sales indicator interval 2 + the matching degree of commission data 3 relative to sales indicator interval 3) / 3. As mentioned above, the matching degree of commission data relative to the sales indicator interval can be an indicator for quantifying the satisfaction of the management party and the satisfaction of the seller, wherein the satisfaction of the management party can specifically be the satisfaction of the management party with the commission data after the sales indicator belonging to the sales indicator interval is generated, and similarly, the satisfaction of the seller can specifically be the satisfaction of the seller with the commission data.

[0092] In order to adjust the commission ladder in a timely manner to ensure that it always meets the actual needs of the management party and the seller, in some embodiments, a feedback mechanism can be established to introduce feedback from the management party and the seller to adjust the commission calculation rules. Specifically, when the commission data is determined based on the sales data and the commission calculation rules, the satisfaction of the management party and the seller with the commission data can be obtained in real time, and the commission data and the sales data can be used as new samples. Based on the above-mentioned adaptive optimization algorithm, with maximizing the matching degree of the commission data relative to the sales data generated in the sales scenario as the optimization goal, iterative optimization calculations are performed on the new samples to achieve the adjustment of the commission calculation rules.

[0093] In order to further improve the accuracy and rationality of the commission paid by the management to the seller, in some embodiments, for a sales scenario, it is also possible to regularly obtain several historical commission data samples and several historical sales data samples generated in this sales scenario according to the preset rule update cycle, and based on the above-mentioned adaptive optimization algorithm, with the optimization goal of maximizing the matching degree of the commission data relative to the sales data generated in the sales scenario, iterative optimization calculations are performed on the historical commission data samples and the historical sales data samples to obtain the commission calculation rules with the highest matching degree with the sales scenario. Among them, the historical commission data samples and the historical sales data samples can be historical commission data samples and historical sales data samples whose generation time is close to the current time, and the commission calculation rules obtained at this time can be used as the updated commission calculation rules corresponding to the sales scenario. In this way, the matching degree of the commission calculation rules with the latest situation of the sales scenario can be guaranteed, so that the accuracy and rationality of the commission paid by the management to the seller can be further improved.

[0094] As for the above-mentioned adaptive optimization algorithm adopted, in some embodiments, the adaptive optimization algorithm may specifically be a genetic algorithm.

[0095] The working principle of genetic algorithms is to imitate the evolutionary process that occurs in natural selection and genetics. In genetic algorithms, population refers to a set of candidate solutions that may solve a problem, and each candidate solution is called an individual. Chromosome represents the encoding form of a solution in genetic algorithms, usually composed of a series of parameters. Gene refers to a site on a chromosome that represents a specific parameter. Fitness function is used to evaluate the quality of a solution. The higher the fitness value, the better the quality of the solution. Selection refers to deciding which individuals can serve as parents for the next generation based on their fitness. Crossover refers to the generation of new offspring by exchanging part of the information of two parent individuals. Mutation refers to the random change of certain genes in certain individuals to maintain population diversity and prevent premature convergence. Replacement refers to replacing old individuals with newly generated offspring to form a new population.

[0096] The genetic algorithm can specifically include the following steps: initialize the population, that is, randomly generate a set of initial solutions; evaluate the fitness, that is, use the fitness function to calculate the fitness of each individual in the population; perform the selection operation, that is, select excellent individuals as parents based on fitness; perform the crossover operation, that is, combine the selected parents to produce new offspring; perform the mutation operation, that is, randomly change certain genes of the offspring with a certain probability, and perform the replacement operation, that is, replace the old individuals with new offspring to form a new population; if the preset termination condition (such as the number of iterations, fitness threshold, etc.) is reached, the iteration is stopped, otherwise the fitness is re-evaluated to continue the iteration.

[0097] Based on the genetic algorithm, with the optimization goal of maximizing the matching degree of the commission data relative to the sales data generated in the sales scenario, the above historical commission data samples and the above historical sales data samples are iteratively optimized and calculated to obtain the commission calculation rule with the highest matching degree with the above sales scenario, the individuals in the population can be commission data, the fitness can be the matching degree score of the commission data relative to the sales data generated in the sales scenario, the crossover operation can be the calculation of the average value, and the mutation operation can be the addition or subtraction of a random value that does not exceed a preset threshold. Specifically, the following steps can be iteratively performed until the commission data with the highest matching degree relative to the sales scenario is generated, and the commission calculation rule with the highest matching degree with the sales scenario is generated based on the commission data: calculate the matching degree score of the commission data relative to the historical sales data sample (the historical sales data sample is essentially the sales data generated in the sales scenario); select the preset number of commission data with the highest matching degree; perform combination processing (i.e., crossover operation, mutation operation, replacement operation) on the selected commission data to generate new commission data (i.e., new population). Among them, in order to simplify the calculation process of the above genetic algorithm, the above historical commission data sample can be directly used as the initial commission data (i.e., initial population).

[0098] As mentioned above, in some embodiments, the matching degree of the commission data with respect to the above-mentioned historical sales data sample generated in the above-mentioned sales scenario may specifically include: a weighted sum of a first score obtained by quantifying the manager's satisfaction and a second score obtained by quantifying the seller's satisfaction. The manager's satisfaction may specifically be the manager's satisfaction with the commission data when the historical sales data sample is generated in the sales scenario; the seller's satisfaction may specifically be the seller's satisfaction with the commission data when the historical sales data sample is generated in the sales scenario.

