A marketing data analysis and processing method based on big data

Abstract

The present invention provides a marketing data analysis and processing method based on big data, and relates to the technical field of marketing data analysis. The present invention collects daily sales data, market search popularity, economic confidence of consumers in each region, population size and per capita income training models, obtains predicted sales for the next three months through the model, obtains optimized predicted sales according to the change factors of the first month, and then divides the marketing investment amount according to volatility analysis and market potential evaluation, adjusts the marketing investment amount of the second month of the second time through the marketing investment amount of the first third month, improves the flexibility of marketing planning according to the optimized predicted sales, and improves the utilization efficiency of marketing investment according to the adjustment of marketing investment amount.

Description

Technical Field

[0001] The present invention relates to the technical field of marketing data analysis, and in particular to a marketing data analysis and processing method based on big data. Background Art

[0002] In the modern business environment, marketing has become an important way for enterprises to increase market share and expand brand influence, but the problem that follows is how to optimize marketing strategies through scientific methods to achieve efficient resource utilization. Traditional marketing strategies often rely on experience judgment or static market data analysis, which makes it difficult to accurately predict future market changes and lack a comprehensive evaluation of marketing effectiveness. In addition, due to the strong market dynamics, the economic conditions, consumer preferences and competitive environment in different regions vary greatly, and the uniformity of marketing strategies is difficult to meet regional needs. In this context, how to combine big data technology and use accurate prediction and optimization models to improve the effectiveness of marketing has become a technical problem that the industry needs to solve urgently.

[0003] In the prior art, the publication number CN118747657A discloses a real-time dynamic marketing decision method and system based on big data, which includes the following steps: the first step: data collection, the second step: data preprocessing, the third step: feature extraction, the fourth step: real-time analysis, the fifth step: decision generation, the sixth step: strategy execution, and the seventh step: effect evaluation. In the present invention, the system generates dynamic marketing strategies and decisions by analyzing market trends and consumer behaviors in real time.

[0004] Although the above-mentioned disclosed technology realizes marketing decision-making, its process is not sophisticated enough and only stays at the level of directly obtaining results, without considering the impact of real-time market changes on marketing investment.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a marketing data analysis and processing method based on big data to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: A marketing data analysis and processing method based on big data, the specific steps include: Step 1: Collect daily sales data, market search popularity, consumer economic confidence, population size and per capita income of each region to form a historical data set, where sales data includes sales volume, extract consumer purchasing power index features and marketing delivery effect index features and add them to the historical data set, input the historical data set into the time training model for training, use sales volume as a label, and obtain the sales prediction model; Step 2: Use the sales forecast model to obtain the daily forecast sales of each region in the next three months, and then continue to collect daily sales data, market search popularity, economic confidence, population size and per capita income of each region in the first month, obtain the change factors of each region between the first month and the previous month, and obtain the forecast correction coefficient through the change factors. Apply the correction coefficient to the forecast sales of the second and third months to obtain the daily optimized forecast sales of each region in the second and third months; Step 3: Perform volatility analysis and market potential assessment on the optimized forecast sales volume in the second and third months, generate investment scores for each region based on the volatility value and market potential, obtain the marketing investment quota for each region in the second and third months based on the investment score and the total marketing investment budget, and determine whether to issue the quota based on the increase rate of the marketing investment quota in the first month; Step 4: Obtain the predicted sales volume through the sales forecast model every month, and adjust the marketing investment amount for the second month of the second period based on the marketing investment amount for the third month of the first period.

[0008] Furthermore, the sales data of each region is obtained through the enterprise sales management system, the sales data including sales volume, marketing investment amount, average selling price and sales volume increase due to marketing, the each region is a sales region divided by the merchant according to the sales system, the market search popularity is obtained through the third-party search engine data, and the consumer economic confidence, population size and per capita income of each region are obtained through the economic database. The sales data and market search popularity are based on a daily cycle, so as to obtain the daily sales data and market search popularity of each region, and according to the statistical cycle of economic confidence, population size and per capita income, the economic confidence, population size and per capita income within the statistical period are divided according to the number of days to form the daily economic confidence, population size and per capita income of each region, and the daily sales data, market search popularity, economic confidence, population size and per capita income of each region are aggregated into a historical data set.

[0009] Furthermore, the historical data set is normalized, and features are extracted from the historical data set to construct the consumer purchasing power index features. The formula is as follows: in, For the Cycle No. Purchasing power index of the region, For the Cycle No. The population of the region, For the Cycle No. The per capita income of the region, is the period index variable, Retrieve variable for region number, , , , ; The characteristics of marketing delivery effect indicators are constructed based on the following formula: in, For the Cycle No. Marketing effectiveness indicators for each region, For the Cycle No. The sales volume increase in each region due to marketing, For the Cycle No. The marketing investment amount for each region, is the period index variable, Retrieve variable for region number, , , , ; The consumer purchasing power index characteristics and marketing delivery effect indicator characteristics are added to the historical data set. The historical data set is input into the time series model. Sales volume is used as the training label to obtain the daily predicted sales volume for each region in the next three months. The trained model is punctuated as the sales volume prediction model.

