Customer quantity prediction method, device, equipment, medium and program product

Abstract

Embodiments of the present application relate to the technical field of big data analysis, and disclose a customer quantity prediction method, device, equipment, medium and program product, wherein the method comprises: obtaining total customer quantity of a target analysis product in a current analysis period; inputting a result of adding one to a period order value of the current analysis period to a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period; wherein the target customer quantity analysis model is a model constructed based on customer quantity change and inter-customer relationship of the target analysis product. The technical solution of the present embodiment can simulate actual customer quantity change law based on customer quantity change and customer relationship information, and predict customer quantity in a future period of time, thereby more efficiently guiding product business development.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of big data analysis, and particularly relate to a customer quantity prediction method and device, equipment, medium and program product. BACKGROUND

[0002] It is very important to dynamically predict the development and change of the customer group quantity of any product in the process of business development and operation, and the result of quantity change prediction can be used to guide the specific work in the process of business development. The more accurate the result of customer quantity dynamic change prediction is, the more guiding significance it has for subsequent business operation activities.

[0003] However, in the current customer quantity dynamic prediction process, although the potential customer quantity and the lost customer quantity are considered, the customers of different sources are not further divided, and further algorithm optimization is needed to be closer to the change characteristics of the product market. SUMMARY

[0004] Embodiments of the present application provide a customer quantity prediction method, device, equipment, medium and program product to simulate the actual customer quantity change rule based on customer quantity change and customer relationship information, predict the customer quantity in a future period of time, and improve the accuracy of customer quantity prediction and more efficiently guide the development of product business.

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

[0006] obtaining the total customer quantity of a target analysis product in a current analysis period;

[0007] inputting the total customer quantity and the result of adding one to the cycle order value of the current analysis period into a target customer quantity analysis model to obtain the total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period;

[0008] The target customer quantity analysis model is a model constructed based on the customer quantity change and the association relationship between customers of the target analysis product.

[0009] In a second aspect, embodiments of the present application provide a customer quantity prediction device, which comprises:

[0010] a prediction data acquisition module configured to obtain the total customer quantity of a target analysis product in a current analysis period;

[0011] a prediction result determination module configured to input a result of adding one to the total customer quantity and the cycle order value of the current analysis cycle into a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in a next analysis cycle adjacent to the current analysis cycle.

[0012] The target customer quantity analysis model is a model constructed based on customer quantity changes of the target analysis product and inter-customer relationships.

[0013] In a third aspect, an embodiment of the present application further provides a computer device, which comprises:

[0014] one or more processors;

[0015] a memory configured to store one or more programs;

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the customer quantity prediction method provided in any embodiment of the present application.

[0017] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the customer quantity prediction method provided in any embodiment of the present application.

[0018] In a fifth aspect, an embodiment of the present application further provides a computer program product having a computer program stored thereon, which, when executed by a processor, implements the customer quantity prediction method provided in any embodiment of the present application.

[0019] The embodiments of the above application have the following advantages or beneficial effects:

[0020] In the embodiments of the present application, the total customer quantity of the target analysis product in a current analysis cycle is obtained; a result of adding one to the total customer quantity and the cycle order value of the current analysis cycle is input into a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in a next analysis cycle adjacent to the current analysis cycle; and the target customer quantity analysis model is a model constructed based on customer quantity changes of the target analysis product and inter-customer relationships. The technical scheme of the embodiments of the present application solves the problem that the current method for predicting product customer quantity does not comprehensively consider factors, resulting in low prediction result accuracy, and can simulate actual customer quantity change rules based on customer quantity changes and customer relationship information, predict customer quantity in a future period of time, and improve customer quantity prediction accuracy to more efficiently guide product business development. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 is a flow chart of a customer quantity prediction method provided by an embodiment of the present application;

[0022] Figure 2 is a flow chart of a customer quantity prediction method provided by an embodiment of the present application;

[0023] Figure 3 is a flow chart of a customer quantity prediction method provided by an embodiment of the present application;

[0024] Figure 4 is a flow chart of a customer quantity prediction device provided by an embodiment of the present application

[0025] Figure 5 is a structural schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0026] The present application will be further described below in conjunction with the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings, but not all the structures.

[0027] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, but cannot be understood as indicating or implying relative importance. The acquisition, storage, use, processing, etc. of data in the technical solution of the present application all comply with the relevant provisions of the national laws and regulations.

[0028] Embodiment I

[0029] Figure 1 A flow chart of a customer quantity prediction method provided by an embodiment of the present application, the present embodiment can be applicable to the prediction of the customer quantity of any product, and is particularly suitable for the prediction of the customer quantity of a banking product. The method can be executed by a customer quantity prediction device, which can be realized by software and / or hardware, and integrated in a computer device with application development function.