[0099] Step 206: Determine the commission data corresponding to the seller according to the commission calculation rule.

[0100] In this embodiment, when the above-mentioned commission calculation rules corresponding to the above-mentioned sales scenario are obtained, the commission data corresponding to the above-mentioned seller can be determined according to the commission calculation rules. Specifically, when the commission calculation rules contain commission data corresponding to the sales scenario, the commission data can be directly determined as the commission data corresponding to the seller. When the sales index generated in the sales scenario is divided into multiple sales index intervals, since the commission calculation rules at this time contain each sales index interval and its corresponding commission data, the corresponding sales index interval can be determined based on the above-mentioned sales data, and then the commission data corresponding to the sales index interval can be determined as the commission data corresponding to the seller.

[0101] Step 208: Transfer the commission indicated by the commission data from the manager's account to the seller's account.

[0102] In this embodiment, when the commission data corresponding to the above-mentioned seller is determined, the commission indicated by the commission data can be transferred from the manager's account to the seller's account to complete the commission payment of the current commission settlement cycle. For example, assuming that the commission data is the commission ratio, the specific amount of the commission can be calculated based on the ratio and the sales amount in the above-mentioned sales data, and the commission of this amount can be transferred from the manager's account to the seller's account; or, assuming that the commission data is the specific amount of the commission, the commission of this amount can be directly transferred from the manager's account to the seller's account. Among them, the form of the commission can be physical resources such as currency, stocks, bonds, bills, accounts receivable, etc. in the real world, or it can be a virtual resource with a specific value.

[0103] In the above technical solution, when the sales data corresponding to the seller is obtained according to the commission settlement cycle, the sales scenario corresponding to the sales data can be determined first, and then the commission calculation rule corresponding to the sales scenario can be obtained, so as to determine the commission data corresponding to the seller according to the commission calculation rule, so that the commission specified by the commission data can be transferred from the account of the administrator to the account of the seller; wherein the commission calculation rule can be based on an adaptive optimization algorithm, and optimization calculations are performed on several historical commission data samples and several historical sales data samples generated in various sales scenarios to obtain the commission calculation rule with the highest matching degree with the sales scenario.

[0104] By adopting the above method, on the one hand, commission calculation rules corresponding to different sales scenarios are introduced, making the commission calculation mechanism more diversified and personalized, thereby providing more accurate and reasonable commission settlement services; on the other hand, the commission calculation rule corresponding to a sales scenario is the commission calculation rule with the highest matching degree with this sales scenario obtained by adaptive optimization. Therefore, for the commission corresponding to the sales data generated in this sales scenario determined according to the commission calculation rule, its accuracy and rationality can be further guaranteed, and because the adaptive optimization algorithm itself has a high computational efficiency, it can improve the generation and update efficiency of the commission calculation rules, thereby improving the response speed and flexibility of the commission calculation.

[0105] 2. Commission calculation method based on machine learning and its corresponding commission payment process

[0106] Please refer to Figure 3 , Figure 3 It is a schematic diagram of another commission payment system shown in an exemplary embodiment of the present application.

[0107] like Figure 3 As shown, the above-mentioned commission payment system may include a user interface facing the management party, a commission calculation component and a commission payment component. Among them, the commission calculation component and the commission payment component can be deployed on the server, and the user interface can be provided to the management party through the client. The server can correspond to a server including an independent physical host, or a server cluster composed of multiple independent physical hosts; or, it can correspond to a virtual server, cloud server, etc. carried by a host cluster. The client can correspond to terminal devices such as smart phones, tablet computers, laptops, desktop computers, etc. The client can access the server through any type of wired or wireless network.

[0108] The administrator can use the commission calculation and commission payment services provided by the above-mentioned commission payment system through the above-mentioned user interface. Specifically, the administrator can execute the process of initiating commission calculation and commission payment in the user interface, input sales data corresponding to the seller, and input other influencing factors (such as promotion intensity, customer interaction intensity, market trends, etc.) that assist in the decision-making of the commission paid to the seller, so as to provide these data to the above-mentioned commission calculation component, triggering the commission calculation component to calculate the commission based on these sales data.

[0109] The above-mentioned commission calculation component can obtain the sales data for commission calculation input by the management party, and calculate the commission data based on the sales data (wherein the commission data refers to the proportion or specific amount of the commission). Alternatively, the above-mentioned commission calculation component can also collect the sales data for commission calculation by itself, and calculate the commission data based on the sales data. For example, the commission calculation component can obtain the sales data for commission calculation generated within a commission settlement cycle, and based on the sales data, calculate the commission data for specifying the commission that the management party pays to the seller after the end of this commission settlement cycle.

[0110] The above-mentioned commission payment component can complete the commission payment from the administrator to the seller through the accounts registered by the administrator and the seller in the commission payment system. Specifically, the commission payment system can transfer the corresponding commission from the administrator's account to the seller's account based on the commission data calculated by the above-mentioned commission calculation component. For example, assuming that the commission data is the proportion of the commission, the specific amount of the commission can be calculated based on the proportion and the sales amount, and the commission of this amount can be transferred from the administrator's account to the seller's account; or, assuming that the commission data is the specific amount of the commission, the commission of this amount can be directly transferred from the administrator's account to the seller's account. Among them, the form of the commission can be physical resources such as currency, stocks, bonds, bills, accounts receivable, etc. in the real world, or it can be a virtual resource with a specific value.

