A marketing decision-making method and system based on market dynamics

Abstract

The present invention relates to a marketing decision method and system based on market dynamics, the method comprising the following steps: collecting historical market data and constructing a historical market data set after preprocessing; constructing a market prediction model, inputting the historical market data set into the market prediction model for training, and the model outputs market prediction data; constructing a marketing decision optimization model, and outputting a marketing decision plan according to the market prediction data; the present invention combines market dynamics prediction and multi-period resource allocation optimization, and can quickly capture consumer behavior, competition dynamics and seasonal changes through the market dynamics prediction model, thereby achieving better adaptation to market complexity and uncertainty; introducing a long-term cumulative profit target, taking into account current and future profits, and supporting enterprises to make forward-looking strategic decisions.

Description

Technical Field

[0001] The invention relates to a marketing decision method and system based on market dynamics, belonging to the technical field of marketing. Background Art

[0002] In the current market competition environment, corporate marketing decisions often need to be adjusted based on a large amount of real-time data (such as market demand, competitor dynamics, and consumer behavior). However, due to the complexity and uncertainty of market changes, traditional marketing decision-making methods have obvious shortcomings, resulting in inefficient resource allocation and delayed market response.

[0003] In existing technologies, static models rely on fixed historical data or market status in a single period of time to formulate marketing strategies, ignoring the real-time changes and nonlinear characteristics of the market, and are unable to cope with rapidly changing consumer demand or competitive landscape; artificial intelligence technology is mostly centered on single-objective optimization (such as maximizing profits or minimizing costs), ignoring the complexity of dynamic market environments (such as multi-channel synergy effects). Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention proposes a marketing decision-making method and system based on market dynamics.

[0005] The technical solution of the present invention is as follows:

[0006] In one aspect, the present invention provides a marketing decision-making method based on market dynamics, comprising the following steps:

[0007] Collect historical market data and construct historical market data sets after preprocessing;

[0008] Build a market prediction model, input historical market data sets into the market prediction model for training, and the model outputs market prediction data;

[0009] Build a marketing decision optimization model and output marketing decision plans based on market forecast data.

[0010] As a preferred embodiment of the present invention, the market forecasting model is constructed based on a long short-term memory network, as shown in the following formula:

[0011] Y t+1 =g(X t ,φ)+λ·sin(ωt)+η

[0012]

[0013] Where: Y t+1 represents the prediction result of the market prediction model t+1; X trepresents the sample at time t in the historical market data set; φ represents the environmental parameter; λ represents the fluctuation amplitude; ω represents the frequency; η represents the random noise; M represents the total number of features of the sample; x m represents the mth feature of the sample; γ and δ represent nonlinear adjustment factors respectively; β m Represents the mth feature weight of the sample.

[0014] As a preferred embodiment of the present invention, the marketing decision optimization model is an objective function with the goal of maximizing revenue. Under a preset budget, an optimal resource allocation decision is made. The objective function is specifically shown in the following formula:

[0015] max R(S,Y t+1 , Φ, Θ)

[0016] Where: R represents the profit function; S represents the resource allocation vector; Φ represents the external interference parameter; Θ represents the endogenous parameter of the model;

[0017] The constraint adjustment of the objective function is specifically shown in the following formula:

[0018]

[0019] P(Y t+1 |X t ,φ)>α

[0020] Where: K represents the total number of samples of resource allocation vector; s i represents the kth sample in the resource allocation vector; P(Y t+1 |X t ,φ) represents the confidence of the predicted market data; α represents the confidence threshold; B represents the preset budget.

[0021] As a preferred embodiment of the present invention, the profit function R is specifically shown as follows:

[0022]

[0023] Where: u k 、v k They represent the nonlinear response coefficient of the kth sample in the resource allocation vector; σ k represents the revenue efficiency of the kth sample in the resource allocation vector; θ k represents the cost penalty factor of the kth sample in the resource allocation vector; e -τ It represents the impact rate of external environment on profit; τ represents the impact coefficient.