[0030] As shown in Figure 1 , the customer quantity prediction method comprises the following steps:

[0031] S110, acquiring the total customer quantity of the target analysis product in the current analysis period.

[0032] The target analysis product can be any product that needs to be concerned about its business development, such as a game application product, a social application product, or a financial product that needs customer quantity or traffic support.

[0033] The analysis period can be understood as a time span for customer quantity and other customer data statistics and analysis, such as every day, every week, every month, or other time steps. The current analysis period can be the analysis period corresponding to the current time, counted from the online time of the target analysis product, and divided according to the length of the analysis period. The time span of customer data can be denoted as t = 0, 1, 2, …, T. When t is 0, it represents the initial time, that is, the online time of the target analysis product. When t is greater than 0, it represents the corresponding analysis period after the initial time, and therefore, t can also be used as the order of the analysis period.

[0034] The total customer data is the total customer quantity accumulated and determined in the current analysis period.

[0035] Specifically, the total customer quantity of the target analysis product in the current analysis period can be obtained by reading the business data of the business system where the target analysis product is located, or can be input to the customer quantity prediction device after the user performs customer quantity statistics.

[0036] In S120, the total customer quantity and the result of adding one to the cycle order value of the current analysis period are input into a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in the next analysis period adjacent to the current analysis period.

[0037] The target customer quantity analysis model is a model constructed based on the customer quantity change and the association between customers of the target analysis product, and is used to predict the total customer quantity in the next analysis period based on the total customer quantity data in the current analysis period. The growth of customer quantity is a complex dynamic process. During the process of product business development, customers will recommend and promote their favorite products to their friends and relatives, and therefore, there will be an association between new customers and old customers in terms of introduction and being introduced. At the same time, there will also be customers who give up continuing to consume the product after experiencing it, that is, a customer can be in a retention state in one period, but can not be a customer of this product in the next period, and becomes a loss state. That is, the dynamic growth and dynamic decline of customer quantity are synchronized. Therefore, the model constructed based on the customer quantity change and the association between customers of the target analysis product is more suitable for the market characteristics of the product.

[0038] The total number of customers in the current period and the result of adding one to the period order value of the current analysis period, i.e. the period order value of the next analysis period, are input into the target customer quantity analysis model, and the corresponding prediction result can be obtained.

[0039] Further, the target customer quantity analysis model includes a non-historical analysis period customer recommendation new customer quantity prediction unit, a historical analysis period customer recommendation new customer quantity prediction unit, and a historical analysis period customer retention state prediction unit. The function model of the non-historical analysis period customer recommendation new customer quantity prediction unit includes a preset logistic regression function based on the total customer quantity of the last analysis period of the target prediction analysis period, and the product of a uniformly distributed random variable function in a preset numerical range. The function model of the historical analysis period customer retention state prediction unit conforms to the Bernoulli binomial distribution. The function model of the historical analysis period customer recommendation new customer quantity prediction unit conforms to the Poisson distribution.

[0040] Now, the formula of the model is described. The target customer quantity analysis model can be represented as: Where t represents the tth analysis period, N(t) is the total customer quantity of the tth analysis period, the total customer quantity at the initial moment can be represented by N(0), denoted as n0. When t>0, for the total customer quantity of each analysis period, the above formula consists of three parts, t-1 can be understood as the current analysis period of the known total customer quantity, and t represents the next analysis period adjacent to the current analysis period which needs to be predicted.

[0041] For the composition of the target customer quantity analysis model, Z(t) is the number of new non-old customers recommended in the tth analysis period, and the number of new customers who become customers spontaneously, corresponding to the non-historical analysis period customer recommendation new customer quantity prediction unit. Z(t) can be preset as f(N(t-1))*I(t). Where I(t) is a random variable subject to 0-1 uniform distribution, and f(N(t-1)) is a function related to N(t-1), i.e. a preset logistic regression function based on the total customer quantity of the last analysis period of the target prediction analysis period, which can be represented as Wherein, β, γ are parameters determined by parameter estimation based on historical customer data in the model construction process. The trend of f(N(t-1)) function is S-shaped curve, which is first rapidly rising and then slowly declining. The function f selected by this logistic regression function configuration can be more matched with the actual product business development scenario. Generally speaking, the total number of customers in an industry is limited. At the initial stage of product marketing, the number of customers generated spontaneously will inevitably increase, but when it reaches a certain level, the market will tend to be saturated, which will inevitably lead to a decrease in the number of new customers. The S-shaped curve configuration of the logistic regression function is exactly matched with this characteristic of first accelerating upward and then slowly downward, so it can better predict the number of customers in the future period.