[0111] Furthermore, the above-mentioned commission calculation component may include a data collection module, a contract management module, a model building module, a model management module, a feedback module, a commission calculation module, etc.

[0112] The data collection module can be used to obtain sales data corresponding to the seller generated in a commission settlement cycle. For example, at the end of a commission settlement cycle, the management party can obtain the sales data counted in this commission settlement cycle input through the user interface, or the management party can collect the sales data generated in this commission settlement cycle by itself. In addition, the data collection module can also be used to obtain historical commission data corresponding to the seller in previous commission settlement cycles.

[0113] The above-mentioned contract management module can be used to manage relevant information of the contract signed by the management party and the seller, such as information of the account used by the management party to pay commissions and information of the account used by the seller to receive commissions, so as to realize transparency and automation of the commission payment process.

[0114] The above-mentioned model building module can be used for the constructed commission prediction model, with the optimization goal of minimizing the difference between the commission data predicted by the commission prediction model based on historical commission data and sales data and the historical commission data. Several historical commission data samples and several historical sales data samples generated in a sales scenario are input into the commission prediction model for training to obtain the trained commission prediction model.

[0115] The above-mentioned model management module can be used to manage the commission prediction model that has been trained, and can regularly update the commission prediction model. For example, an indicator indicating the model performance of the commission prediction model can be regularly calculated, and it can be determined based on the indicator whether the commission prediction model needs to be updated. If so, the model parameters or input features of the commission prediction model can be adjusted to achieve the update of the commission prediction model.

[0116] The above-mentioned commission calculation module can be used to use the trained commission prediction model, input the historical commission data and sales data corresponding to the seller into the commission prediction model for calculation, and the commission prediction model predicts the commission data corresponding to the seller.

[0117] Please refer to 4, Figure 4 It is a flow chart of another commission payment method shown in an exemplary embodiment of the present application.

[0118] In this embodiment, the above commission payment method can be applied to Figure 3 The server shown in Figure 1 is as follows. Figure 4 As shown, the commission payment method may include the following steps:

[0119] Step 402: According to a preset commission settlement cycle, regularly obtain sales data corresponding to the seller, and obtain historical commission data corresponding to the seller.

[0120] In this embodiment, the sales data corresponding to the seller can be obtained regularly according to the preset commission settlement cycle, and the historical commission data corresponding to the seller (wherein the commission data refers to the proportion or specific amount of the commission) can be obtained, so that the commission data corresponding to the seller can be calculated based on the sales data and the historical commission data, and the proportion or specific amount of the commission paid by the management party to the seller can be determined through the commission data. For example, at the end of a commission settlement cycle, the sales data counted in this commission settlement cycle input by the management party through the user interface can be obtained, or the sales data generated in this commission settlement cycle can be collected by the management party itself; and the historical commission data of several commission settlement cycles before this commission settlement cycle can be obtained.

[0121] It should be noted that the above commission settlement cycle can be set and adjusted according to the actual situation, the needs of the management party and the seller; that is, the commission settlement cycle can be fixed or change according to the actual situation and needs. For example, the commission settlement cycle can be shortened to one month after entering the peak sales period, and extended to one quarter after entering the trough sales period, and so on.

[0122] In some embodiments, the sales data may include one or more of the following: sales amount; market activity data; competitor data; seasonal data. Among them, the market activity data may further include numerical data such as promotion intensity and customer activity intensity, the competitor data may further include numerical data such as competitor sales amount and competitor promotion intensity, and the seasonal data may further include statistical data such as customer active time distribution and purchase intention time distribution.

[0123] In order to ensure the integrity and accuracy of the data and to guarantee the quality of the model input, in some embodiments, the historical commission data and the sales data may be cleaned first, that is, the missing values ​​and outliers in these data may be processed and the data may be standardized.

[0124] Step 404: Input the historical commission data and the sales data into a trained commission prediction model, and the commission prediction model predicts the commission data corresponding to the seller based on the historical commission data and the sales data.

[0125] In order to improve the accuracy and rationality of the commission paid by the management to the seller, a commission prediction model can be constructed by machine learning to use the commission prediction model to predict commission data. Specifically, in some embodiments, a number of historical commission data samples and a number of historical sales data samples can be obtained, and the degree of difference between the commission data predicted by the commission prediction model based on the historical commission data and sales data and the historical commission data (i.e., the historical commission data used for prediction) is minimized as the optimization goal, and the historical commission data samples and the historical sales data samples are input into the constructed commission prediction model for training to obtain the trained commission prediction model.

[0126] It should be noted that the degree of difference between the commission data predicted by the commission prediction model based on the historical commission data and sales data and the historical commission data can be measured using the residual sum of squares.

[0127] For a commission settlement cycle, the historical commission data of several commission settlement cycles before this commission settlement cycle, as well as the sales data generated in this commission settlement cycle, may have an impact on the commission data of this commission settlement cycle alone. Therefore, in order to further improve the accuracy and rationality of the commission paid by the management to the seller, in some embodiments, the above-mentioned commission prediction model may include an autoregressive sub-model and a linear regression sub-model, wherein the autoregressive sub-model can be used to predict new commission data based on historical commission data, and the linear regression sub-model can be used to predict commission data based on sales data. In this case, the output of the commission prediction model can be the sum of the output of the autoregressive sub-model and the output of the linear regression sub-model.