[0024] As a preferred implementation mode of the present invention, the optimization resource allocation decision is constructed by taking into account the maximization of long-term cumulative benefits as the objective function, which is specifically shown in the following formula:

[0025]

[0026] Where: J(S,t) represents the total benefit from the current time t to the future time T; μ t represents the discount factor at time t; R t (S,Y t+1 ,Φ,Θ) represents the profit at time t;

[0027] The optimal resource allocation vector S is solved by gradient update, as shown in the following formula:

[0028]

[0029]

[0030] Where: S t+1 represents the resource allocation vector at time t+1; S t represents the resource allocation vector at time t. ∈ represents the learning rate.

[0031] On the other hand, the present invention also provides a marketing decision system based on market dynamics, including a data collection module, a market forecasting module and an impact decision optimization module;

[0032] The data collection module is used to collect historical market data and construct a historical market data set after preprocessing;

[0033] The market prediction module is used to build a market prediction model, input the historical market data set into the market prediction model for training, and the model outputs market prediction data;

[0034] The impact decision optimization module is used to build a marketing decision optimization model and output a marketing decision plan based on market forecast data.

[0035] As a preferred embodiment of the present invention, the market forecasting model is constructed based on a long short-term memory network, as shown in the following formula:

[0036] Y t+1 =g(X t ,φ)+λ·sin(ωt)+η

[0037]

[0038] Where: Y t+1 represents the prediction result of the market prediction model t+1; X t represents the sample at time t in the historical market data set; φ represents the environmental parameter; λ represents the fluctuation amplitude; ω represents the frequency; η represents the random noise; M represents the total number of features of the sample; x mrepresents the mth feature of the sample; γ and δ represent nonlinear adjustment factors respectively; β m Represents the mth feature weight of the sample.

[0039] As a preferred embodiment of the present invention, the marketing decision optimization model is an objective function with the goal of maximizing revenue. Under a preset budget, an optimal resource allocation decision is made. The objective function is specifically shown in the following formula:

[0040] maxR(S,Y t+1 ,Φ,Θ)

[0041] Where: R represents the profit function; S represents the resource allocation vector; Φ represents the external interference parameter; Θ represents the endogenous parameter of the model;

[0042] The constraint adjustment of the objective function is specifically shown in the following formula:

[0043]

[0044] P(Y t+1 |X t ,φ)>α

[0045] Where: K represents the total number of samples of resource allocation vector; s i represents the kth sample in the resource allocation vector; P(Y t+1 |X t ,φ) represents the confidence of the predicted market data; α represents the confidence threshold; B represents the preset budget.

[0046] On the other hand, the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any embodiment of the present invention is implemented.

[0047] In yet another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any embodiment of the present invention.

[0048] The present invention has the following beneficial effects:

[0049] 1. The present invention combines market dynamics prediction and multi-period resource allocation optimization. The market dynamics prediction model can quickly capture consumer behavior, competition dynamics and seasonal changes, thereby achieving better adaptation to market complexity and uncertainty. It introduces long-term cumulative profit goals, takes into account current and future profits, and supports enterprises in making forward-looking strategic decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0053] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0054] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0055] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0056] Embodiment 1:

[0057] See also Figure 1 , a marketing decision-making method based on market dynamics, comprising the following steps:

[0058] Collect historical market data and construct a historical market data set after preprocessing, wherein the historical market data includes social media sentiment, search volume, sales trend, market growth rate, and historical sales, etc.;

[0059] Build a market prediction model, input historical market data sets into the market prediction model for training, and the model outputs market prediction data;

[0060] Build a marketing decision optimization model and output marketing decision plans based on market forecast data.

[0061] As a preferred implementation of this embodiment, the market prediction model is constructed based on a long short-term memory network, as shown in the following formula:

[0062] Y t+1 =g(X t,φ)+λ·sin(ωt)+η

[0063]

[0064] Where: Y t+1 represents the prediction result of the market prediction model t+1; X t represents the sample at time t in the historical market data set; φ represents the environmental parameter; λ represents the fluctuation amplitude; ω represents the frequency; η represents the random noise; M represents the total number of features of the sample; x m represents the mth feature of the sample; γ and δ represent nonlinear adjustment factors respectively; β m represents the mth feature weight of the sample; g(X t ,φ) through X t And φ describes the main market trend; λ·sin(ωt) is a periodic fluctuation term that simulates seasonal or cyclical demand.