[0042] Further, N(t-1) represents the total number of customers in the t-1th analysis period, wherein, X i (t) represents the i th customer in N(t-1), and the individual state in the t th analysis period, corresponding to the customer retention state prediction unit of the historical analysis period. X i (t) conforms to Bernoulli binomial distribution, that is, there are only two values of 0 and 1, 0 represents that the customer has left, and 1 represents that the customer still remains, that is, X i (t) ~ B(1, p(t)). In this embodiment, p(t) is preset The value of this function is a value that gradually decreases with time t, indicating that the probability of customer retention is potentially decreasing over time. That is, as market competition intensifies and there are more choices, customers are more likely to leave, which is more in line with the general situation in the market. α is a parameter determined by parameter estimation based on historical customer data in the model construction process.

[0043] Y i (t) represents the i th customer in N(t-1), and the recommended new customer number in the t th analysis period, corresponding to the customer recommended new customer number prediction unit of the historical analysis period. Specifically, in this embodiment, Y i (t) is a Poisson distribution with parameter λ, that is:

[0044] Wherein, k = 1, 2, 3,...; λ is a parameter determined by parameter estimation based on historical customer data in the model construction process. This is because whether a customer in a historical analysis period will recommend a new customer and the number of new customers recommended are random and uncertain.

[0045] The target customer number analysis model composed of the above parts is in line with the market development law of the product, and can more accurately predict the total number of customers in the target prediction time period, so as to more efficiently guide the business development of the target analysis product.

[0046] The technical scheme of the embodiment obtains the total customer quantity of the target analysis product in a current analysis period, inputs the total customer quantity and a result of adding one to a period order value of the current analysis period into a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period, wherein the target customer quantity analysis model is a model constructed based on customer quantity changes of the target analysis product and a correlation between customers. The technical scheme of the embodiment solves the problem that factors are not comprehensive in a current product customer quantity prediction method, which leads to low prediction result accuracy, and can realize simulation of actual customer quantity change rules based on customer quantity changes and customer relationship information, prediction of customer quantity in a future period of time, and improvement of customer quantity prediction accuracy to more efficiently guide product business development.

[0047] Embodiment Two

[0048] Figure 2 A flowchart of a customer quantity prediction method provided by the embodiment two, the embodiment and the customer quantity prediction method in the above embodiment belong to the same inventive concept, and further describes a process of continuously predicting customer quantity by using a target customer quantity analysis model. The method can be executed by a customer quantity prediction device, and the device can be realized by software and / or hardware and integrated in a computer device with application development function.

[0049] As shown in the figure, the customer quantity prediction method includes the following steps: Figure 2

[0050] S210, obtaining a total customer quantity of a target analysis product in a current analysis period.

[0051] The total customer quantity in the current analysis period is a result of actual customer quantity statistics in the current analysis period.

[0052] S220, inputting a result of adding one to a period order value of the current analysis period to a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period.

[0053] Suppose that the period order value of the current analysis period is t, then the total customer quantity prediction result obtained in the step is a total customer quantity prediction value of the t+1th analysis period.

[0054] ​S230, inputting the total customer quantity prediction result and the value of the period order value of the next analysis period plus one into the target customer quantity analysis model to obtain the predicted total customer quantity of the next analysis period adjacent to the next analysis period, and repeating the process until the total customer quantity prediction result of the analysis period containing the first preset number of analysis periods is obtained.

[0055] In this step, the total customer quantity prediction value of the t+1th analysis period can be used to predict the total customer quantity prediction value of the t+2th analysis period. Then, the total customer quantity prediction value of the t+2th analysis period can be used to predict the total customer quantity prediction value of the t+3th analysis period, and even more (such as the first preset number of) analysis period total customer quantity prediction values can be predicted.

[0056] It can be understood that the accuracy of the prediction result obtained by predicting based on the prediction value will decrease to a certain extent, but the trend of data change can be used as a reference for customer quantity analysis.

[0057] S240, determining the change trend of the customer quantity of the target analysis product in the first preset number of analysis periods according to the total customer quantity prediction result of the first preset number of analysis periods.

[0058] For the change in the value of the total customer quantity prediction result in the first preset number of consecutive analysis periods, the change trend of the total customer quantity prediction value can be determined by dotting and curve fitting. According to the trend of the curve, the rising or falling trend of the total customer quantity in the future few analysis periods and whether the rising or falling speed is slow or fast can be determined.

[0059] S250, determining the development period of the target analysis product based on the change trend.

[0060] Specifically, the development period includes a growth period and a shrinkage period. The total customer quantity prediction value shows an upward trend, which corresponds to the growth period of the development period, and conversely, the total customer quantity prediction value shows a downward trend, which corresponds to the shrinkage period of the development period. Accordingly, the corresponding business development strategy can be selected according to the judgment result of the development period or the shrinkage period to maintain the business development of the target analysis product.