[0128] When training the above commission prediction model, the above historical commission data samples can be input into the above autoregressive sub-model for calculation to obtain a portion of the commission data predicted by the autoregressive sub-model based on the historical commission data samples (which can be referred to as the first commission data); and the above historical sales data samples can be input into the above linear regression sub-model for calculation to obtain a portion of the commission data predicted by the linear regression sub-model based on the historical sales data samples (which can be referred to as the second commission data). Furthermore, the sum of the first commission data and the second commission data can be determined as the commission data predicted by the commission prediction model based on the historical commission data samples and the historical sales data samples, and the degree of difference between the commission data predicted by the commission prediction model and the historical commission data samples can be calculated, so as to adjust the model parameters of the commission prediction model according to the degree of difference and complete the training of the commission prediction model.

[0129] In some embodiments, the above commission prediction model may include the following formula:

[0130]

[0131] in, represents the commission data of the tth commission settlement period predicted by the above commission prediction model; represents the first commission data of the tth commission settlement period predicted by the above autoregressive sub-model; represents the second commission data of the tth commission settlement period predicted by the above linear regression sub-model; β represents the basic commission data; p represents the number of historical commission data; The coefficient representing the historical commission data of the ti-th commission settlement period; Y t-i represents the historical commission data of the ti-th commission settlement period; q represents the number of sales indicators in the sales data; θ j represents the coefficient of the j-th sales indicator; X j,tRepresents the j-th sales indicator generated in the t-th commission settlement cycle.

[0132] It should be noted that the above autoregressive sub-model may also include an error adjustment term, that is, the autoregressive sub-model may include the following formula:

[0133]

[0134] Among them, ∈ represents the error adjustment term.

[0135] The sales index can be an index converted from various types of data in the sales data. For example, the sales index can be sales amount, promotion intensity, customer interaction intensity, etc.; it can also be an index used to indicate the season, where the value of this sales index is 1 for spring, 2 for summer, 3 for autumn, and 4 for winter.

[0136] It should be noted that in the commission prediction model shown in the above formula, β, θ j are adjustable model parameters, and p and q reflect the number of input features of the model.

[0137] Furthermore, the optimization objective of the above commission prediction model can be expressed as the following formula:

[0138]

[0139] in, represents the commission data of the tth commission settlement period predicted by the above commission prediction model; Y t represents the commission data of the tth commission settlement period, and N represents the number of model predictions. During the training process, Y t and Y t-i It can be derived from the above historical commission data sample, X j,t It can be derived from the above historical sales data sample.

[0140] When the above-mentioned historical commission data and the above-mentioned sales data are obtained, the historical commission data and the sales data can be input into the above-mentioned trained commission prediction model, and the commission prediction model predicts the commission data corresponding to the above-mentioned seller based on the historical commission data and the sales data.

[0141] It should be noted that it is not necessary to divide the sales scenarios and a universal commission prediction model can be obtained through training.

[0142] Alternatively, in order to introduce a more diversified and personalized commission calculation mechanism so as to improve the accuracy and rationality of the commission paid by the management to the seller, different sales scenarios can be divided, and for each sales scenario, a commission prediction model corresponding to the sales scenario can be trained based on a number of historical commission data and a number of historical sales data generated in the sales scenario. In this case, the sales scenario corresponding to the above-mentioned sales data can be determined first, and then the commission prediction model corresponding to the sales scenario can be obtained, so that the above-mentioned historical commission data and the sales data can be input into the trained commission prediction model corresponding to the sales scenario, and the commission prediction model predicts the commission data corresponding to the seller based on the historical commission data and the sales data.

[0143] In order to adjust the commission ladder in a timely manner to ensure that it always meets the actual needs of the management party and the seller, in some embodiments, a feedback mechanism can be established to introduce feedback from the management party and the seller to adjust the commission calculation rules. Specifically, when the commission data is determined based on the sales data, the commission actually paid by the management party to the seller can be obtained in real time, and the corresponding commission data and the sales data can be used as new samples, and the new samples can be input into the above-mentioned commission prediction model for continued training to obtain the commission prediction model that has been trained.

[0144] In order to further improve the accuracy and rationality of the commission paid by the management to the seller, in some embodiments, it is also possible to regularly calculate an indicator indicating the model performance of the commission prediction model according to a preset model update cycle, and determine whether the commission prediction model needs to be updated based on the indicator. Among them, the indicators used to indicate the model performance of the commission prediction model may include accuracy, precision, recall, etc. When the indicator does not reach the standard threshold, it can be considered that the commission prediction model needs to be updated. If it is determined that the commission prediction model needs to be updated, the model parameters or input features of the commission prediction model can be adjusted to achieve the update of the commission prediction model.

[0145] Specifically, in some embodiments, the model parameters of the above-mentioned commission prediction model can be adjusted to achieve the update of the commission prediction model. For example, the commission prediction model can be retrained, and the retrained commission prediction model can be used as the updated commission prediction model. That is, a number of historical commission data samples and a number of historical sales data samples can be reacquired, and the degree of difference between the commission data predicted by the commission prediction model based on the historical commission data and sales data and the historical commission data can be minimized as the optimization goal, and the historical commission data samples and the historical sales data samples can be input into the commission prediction model for retraining to obtain the retrained commission prediction model to update the commission prediction model.