[0065] As a preferred implementation of this embodiment, the marketing decision optimization model is an objective function with the goal of maximizing revenue. Under a preset budget, an optimal resource allocation decision is made. The objective function is specifically shown in the following formula:

[0066] maxR(S,Y t+1 ,Φ,Θ)

[0067] Where: R represents the revenue function; S represents the resource allocation vector, which is the allocation of marketing resources in different channels (such as advertising investment, activity cost), and each sample in the vector represents the investment amount of different channels; Φ represents external interference parameters (such as macroeconomic factors, competitive product dynamics); Θ represents the endogenous parameters of the model (such as pricing, advertising effect coefficient);

[0068] The constraint adjustment of the objective function is specifically shown in the following formula:

[0069]

[0070] P(Y t+1 |X t ,φ)>α

[0071] Where: K represents the total number of samples of resource allocation vector; s i represents the kth sample in the resource allocation vector; P(Y t+1 |X t ,φ) represents the confidence of the predicted market data; α represents the confidence threshold; B represents the preset budget.

[0072] As a preferred implementation of this embodiment, the profit function R is specifically shown in the following formula:

[0073]

[0074] Where: u k 、v k They represent the nonlinear response coefficient of the kth sample in the resource allocation vector; σ k represents the revenue efficiency of the kth sample in the resource allocation vector; θ k represents the cost penalty factor of the kth sample in the resource allocation vector, which is used to constrain the decreasing marginal effect of high investment; e -τ It represents the impact rate of external environment on profit; τ represents the impact coefficient.

[0075] As a preferred implementation of this embodiment, the optimization resource allocation decision is constructed by taking into account the maximization of long-term cumulative benefits as the objective function. The objective function is used to optimize resource allocation at multiple time points to maximize the cumulative benefits. Specifically, it is shown in the following formula:

[0076]

[0077] Where: J(S,t) represents the total benefit from the current time t to the future time T; μ t Represents the discount factor at time t, which is used to reduce the weight of future earnings, reflecting that "current earnings are more important"; R t (S,Y t+1 ,Φ,Θ) represents the profit at time t;

[0078] The optimal resource allocation vector S is solved by gradient update, as shown in the following formula:

[0079]

[0080] Where: S t+1 represents the resource allocation vector at time t+1; S t Represents the resource allocation vector at time t. ∈ represents the learning rate, which determines the magnitude of each adjustment.

[0081] Embodiment 2:

[0082] A marketing decision-making system based on market dynamics, including a data collection module, a market forecasting module and an influencing decision optimization module;

[0083] The data collection module is used to collect historical market data and construct a historical market data set after preprocessing;

[0084] The market prediction module is used to build a market prediction model, input the historical market data set into the market prediction model for training, and the model outputs market prediction data;

[0085] The impact decision optimization module is used to build a marketing decision optimization model and output a marketing decision plan based on market forecast data.

[0086] This system is used to implement the functions in the first embodiment, which will not be described in detail here.

[0087] Embodiment three:

[0088] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any embodiment of the present invention when executing the program.

[0089] Embodiment 4:

[0090] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any embodiment of the present invention is implemented.

[0091] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0092] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0093] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0094] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0095] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A marketing decision-making method based on market dynamics, characterized in that: The following steps are involved: Collect historical market data and construct historical market data sets after preprocessing; Build a market prediction model, input historical market data sets into the market prediction model for training, and the model outputs market prediction data; Build a marketing decision optimization model and output marketing decision plans based on market forecast data.