[0061] The technical scheme of the embodiment obtains the total customer quantity of the target analysis product in the current analysis period, inputs the total customer quantity and the result of adding one to the period order value of the current analysis period into the target customer quantity analysis model, obtains the total customer quantity prediction result of the target analysis product in the next analysis period adjacent to the current analysis period, and then continuously predicts the total customer quantity prediction result of future multiple analysis periods based on the customer quantity prediction result, so that the specific business development strategy can be determined according to the change trend of the total customer quantity prediction result. The technical scheme of the embodiment solves the problem that the current product customer quantity prediction method does not comprehensively consider factors, resulting in low prediction result accuracy, can simulate the actual customer quantity change rule based on customer quantity change and customer relationship information, predict the customer quantity in the future period of time, and improve the customer quantity prediction accuracy to more efficiently guide the product business development work.

[0062] Embodiment three

[0063] Figure 3 A flowchart of a customer quantity prediction method provided by the third embodiment of the application, the embodiment and the customer quantity prediction method in the above embodiments belong to the same inventive concept, and further describes the construction process of the target customer quantity analysis model. The method can be executed by a customer quantity prediction device, which can be realized by software and / or hardware and integrated in a computer device with application development function.

[0064] As shown in Figure 3 The customer quantity prediction method includes the following steps:

[0065] S310, obtaining customer quantity information and inter-customer relationship information of a target analysis product in a second preset number of historical analysis periods.

[0066] For the customer quantity information and the inter-customer relationship information in the historical stage of the target analysis product, a preset time span can be used as an analysis period to sort the customer quantity information and the inter-customer relationship information in the historical stage. The customer quantity in the time range corresponding to the start point and the end point of an analysis period time is counted, and the association relationship between customers is recorded.

[0067] The association relationship can be an introduction and referral relationship. For example, customer A introduces customer B to use or experience the target analysis product, then customer A is the introducer and customer B is the introduced person.

[0068] S320, determining, according to the customer quantity information and the inter-customer association relationship information, a customer retention state of a previous historical analysis period in each historical analysis period, a customer recommendation new customer quantity of the previous historical analysis period, and a non-historical analysis period customer recommendation new customer quantity.

[0069] Based on the data collation result of the previous step, the customer quantity information and the inter-customer association relationship information in the continuous historical analysis period can be determined as sample data for model construction. By comparing the customer quantity information and the customer association relationship in two continuous analysis periods, the retention state of the customers in the previous historical analysis period in the adjacent next historical analysis period can be determined, and whether the customers in the previous historical analysis period introduced customers to become new customers in the adjacent next historical analysis period. Accordingly, the total customer quantity in the next historical analysis period minus the number of old customers remaining from the previous historical analysis period and the number of new customers introduced by the old customers is the non-historical analysis period customer recommendation new customer quantity, that is, the number of customers who spontaneously accept the target analysis product to become new customers.

[0070] S330, determining model parameters in the target customer quantity analysis model based on the customer retention state of the previous historical analysis period in each historical analysis period, the customer recommendation new customer quantity of the previous historical analysis period, and the non-historical analysis period customer recommendation new customer quantity, and completing the model construction process.

[0071] Specifically, in the process of determining the model parameters in the target customer quantity analysis model, the customer retention state of the previous historical analysis period in each historical analysis period, the customer recommendation new customer quantity of the previous historical analysis period, and the non-historical analysis period customer recommendation new customer quantity are input into the initial customer quantity analysis model to predict the total customer quantity of each historical analysis period. Since the total customer quantity in each historical analysis period is known, the parameters in the initial customer quantity analysis model can be estimated to determine the final parameter value.

[0072] The initial customer quantity analysis model is a model constructed based on the customer quantity change of the target analysis product and the association relationship between customers. The initial customer quantity analysis model includes a customer recommendation new customer quantity prediction unit in a non-history analysis period, a customer recommendation new customer quantity prediction unit in a history analysis period, and a customer retention state prediction unit in the history analysis period. The function model of the customer recommendation new customer quantity prediction unit in the non-history analysis period includes a preset logistic regression function established based on the total customer quantity of the last analysis period of the target prediction analysis period, and the product of a random variable function uniformly distributed in a preset numerical range. The function model of the customer retention state prediction unit in the history analysis period conforms to a Bernoulli binomial distribution. The function model of the customer recommendation new customer quantity prediction unit in the history analysis period conforms to a Poisson distribution. Specifically, reference can be made to the formula expression of the model in Embodiment 1.

[0073] The parameters in the initial customer quantity analysis model can be determined by a parameter determination method as follows to obtain the target customer quantity analysis model.