[0146] Alternatively, in some embodiments, the input features of the above-mentioned commission prediction model may be adjusted to update the commission prediction model. For example, input feature selection may be performed for the commission prediction model based on an input feature selection algorithm, and the input features of the commission prediction model may be adjusted according to the selected input features to update the commission prediction model.

[0147] In some embodiments, the input feature selection algorithm may include a stepwise regression algorithm or a LASSO (Least Absolute Shrinkage and Selection Operator) regression algorithm.

[0148] Stepwise regression and LASSO regression are two methods used to select the set of predictor variables in a model, where the variables are the input features of the model. Stepwise regression is a statistical method for building a regression model by iteratively adding or removing predictor variables. It decides whether to keep a predictor variable in the model based on a specific criterion (such as AIC, BIC, or p-value). Stepwise regression has three main forms: forward selection (starting with no predictors and adding the predictor that improves the model the most in each iteration until no more variables can significantly improve the model), backward elimination (starting with all possible predictors and removing the predictor that contributes the least to the model in each iteration until all remaining variables contribute significantly to the model), and two-way elimination (i.e., variables can be added or removed in each iteration). LASSO regression performs variable selection and regularization while estimating model parameters. It does this by adding a penalty term (i.e., L1 regularization) to the loss function, shrinking some coefficients to zero, effectively excluding those corresponding predictor variables from the model.

[0149] Step 406: Transfer the commission specified by the commission data from the manager's account to the seller's account.

[0150] In this embodiment, when the commission data corresponding to the above-mentioned seller is determined, the commission indicated by the commission data can be transferred from the manager's account to the seller's account to complete the commission payment of the current commission settlement cycle. For example, assuming that the commission data is the commission ratio, the specific amount of the commission can be calculated based on the ratio and the sales amount in the above-mentioned sales data, and the commission of this amount can be transferred from the manager's account to the seller's account; or, assuming that the commission data is the specific amount of the commission, the commission of this amount can be directly transferred from the manager's account to the seller's account. Among them, the form of the commission can be physical resources such as currency, stocks, bonds, bills, accounts receivable, etc. in the real world, or it can be a virtual resource with a specific value.

[0151] In the above technical solution, when the sales data corresponding to the seller is obtained according to the commission settlement cycle and the historical commission data corresponding to the seller is obtained, the historical commission data and the sales data can be input into a trained commission prediction model, and the commission prediction model predicts the commission data corresponding to the seller based on the historical commission data and the sales data, so that the commission specified by the commission data can be transferred from the account of the administrator to the account of the seller.

[0152] By adopting the above method, a commission prediction model is constructed through machine learning, and the commission prediction model is used to determine the commission corresponding to the sales data. This can ensure the accuracy and rationality of the determined commission, thereby providing a more accurate and reasonable commission settlement service.

[0153] Corresponding to the aforementioned method embodiments, the present application also provides device embodiments.

[0154] Please refer to Figure 5 , Figure 5 It is a structural diagram of a device shown in an exemplary embodiment of the present application. At the hardware level, the device includes a processor 502, an internal bus 504, a network interface 506, a memory 508 and a non-volatile memory 510, and of course may also include other required hardware. One or more embodiments of the present application can be implemented based on software, such as the processor 502 reading the corresponding computer program from the non-volatile memory 510 into the memory 508 and then running it. Of course, in addition to the software implementation, one or more embodiments of the present application do not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is to say, the execution subject of the following processing flow is not limited to each logic module, but can also be hardware or logic devices.

[0155] Please refer to Figure 6 , Figure 6It is a block diagram of a commission payment device shown in an exemplary embodiment of the present application.

[0156] The above commission payment device can be applied to Figure 5 The device shown is used to implement the technical solution of the present application. The device includes:

[0157] The acquisition module 602 periodically acquires sales data corresponding to the seller according to a preset commission settlement cycle;

[0158] Matching module 604, determining a sales scenario corresponding to the sales data, and obtaining a commission calculation rule corresponding to the sales scenario; wherein the commission calculation rule is a commission calculation rule with the highest matching degree with the sales scenario obtained by optimizing and calculating a number of historical commission data samples and a number of historical sales data samples generated in the sales scenario based on an adaptive optimization algorithm;

[0159] A calculation module 606, which determines commission data corresponding to the seller according to the commission calculation rule;

[0160] The payment module 608 transfers the commission specified by the commission data from the account of the management party to the account of the seller.

[0161] In some embodiments, the apparatus further comprises a generating module, configured to:

[0162] Obtaining several historical commission data samples and several historical sales data samples generated in the sales scenario;

[0163] Based on the adaptive optimization algorithm, with the optimization goal of maximizing the matching degree of commission data relative to the sales data generated in the sales scenario, iterative optimization calculations are performed on the historical commission data samples and the historical sales data samples to obtain the commission calculation rules with the highest matching degree with the sales scenario.

[0164] In some embodiments, the sales data includes a sales index; the sales index includes a plurality of sales index intervals; the commission calculation rule includes a correspondence between each sales index interval in the plurality of sales index intervals and the commission data;

[0165] The determining the commission data corresponding to the seller according to the commission calculation rule includes:

[0166] According to the commission calculation rule, the sales index interval to which the sales index corresponding to the seller belongs is determined, and the commission data corresponding to the sales index interval is determined as the commission data corresponding to the seller.