2. A marketing decision-making method based on market dynamics according to claim 1, characterized in that: The market prediction model is constructed based on the long short-term memory network, as shown in the following formula: Y t+1 =g(X t ,φ)+λ·sin(ωt)+η Where: Y t+1 represents the prediction result of the market prediction model t+1; X t represents the sample at time t in the historical market data set; φ represents the environmental parameter; λ represents the fluctuation amplitude; ω represents the frequency; η represents the random noise; M represents the total number of features of the sample; x m represents the mth feature of the sample; γ and δ represent nonlinear adjustment factors respectively; β m Represents the mth feature weight of the sample.

3. A marketing decision-making method based on market dynamics according to claim 2, characterized in that: The marketing decision optimization model is an objective function with the goal of maximizing revenue. Under a preset budget, an optimal resource allocation decision is made. The objective function is specifically shown in the following formula: maxR(S,Y t+1 ,Φ,Θ) Where: R represents the profit function; S represents the resource allocation vector; Φ represents the external interference parameter; Θ represents the endogenous parameter of the model; The constraint adjustment of the objective function is specifically shown in the following formula: P(Y t+1 |X t ,φ)>a Where: K represents the total number of samples of resource allocation vector; s i represents the kth sample in the resource allocation vector; P(Y t+1 |X t ,φ) represents the confidence of the predicted market data; α represents the confidence threshold; B represents the preset budget.

4. A marketing decision-making method based on market dynamics according to claim 3, characterized in that: The profit function R is specifically shown in the following formula: Where: u k 、v k They represent the nonlinear response coefficient of the kth sample in the resource allocation vector; σ k represents the revenue efficiency of the kth sample in the resource allocation vector; θ k represents the cost penalty factor of the kth sample in the resource allocation vector; e -τ It represents the impact rate of external environment on profit; τ represents the impact coefficient.

5. A marketing decision-making method based on market dynamics according to claim 4, characterized in that: The optimization resource allocation decision is constructed by taking into account the maximization of long-term cumulative benefits as the objective function, which is specifically shown in the following formula: Where: J(S,t) represents the total benefit from the current time t to the future time T; μ t represents the discount factor at time t; R t (S,Y t+1 ,Φ,Θ) represents the profit at time t; The optimal resource allocation vector S is solved by gradient update, as shown in the following formula: Where: S t+1 represents the resource allocation vector at time t+1; S t represents the resource allocation vector at time t. ∈ represents the learning rate.

6. A marketing decision system based on market dynamics, characterized in that: It includes data collection module, market forecasting module and decision-making optimization module; The data collection module is used to collect historical market data and construct a historical market data set after preprocessing; The market prediction module is used to build a market prediction model, input the historical market data set into the market prediction model for training, and the model outputs market prediction data; The impact decision optimization module is used to build a marketing decision optimization model and output a marketing decision plan based on market forecast data.

7. A marketing decision-making method based on market dynamics according to claim 6, characterized in that: The market prediction model is constructed based on the long short-term memory network, as shown in the following formula: Y t+1 =g(X t ,φ)+λ·sin(ωt)+η Where: Y t+1 represents the prediction result of the market prediction model t+1; X t represents the sample at time t in the historical market data set; φ represents the environmental parameter; λ represents the fluctuation amplitude; ω represents the frequency; η represents the random noise; M represents the total number of features of the sample; x m represents the mth feature of the sample; γ and δ represent nonlinear adjustment factors respectively; β m Represents the mth feature weight of the sample.

8. A marketing decision-making method based on market dynamics according to claim 7, characterized in that: The marketing decision optimization model is an objective function with the goal of maximizing revenue. Under a preset budget, an optimal resource allocation decision is made. The objective function is specifically shown in the following formula: maxR(S,Y t+1 ,Φ,Θ) Where: R represents the profit function; S represents the resource allocation vector; Φ represents the external interference parameter; Θ represents the endogenous parameter of the model; The constraint adjustment of the objective function is specifically shown in the following formula: P(Y t+1 |X t ,φ)>a Where: K represents the total number of samples of resource allocation vector; s i represents the kth sample in the resource allocation vector; P(Y t+1 |X t ,φ) represents the confidence of the predicted market data; α represents the confidence threshold; B represents the preset budget.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images