[0074] First, the parameters in the initial customer quantity analysis model can be determined by a sample estimation algorithm to obtain the target customer quantity analysis model.

[0075] For example, for the parameter α, the probability p(t) of the customer quantity retained in the t-1th history analysis period in the tth history analysis period is estimated, and then the value of t is brought into the probability formula to calculate the estimated value of α by using The estimated value of α can be directly calculated by back calculation.

[0076] For another example, for the parameter λ, which is a parameter of the probability distribution function of Y i (t) conforming to the Poisson distribution, the mean value can be calculated according to the sample value of Y i (t) in the tth history analysis period, and the estimated value of λ can be obtained by back calculation using the mean value of the Poisson distribution being 1 / λ.

[0077] Similarly, the estimated value of f(N(t-1)) can be obtained by using the sample observation value of Z(t), and then the estimated values of β and γ can be obtained by back calculation using the calculation formula of f(N(t-1)).

[0078] After the values of the parameters are determined, the structure of the target customer quantity analysis model is uniquely determined.

[0079] Second, the parameters in the initial customer quantity analysis model can also be determined by a maximum likelihood algorithm to obtain the target customer quantity analysis model.

[0080] In the parameter estimation by using the maximum likelihood method, firstly, the calculation formula of the likelihood function is obtained, and under the model assumption of the embodiment, the following likelihood function value about the sample can be constructed, that is, the joint probability density logarithmic likelihood value of the model constructed sample:

[0081]

[0082] Wherein, the order value of the historical analysis period corresponding to the sample data of the model is t, and the value is 1, …, T.m t is the number of old customers in N(t-1) that do not flow away and are saved in the tth historical analysis period.k it is the number of new customers recommended by the ith customer in the total number of customers in the t-1th historical period in the tth historical analysis period. The to-be-estimated parameters are contained in p(t) and f(N(t-1)). In the case of fixed sample, changing the values of the parameters α, β, γ and λ can change the value of the likelihood value loglik, and the combination of α, β, γ and λ that can make loglik reach the maximum value is searched, and the value at this time is the maximum likelihood estimation value of the parameters. The estimation value can be used as the final determination result of the parameters as the parameters of the target customer quantity analysis model.

[0083] Thirdly, or, the parameters in the initial customer quantity analysis model can also be evaluated and determined by the precision correction algorithm to obtain the target customer quantity analysis model.

[0084] The precision correction method is a method of directly skipping the sample data evaluation and selecting the parameters by using the simulation precision. According to the growth mechanism of N(t) in the model setting, the estimation value of N(t) at each moment is obtained by using the random number generation method of the corresponding distribution for the variables X i (t), Y i (t) and Z(t) that obey a specific distribution. By comparing the real data, the error is calculated. Since the simulation result is affected by the values of the parameters α, β, γ and λ, the parameter combination that can make the error reach the minimum is searched by changing the values of α, β, γ and λ, and is used as the parameters of the target customer quantity analysis model. It is a method of directly using the error change to adjust the model parameter structure.

[0085] S340, obtaining the total customer quantity of the target analysis product in the current analysis period.

[0086] After the above model construction process, the parameters of the target customer quantity analysis model can be determined, and can be applied to the business development analysis of the target analysis product. The total customer quantity in the future analysis period is predicted.

[0087] S350, inputting the result of adding one to the period order value of the current analysis period to the target customer quantity analysis model to obtain the total customer quantity prediction result of the target analysis product in the next analysis period adjacent to the current analysis period.

[0088] The technical scheme of the embodiment obtains the customer information of the target analysis product in the historical stage, obtains the customer retention state of the previous historical analysis period in the plurality of historical analysis periods, the customer recommended new customer quantity of the previous historical analysis period, and the customer recommended new customer quantity in the non-historical analysis period, and performs parameter estimation to obtain the target customer quantity analysis model; then in the process of using the target customer quantity analysis model, the total customer quantity of the target analysis product in the current analysis period is obtained; the result of adding one to the period order value of the current analysis period is input into the target customer quantity analysis model to obtain the total customer quantity prediction result of the target analysis product in the next analysis period adjacent to the current analysis period; wherein the target customer quantity analysis model is a model constructed based on the customer quantity change and the correlation between customers of the target analysis product. The technical scheme of the embodiment solves the problem that the factors are not comprehensive in the current product customer quantity prediction method, which leads to low prediction result accuracy, can simulate the actual customer quantity change rule based on the customer quantity change and customer relationship information, predict the customer quantity in the future period of time, and improve the customer quantity prediction accuracy to more efficiently guide the product business development work.

[0089] Embodiment four

[0090] Figure 4 The structure diagram of a customer quantity prediction device provided by the embodiment four of the application, the embodiment can be applied to the case of predicting the customer quantity of any product, and is particularly suitable for predicting the customer quantity of a bank business product. The device can be realized by software and / or hardware, and integrated in a computer device with application development function.