[0167] In some embodiments, the generating module is further configured to:

[0168] According to the preset rule update cycle, several historical commission data samples and several historical sales data samples generated in the sales scenario are regularly obtained, and with maximizing the matching degree of the commission data relative to the sales data generated in the sales scenario as the optimization goal, the historical commission data samples and the historical sales data samples are iteratively optimized and calculated again to obtain the commission calculation rule with the highest matching degree with the sales scenario.

[0169] In some embodiments, the adaptive optimization algorithm is a genetic algorithm;

[0170] Based on the adaptive optimization algorithm, taking maximizing the matching degree of commission data with respect to sales data generated in the sales scenario as the optimization goal, performing iterative optimization calculations on the historical commission data samples and the historical sales data samples to obtain the commission calculation rule with the highest matching degree with the sales scenario includes:

[0171] The following steps are iteratively performed until the commission data with the highest matching degree with respect to the sales scenario is generated, and the commission calculation rule with the highest matching degree with the sales scenario is generated based on the commission data:

[0172] Calculating a matching score of the commission data relative to the historical sales data sample generated in the sales scenario; wherein the initial commission data is the historical commission data sample;

[0173] Selecting a preset number of commission data with the highest matching degree;

[0174] The selected commission data are combined and processed to generate new commission data.

[0175] In some embodiments, the matching degree of the commission data relative to the historical sales data sample generated in the sales scenario includes: a weighted sum of a first score obtained by quantifying the satisfaction of the management party and a second score obtained by quantifying the satisfaction of the seller;

[0176] Among them, the management party satisfaction is the management party's satisfaction with the commission data when the historical sales data sample is generated in the sales scenario; the seller party satisfaction is the seller's satisfaction with the commission data when the historical sales data sample is generated in the sales scenario.

[0177] In some embodiments, the sales data includes one or more of the following: sales amount; market activity data; competitor data; seasonal data.

[0178] Please refer to Figure 7 , Figure 7It is a block diagram of another commission payment device shown in an exemplary embodiment of the present application.

[0179] The above commission payment device can be applied to Figure 5 The device shown is used to implement the technical solution of the present application. The device includes:

[0180] The acquisition module 702 periodically acquires sales data corresponding to the seller according to a preset commission settlement cycle, and acquires historical commission data corresponding to the seller;

[0181] A calculation module 704 inputs the historical commission data and the sales data into a trained commission prediction model, and the commission prediction model predicts commission data corresponding to the seller based on the historical commission data and the sales data;

[0182] The payment module 706 transfers the commission specified by the commission data from the account of the management party to the account of the seller.

[0183] In some embodiments, the apparatus further comprises a training module for:

[0184] Obtain several historical commission data samples and several historical sales data samples;

[0185] With minimizing the difference between the commission data predicted by the commission prediction model based on historical commission data and sales data and the historical commission data as the optimization goal, the historical commission data samples and the historical sales data samples are input into the commission prediction model for training to obtain the trained commission prediction model.

[0186] In some embodiments, the commission prediction model includes an autoregressive sub-model and a linear regression sub-model;

[0187] The optimization goal is to minimize the difference between the commission data predicted by the commission prediction model based on the historical commission data and sales data and the historical commission data, input the historical commission data samples and the historical sales data samples into the commission prediction model for training, and obtain the trained commission prediction model, including:

[0188] Inputting the historical commission data sample into the autoregressive sub-model for calculation to obtain first commission data predicted by the autoregressive sub-model based on the historical commission data sample;

[0189] Inputting the historical sales data sample into the linear regression sub-model for calculation to obtain second commission data predicted by the linear regression sub-model based on the historical sales data sample;

[0190] The sum of the first commission data and the second commission data is determined as the commission data predicted by the commission prediction model based on the historical commission data samples and the historical sales data samples, and the degree of difference between the commission data predicted by the commission prediction model and the historical commission data samples is calculated, so as to adjust the model parameters of the commission prediction model according to the degree of difference to complete the training of the commission prediction model.

[0191] In some embodiments, the commission prediction model includes the following formula:

[0192] Y t =Y t1 +Y t2 ;

[0193]

[0194] Among them, Y t represents the commission data of the tth commission settlement period predicted by the commission prediction model; Y t1 represents the first commission data of the tth commission settlement period predicted by the autoregressive sub-model; Y t2 represents the second commission data of the t-th commission settlement period predicted by the linear regression sub-model; β represents the basic commission data; p represents the number of historical commission data; The coefficient representing the historical commission data of the ti-th commission settlement period; Y t-i represents the historical commission data of the ti-th commission settlement period; q represents the number of sales indicators in the sales data; θ j represents the coefficient of the j-th sales indicator; X j,t Represents the j-th sales indicator generated in the t-th commission settlement cycle.

[0195] In some embodiments, the training module is further configured to:

[0196] According to a preset model update cycle, regularly calculating an indicator indicating the model performance of the commission prediction model, and determining whether the commission prediction model needs to be updated according to the indicator;

[0197] If it is determined that the commission prediction model needs to be updated, the model parameters or input features of the commission prediction model are adjusted to update the commission prediction model.