[0091] As shown in Figure 4 , the customer quantity prediction device comprises a prediction data acquisition module 410 and a prediction result determination module 420.

[0092] The prediction data acquisition module 410 is configured to acquire a total customer quantity of a target analysis product in a current analysis period; and the prediction result determination module 420 is configured to input a result of adding one to a period order value of the current analysis period to a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period, wherein the target customer quantity analysis model is a model constructed based on a customer quantity change of the target analysis product and an association relationship between customers.

[0093] The technical scheme of the embodiment acquires a total customer quantity of a target analysis product in a current analysis period, inputs a result of adding one to a period order value of the current analysis period to a target customer quantity analysis model to obtain a total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period, wherein the target customer quantity analysis model is a model constructed based on a customer quantity change of the target analysis product and an association relationship between customers. The technical scheme of the embodiment solves the problem that a current product customer quantity prediction method does not comprehensively consider factors, resulting in low prediction result accuracy, and can simulate actual customer quantity change rules based on customer quantity change and customer relationship information, predict customer quantity in a future period of time, and improve customer quantity prediction accuracy to more efficiently guide product business development.

[0094] Optionally, the target customer quantity analysis model comprises a non-historical analysis period customer recommendation new customer quantity prediction unit, a historical analysis period customer recommendation new customer quantity prediction unit and a historical analysis period customer retention state prediction unit.

[0095] Optionally, a function model of the non-historical analysis period customer recommendation new customer quantity prediction unit comprises a product of a preset logistic regression function established based on a total customer quantity of a previous analysis period of a target prediction analysis period and a random variable function uniformly distributed in a preset numerical range.

[0096] Optionally, a function model of the historical analysis period customer retention state prediction unit conforms to a Bernoulli binomial distribution.

[0097] Optionally, a function model of the historical analysis period customer recommendation new customer quantity prediction unit conforms to a Poisson distribution.

[0098] Optionally, the prediction result determination module 420 is further configured to:

[0099] inputting the total customer quantity prediction result and a value of the cycle order value of the next analysis cycle plus one into the target customer quantity analysis model to obtain a predicted total customer quantity of a next analysis cycle adjacent to the next analysis cycle, and repeating the process until a total customer quantity prediction result of an analysis cycle containing the first preset number of analysis cycles is obtained.

[0100] Optionally, the prediction result determination module 420 is further configured to:

[0101] determine a variation trend of the customer quantity of the target analysis product in the first preset number of analysis cycles according to the total customer quantity prediction result of the first preset number of analysis cycles;

[0102] determine a development cycle of the target analysis product based on the variation trend, wherein the development cycle includes a growth period and a shrinkage period.

[0103] Optionally, the customer quantity prediction device further includes a model construction module configured to implement a construction process of the target customer quantity analysis model, and specifically includes:

[0104] obtain customer quantity information and inter-customer association relationship information of the target analysis product in a second preset number of historical analysis cycles;

[0105] determine a customer retention state of a previous historical analysis cycle in each historical analysis cycle, a customer recommendation new customer quantity of the previous historical analysis cycle, and a non-historical analysis cycle customer recommendation new customer quantity according to the customer quantity information and the inter-customer association relationship information;

[0106] determine model parameters in the target customer quantity analysis model based on the customer retention state of the previous historical analysis cycle in each historical analysis cycle, the customer recommendation new customer quantity of the previous historical analysis cycle, and the non-historical analysis cycle customer recommendation new customer quantity, and complete the model construction process.

[0107] Optionally, the model construction module is specifically configured to:

[0108] input the customer retention state of the previous historical analysis cycle in each historical analysis cycle, the customer recommendation new customer quantity of the previous historical analysis cycle, and the non-historical analysis cycle customer recommendation new customer quantity into an initial customer quantity analysis model respectively;

[0109] determine each parameter in the initial customer quantity analysis model through a sample estimation algorithm to obtain the target customer quantity analysis model.

[0110] Optionally, the model construction module is specifically configured to:

[0111] The customer retention status of the previous historical analysis period, the number of new customers referred by customers in the previous historical analysis period, and the number of new customers referred by customers in non-historical analysis periods are respectively input into the initial customer number analysis model.

[0112] The parameters in the initial customer number analysis model are evaluated and determined by the maximum likelihood algorithm, and the target customer number analysis model is obtained.

[0113] Optionally, the model building module is specifically used for:

[0114] The customer retention status of the previous historical analysis period, the number of new customers recommended by customers in the previous historical analysis period, and the number of new customers recommended by customers in non-historical analysis periods are respectively input into the initial customer number analysis model.