[0198] In some embodiments, adjusting the model parameters of the commission prediction model to update the commission prediction model includes:

[0199] Re-obtain several historical commission data samples and several historical sales data samples;

[0200] Taking minimizing the difference between the commission data predicted by the commission prediction model based on historical commission data and sales data and the historical commission data as the optimization goal, the historical commission data samples and the historical sales data samples are input into the commission prediction model for retraining to obtain the retrained commission prediction model, so as to update the commission prediction model.

[0201] In some embodiments, adjusting the input features of the commission prediction model to update the commission prediction model includes:

[0202] Based on the input feature selection algorithm, input feature selection is performed for the commission prediction model, and the input features of the commission prediction model are adjusted according to the selected input features to update the commission prediction model.

[0203] In some embodiments, the input feature selection algorithm includes a stepwise regression algorithm or a LASSO regression algorithm.

[0204] In some embodiments, the sales data includes one or more of the following: sales amount; market activity data; competitor data; seasonal data.

[0205] For the device embodiment, it basically corresponds to the method embodiment, so the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the technical solution of the present application.

[0206] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which may be in the form of a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver, a game console, a tablet computer, a wearable device or a combination of any of these devices.

[0207] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0208] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0209] Computer-readable media include permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0210] It should be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0211] The above specific embodiments of the present application are described. Other embodiments are within the scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0212] The terms used in one or more embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present application. The singular forms of "a", "said" and "the" are also intended to include plural forms, unless the context clearly indicates other meanings. The term "and / or" refers to and includes any or all possible combinations of one or more associated listed items.

[0213] The description of the terms "one embodiment", "some embodiments", "example", "specific example" or "an implementation method" etc. used in one or more embodiments of the present application means that the specific features or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present application. The schematic description of these terms is not necessarily for the same embodiment. Moreover, the specific features or characteristics described can be combined in a suitable manner in one or more embodiments of the present application. In addition, different embodiments and specific features or characteristics in different embodiments can be combined without contradicting each other.

[0214] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in one or more embodiments of the present application, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determination".

[0215] The above description is merely a preferred embodiment of one or more embodiments of the present application and is not intended to limit one or more embodiments of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present application shall be included in the scope of protection of one or more embodiments of the present application.

[0216] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

Claims

1. A commission payment method, comprising: According to the preset commission settlement cycle, regularly obtain sales data corresponding to the seller; Determine a sales scenario corresponding to the sales data, and obtain a commission calculation rule corresponding to the sales scenario; wherein the commission calculation rule is a commission calculation rule with the highest matching degree with the sales scenario obtained by optimizing and calculating a number of historical commission data samples and a number of historical sales data samples generated in the sales scenario based on an adaptive optimization algorithm; Determining commission data corresponding to the seller according to the commission calculation rule; The commission specified by the commission data is transferred from the account of the management party to the account of the seller.

2. The method according to claim 1, further comprising: Obtaining several historical commission data samples and several historical sales data samples generated in the sales scenario; Based on the adaptive optimization algorithm, with the optimization goal of maximizing the matching degree of commission data relative to the sales data generated in the sales scenario, iterative optimization calculations are performed on the historical commission data samples and the historical sales data samples to obtain the commission calculation rules with the highest matching degree with the sales scenario.

3. According to the method of claim 1 or 2, the sales data includes a sales index; the sales index includes a plurality of sales index intervals; the commission calculation rule includes a corresponding relationship between each sales index interval in the plurality of sales index intervals and the commission data; The determining the commission data corresponding to the seller according to the commission calculation rule includes: According to the commission calculation rule, the sales index interval to which the sales index corresponding to the seller belongs is determined, and the commission data corresponding to the sales index interval is determined as the commission data corresponding to the seller.

4. The method according to claim 2, further comprising: According to the preset rule update cycle, several historical commission data samples and several historical sales data samples generated in the sales scenario are regularly obtained, and with maximizing the matching degree of the commission data relative to the sales data generated in the sales scenario as the optimization goal, the historical commission data samples and the historical sales data samples are iteratively optimized and calculated again to obtain the commission calculation rule with the highest matching degree with the sales scenario.

5. The method according to claim 2, wherein the adaptive optimization algorithm is a genetic algorithm; Based on the adaptive optimization algorithm, taking maximizing the matching degree of commission data with respect to sales data generated in the sales scenario as the optimization goal, performing iterative optimization calculations on the historical commission data samples and the historical sales data samples to obtain the commission calculation rule with the highest matching degree with the sales scenario includes: The following steps are iteratively performed until the commission data with the highest matching degree with respect to the sales scenario is generated, and the commission calculation rule with the highest matching degree with the sales scenario is generated based on the commission data: Calculating a matching score of the commission data relative to the historical sales data sample generated in the sales scenario; wherein the initial commission data is the historical commission data sample; Selecting a preset number of commission data with the highest matching degree; The selected commission data are combined and processed to generate new commission data.

6. According to the method of claim 5, the matching degree of the commission data relative to the historical sales data sample generated in the sales scenario comprises: A weighted sum of a first score obtained by quantifying the satisfaction of the management party and a second score obtained by quantifying the satisfaction of the sales party; Among them, the management party satisfaction is the management party's satisfaction with the commission data when the historical sales data sample is generated in the sales scenario; the seller party satisfaction is the seller's satisfaction with the commission data when the historical sales data sample is generated in the sales scenario.