[0115] The parameters in the initial customer number analysis model are evaluated and determined by the accuracy correction algorithm to obtain the target customer number analysis model.

[0116] The customer number prediction device provided in the embodiments of the present invention can execute the customer number prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0117] It is worth noting that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of the present invention.

[0118] Example 5

[0119] Figure 5 This is a schematic diagram of the structure of a computer device provided in Embodiment 5 of the present invention. Figure 5 A block diagram of an exemplary computer device 12 suitable for implementing embodiments of the present invention is shown. Figure 5 The computer device 12 shown is merely an example and should not be construed as limiting the functionality or scope of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capabilities, such as intelligent controllers and servers, mobile phones, and other terminal devices.

[0120] like Figure 5 As shown, the computer device 12 is represented in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and a bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0121] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0122] Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0123] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 5 Not shown; usually referred to as a "hard drive"). Although Figure 5 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0124] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of the present invention.

[0125] Computer device 12 can also communicate with one or more external devices 14 such as a keyboard, a pointing device, a display 24, etc.; one or more devices that enable a user to interact with computer device 12; and / or any devices (e.g., network card, modem, etc.) that enable computer device 12 to communicate with one or more other computing devices. Such communication can occur via Input / Output (I / O) interface(s) 22. Still yet, computer device 12 can communicate with one or more networks, such as a local area network (LAN), a general wide area network (WAN), and / or a public network (e.g., the Internet) via network adapter 20. As depicted, network adapter 20 communicates with the other components of computer device 12 via bus 18. It should be appreciated that although not shown, other hardware and / or software modules could be used in conjunction with computer device 12. Examples, include, but are not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc. Figure 5

[0126] Processing unit 16 performs various function applications and data processing by running programs stored in system memory 28, such as implementing the customer quantity prediction method provided by the embodiments of the present application, which comprises:

[0127] obtaining a total customer quantity of a target analysis product in a current analysis period;

[0128] inputting a result of adding one to a period order value of the current analysis period to the total customer quantity into a target customer quantity analysis model, to obtain a total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period;

[0129] The target customer quantity analysis model is a model constructed based on a customer quantity change of the target analysis product and a correlation between customers.

[0130] Embodiment six

[0131] The embodiment six provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement a customer quantity prediction method provided by any of the embodiments of the present application, which comprises:

[0132] obtaining a total customer quantity of a target analysis product in a current analysis period;

[0133] inputting a result of adding one to a period order value of the current analysis period to the total customer quantity into a target customer quantity analysis model, to obtain a total customer quantity prediction result of the target analysis product in a next analysis period adjacent to the current analysis period;

[0134] ​The target customer quantity analysis model is a model constructed based on the customer quantity change of the target analysis product and the correlation between customers.

[0135] The computer storage medium of the embodiments of the present application can adopt any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of the computer readable storage medium include: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus.

[0136] The computer readable signal medium can include a data signal propagating in a baseband or as part of a carrier wave propagating through a transmission medium, in which the computer readable program code is embodied. Such a propagating data signal can take many forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. The computer readable signal medium can also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate or transport program for use by or in connection with an instruction execution system, apparatus or device.

[0137] The program code contained on the computer readable medium can be transmitted in any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination thereof.

[0138] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0139] Those skilled in the art will appreciate that the modules or steps of the application described above can be implemented in a general purpose computer, which can be centralized or distributed over a network of multiple computers, and optionally implemented as program code executable by a computer and stored in a storage device for execution by the computer, or implemented as individual integrated circuit modules, or implemented as a combination of some of the modules or steps in a single integrated circuit module. Thus, the present application is not limited to any particular combination of hardware and software.

[0140] Embodiment Seven

[0141] The embodiments of the present application further provide a computer program product, comprising a computer program which, when executed by a processor, implements the customer quantity prediction method according to any of the embodiments of the present application.

[0142] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0143] Note that the above merely describes preferred embodiments of the present application and the principles of the technology applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, modifications and substitutions can be made without departing from the scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and can include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. A method for predicting the number of customers, characterized in that, include: Obtain the total number of customers for the target analytics product during the current analytics period; The result of adding one to the total number of customers and the cycle order value of the current analysis cycle is input into the target customer number analysis model to obtain the predicted total number of customers of the target analysis product in the next analysis cycle adjacent to the current analysis cycle. The target customer number analysis model is a model constructed based on the changes in the number of customers of the target analysis product and the relationship between customers. The target customer quantity analysis model includes a customer referral new customer quantity prediction unit in non-historical analysis period, a customer referral new customer quantity prediction unit in historical analysis period, and a customer retention status prediction unit in historical analysis period. The function model of the customer recommendation new customer number prediction unit in the non-historical analysis period includes the product of a preset logistic regression function based on the total number of customers in the previous analysis period and a function of random variables uniformly distributed within a preset numerical range; the function model of the customer retention status prediction unit in the historical analysis period conforms to a Bernoulli binomial distribution; and the function model of the customer recommendation new customer number prediction unit in the historical analysis period conforms to a Poisson distribution.