7. According to the method of claim 1, the sales data includes one or more of the following: sales amount; market activity data; competitor data; seasonal data.

8. A commission payment method, comprising: According to a preset commission settlement cycle, regularly obtain sales data corresponding to the seller, and obtain historical commission data corresponding to the seller; Inputting the historical commission data and the sales data into a trained commission prediction model, and having the commission prediction model predict the commission data corresponding to the seller based on the historical commission data and the sales data; The commission specified by the commission data is transferred from the account of the management party to the account of the seller.

9. The method according to claim 8, further comprising: Obtain several historical commission data samples and several historical sales data samples; With minimizing the difference between the commission data predicted by the commission prediction model based on historical commission data and sales data and the historical commission data as the optimization goal, the historical commission data samples and the historical sales data samples are input into the commission prediction model for training to obtain the trained commission prediction model.

10. The method according to claim 9, wherein the commission prediction model comprises an autoregressive sub-model and a linear regression sub-model; The optimization goal is to minimize the difference between the commission data predicted by the commission prediction model based on the historical commission data and sales data and the historical commission data, input the historical commission data samples and the historical sales data samples into the commission prediction model for training, and obtain the trained commission prediction model, including: Inputting the historical commission data sample into the autoregressive sub-model for calculation to obtain first commission data predicted by the autoregressive sub-model based on the historical commission data sample; Inputting the historical sales data sample into the linear regression sub-model for calculation to obtain second commission data predicted by the linear regression sub-model based on the historical sales data sample; The sum of the first commission data and the second commission data is determined as the commission data predicted by the commission prediction model based on the historical commission data samples and the historical sales data samples, and the degree of difference between the commission data predicted by the commission prediction model and the historical commission data samples is calculated, so as to adjust the model parameters of the commission prediction model according to the degree of difference to complete the training of the commission prediction model.

11. The method according to claim 10, wherein the commission prediction model comprises the following formula: AND t =And t1 +Y t2 ; in, Y t represents the commission data of the tth commission settlement period predicted by the commission prediction model; Y t1 represents the first commission data of the tth commission settlement period predicted by the autoregressive sub-model; Y t2 represents the second commission data of the t-th commission settlement period predicted by the linear regression sub-model; β represents the basic commission data; p represents the number of historical commission data; The coefficient representing the historical commission data of the ti-th commission settlement period; Y t-i represents the historical commission data of the ti-th commission settlement period; q represents the number of sales indicators in the sales data; θ j represents the coefficient of the j-th sales indicator; X j,t Represents the j-th sales indicator generated in the t-th commission settlement cycle.

12. The method according to claim 9, further comprising: According to a preset model update cycle, regularly calculating an indicator indicating the model performance of the commission prediction model, and determining whether the commission prediction model needs to be updated according to the indicator; If it is determined that the commission prediction model needs to be updated, the model parameters or input features of the commission prediction model are adjusted to update the commission prediction model.

13. The method according to claim 12, wherein adjusting the model parameters of the commission prediction model to update the commission prediction model comprises: Re-obtain several historical commission data samples and several historical sales data samples; Taking minimizing the difference between the commission data predicted by the commission prediction model based on historical commission data and sales data and the historical commission data as the optimization goal, the historical commission data samples and the historical sales data samples are input into the commission prediction model for retraining to obtain the retrained commission prediction model, so as to update the commission prediction model.

14. The method according to claim 12, wherein adjusting the input features of the commission prediction model to update the commission prediction model comprises: Based on the input feature selection algorithm, input feature selection is performed for the commission prediction model, and the input features of the commission prediction model are adjusted according to the selected input features to update the commission prediction model.

15. The method according to claim 14, wherein the input feature selection algorithm comprises a stepwise regression algorithm or a LASSO regression algorithm.

16. The method according to claim 8, wherein the sales data comprises one or more of the following: sales amount; market activity data; competitor data; seasonality data.

17. A commission payment device, comprising: The acquisition module regularly acquires the sales data corresponding to the seller according to the preset commission settlement cycle; A matching module determines a sales scenario corresponding to the sales data, and obtains a commission calculation rule corresponding to the sales scenario; wherein the commission calculation rule is a commission calculation rule with the highest matching degree with the sales scenario obtained by optimizing and calculating a number of historical commission data samples and a number of historical sales data samples generated in the sales scenario based on an adaptive optimization algorithm; A calculation module, which determines commission data corresponding to the seller according to the commission calculation rule; The payment module transfers the commission specified by the commission data from the account of the management party to the account of the seller.

18. A commission payment device, comprising: An acquisition module, which regularly acquires sales data corresponding to the seller according to a preset commission settlement cycle, and acquires historical commission data corresponding to the seller; A calculation module, inputting the historical commission data and the sales data into a trained commission prediction model, and having the commission prediction model predict the commission data corresponding to the seller based on the historical commission data and the sales data; The payment module transfers the commission specified by the commission data from the account of the management party to the account of the seller.

19. An electronic device comprising: processor; a memory for storing processor-executable instructions; The processor implements the method according to any one of claims 1 to 16 by running the executable instructions.

20. A computer-readable storage medium having computer instructions stored thereon, wherein the instructions are executed by a processor to implement the method according to any one of claims 1 to 16.

Images