2. The method according to claim 1, characterized in that, The method further includes: The predicted total number of customers and the value of the next analysis period plus one are input into the target customer number analysis model to obtain the predicted total number of customers for the next analysis period adjacent to the next analysis period. This process is repeated until the predicted total number of customers for the analysis period containing the first preset number of analysis periods is obtained.

3. The method according to claim 2, characterized in that, The method further includes: Based on the total number of customers predicted for the first preset number of analysis periods, determine the trend of the number of customers of the target analysis product during the first preset number of analysis periods; The development cycle of the target analytical product is determined based on the aforementioned trend, wherein the development cycle includes a growth period and a decline period.

4. The method according to any one of claims 1-3, characterized in that, The process of constructing the target customer quantity analysis model includes: Obtain customer quantity information and customer relationship information for the target analysis product within a second preset historical analysis period; Based on the customer quantity information and the customer relationship information, determine the customer retention status of the previous historical analysis period, the number of new customers recommended by the customer in the previous historical analysis period, and the number of new customers recommended by the customer in non-historical analysis periods. Based on the customer retention status of the previous historical analysis period, the number of new customers recommended by customers in the previous historical analysis period, and the number of new customers recommended by customers in non-historical analysis periods, the model parameters in the target customer quantity analysis model are determined, and the model construction process is completed.

5. The method according to claim 4, characterized in that, The determination of model parameters in the target customer quantity analysis model based on the customer retention status of the previous historical analysis period, the number of new customers referred by customers in the previous historical analysis period, and the number of new customers referred by customers in non-historical analysis periods includes: The customer retention status of the previous historical analysis period, the number of new customers recommended by customers in the previous historical analysis period, and the number of new customers recommended by customers in non-historical analysis periods are respectively input into the initial customer number analysis model. The parameters in the initial customer number analysis model are evaluated and determined by the sample estimation algorithm to obtain the target customer number analysis model.

6. The method according to claim 4, characterized in that, The determination of model parameters in the target customer quantity analysis model based on the customer retention status of the previous historical analysis period, the number of new customers referred by customers in the previous historical analysis period, and the number of new customers referred by customers in non-historical analysis periods includes: The customer retention status of the previous historical analysis period, the number of new customers referred by customers in the previous historical analysis period, and the number of new customers referred by customers in non-historical analysis periods are respectively input into the initial customer number analysis model. The parameters in the initial customer number analysis model are evaluated and determined by the maximum likelihood algorithm, and the target customer number analysis model is obtained.

7. The method according to claim 4, characterized in that, The determination of model parameters in the target customer quantity analysis model based on the customer retention status of the previous historical analysis period, the number of new customers referred by customers in the previous historical analysis period, and the number of new customers referred by customers in non-historical analysis periods includes: The customer retention status of the previous historical analysis period, the number of new customers recommended by customers in the previous historical analysis period, and the number of new customers recommended by customers in non-historical analysis periods are respectively input into the initial customer number analysis model. The parameters in the initial customer number analysis model are evaluated and determined by the accuracy correction algorithm to obtain the target customer number analysis model.

8. A customer quantity prediction device, characterized in that, The device includes: The predictive data acquisition module is used to obtain the total number of customers of the target analysis product within the current analysis period; The prediction result determination module is used to input the total number of customers and the cycle order value of the current analysis cycle plus one into the target customer number analysis model to obtain the prediction result of the total number of customers of the target analysis product in the next analysis cycle adjacent to the current analysis cycle. The target customer number analysis model is a model constructed based on the changes in the number of customers of the target analysis product and the relationship between customers. The target customer quantity analysis model includes a customer referral new customer quantity prediction unit in non-historical analysis period, a customer referral new customer quantity prediction unit in historical analysis period, and a customer retention status prediction unit in historical analysis period. The function model of the customer recommendation new customer number prediction unit in the non-historical analysis period includes the product of a preset logistic regression function based on the total number of customers in the previous analysis period and a function of random variables uniformly distributed within a preset numerical range; the function model of the customer retention status prediction unit in the historical analysis period conforms to a Bernoulli binomial distribution; and the function model of the customer recommendation new customer number prediction unit in the historical analysis period conforms to a Poisson distribution.

9. A computer device, characterized in that, The computer device includes: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the customer number prediction method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the customer number prediction method as described in any one of claims 1-7.

11. A computer program product comprising a computer program that, when executed by a processor, implements the customer number prediction method as described in any one of claims 1-7.

Images