[0010] Furthermore, daily sales data, market search heat, economic confidence, population size and per capita income of each region in the past time period and the current time are obtained, and the historical data set is mapped. The historical data set is input into the sales forecast model to obtain the daily forecast sales of each region in the next three months, and the actual daily sales data, market search heat, economic confidence, population size and per capita income in each region in the first month are continuously collected. The change factors between the first month and the previous month are obtained for each region respectively. The change factors are the change rate of marketing investment amount, the change rate of average selling price, the change rate of sales increment due to marketing, the change rate of market search heat, the change rate of consumer economic confidence, the change rate of population size and the change rate of per capita income between the first month and the previous month. The change factors with the same sign as the mean error between the forecast sales and actual sales in the first month are determined. The same sign means that the change factor and the mean error between the forecast sales and actual sales in the first month are both greater than 0 or less than 0, and the forecast correction coefficient is obtained. The formula based on it is as follows: in, is the prediction correction coefficient, and its specific value is Same number, For the The error between the predicted sales volume in the first month and the actual sales volume in the second month has the same sign as the change factor. is the average rate of change between the predicted sales and the actual sales in the first month, is the total number of changing factors with the same sign as the mean error between the predicted sales and the actual sales in the first month, Index variables for changing factors, , ; Add 1 to the forecast correction coefficient of each region and multiply it by the forecast sales of each region in the second and third months. The result is the optimized forecast sales of each region on a daily basis in the second and third months. If the forecast correction coefficient is 0, retrain the model.

[0011] Furthermore, the volatility analysis of the optimized predicted sales volume of each region in the second and third months is performed to obtain the daily optimized predicted sales volume of each region in the second and third months, and the volatility of the daily optimized predicted sales volume of each region in the second and third months is obtained respectively, and the formula is as follows: in, For the number The volatility value of the region, For the number Region No. Optimized sales forecast for the day, For the number The average sales volume of regional optimization forecast, Retrieve variable for region number, is the number of days in this month, is the day index variable, , ; Evaluate the market potential of each region, obtain the average sales volume of the last three months through the historical data set, and obtain the market potential of each region based on the following formula: in, For Region market potential, For Region Optimized sales forecast, For Region Average sales volume in the last three months, Retrieve variables for regions, .

[0012] Furthermore, based on the volatility analysis and market potential assessment of each region, an investment score for each region is formed, based on the following formula: in, For Region Investment ratings, For Region The volatility of For Region market potential, and is the weight coefficient, is the region index variable, ; The weight coefficient is obtained by fitting the historical data set. The sales data and marketing investment quota data of the last three months in the historical data set are extracted, and the volatility analysis and market potential evaluation are performed on the sales data. The volatility value and market potential are input into the linear regression model, and the marketing investment quota is used as a label to obtain the weight coefficient through fitting. and , the weight coefficient obtained and Substitute this into the investment score formula; The total budget for marketing investment in the second and third months is obtained, and the marketing investment amount is divided for each region according to the investment score to form the marketing investment amount for each region in the second and third months. The formula is as follows: in, For Region No. Monthly marketing investment amount, For Region No. Monthly investment rating, It is Total budget for monthly marketing investment, is the total number of regions, is the region index variable, , , Retrieve a variable for the month, , ; The marketing investment quota obtained for each region in the second month will be compared with the marketing investment quota in the first month. For regions where the absolute value of the increase rate is less than 5%, no marketing investment quota will be actually issued. Instead, it will be issued according to the marketing investment quota of the first month. The part of the marketing investment quota that exceeds that of the first month will be marked as marketing investment reserve funds. For regions where the absolute value of the increase rate is greater than 5%, the marketing investment quota will be issued according to the marketing investment quota already obtained.

[0013] Furthermore, taking a month as a cycle, at the end of each month, the sales forecast model is used to obtain the daily forecast sales of each region for the next three months, and the marketing investment quotas for the second and third months are obtained. Among the marketing investment quotas for the second and third months obtained in the adjacent two-month cycles, the marketing investment quotas for the second and third months obtained in the previous month are marked as the marketing investment quotas for the first second and third months, and the marketing investment quotas for the second and third months obtained in the next month are marked as the marketing investment quotas for the second and third months. When the marketing investment quota for the second month is obtained, the marketing investment quota for the first second month has been issued, and the marketing investment quota for the first third month has not yet been issued, and is aligned in time with the marketing investment quota for the second month of the second month.

[0014] Furthermore, the marketing investment amount in the second month of the second time is compared with the marketing investment amount in the third month of the first time to determine the change rate of the marketing investment amount. The formula is as follows: in, is the change rate of marketing investment amount, This is the marketing investment amount for the first three months. The marketing investment amount for the second month of the second session; For areas where the absolute value of the change rate of marketing investment quota is less than 5%, the marketing investment quota of the second month of the second period will be replaced by the marketing investment quota of the third month of the first period; for areas where the change rate of marketing investment quota is negative and the absolute value is greater than or equal to 5%, the marketing investment quota of the second month of the second period will be replaced by the average of the marketing investment quotas of the second month of the second period and the marketing investment quotas of the third month of the first period; for areas where the change rate of marketing investment quota is greater than or equal to 5%, the marketing investment quota of the second month of the second period shall prevail.

[0015] Furthermore, when the second month's marketing investment quota is first issued, if there is no first third month's marketing investment quota data to determine the marketing investment quota change rate, the marketing investment quota change rate will not be determined, and step 6 will be directly followed.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention obtains predicted sales by acquiring a sales forecast model, and dynamically optimizes the predicted sales according to the change factors of the first month, so that the predicted sales can be adjusted in time according to real-time market changes, thereby improving the flexibility of marketing planning, and optimizing the volatility analysis and market potential evaluation of the predicted sales, further realizing the scientific allocation of marketing investment quotas in different regions, and dynamically adjusting the marketing investment quotas actually issued according to the marketing investment quotas obtained in adjacent periods, thereby improving the utilization efficiency of marketing investment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION

[0018] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0019] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0020] Example

[0021] See also Figure 1 , the present invention provides a technical solution: A marketing data analysis and processing method based on big data, the specific steps include: Step 1: Collect daily sales data, market search popularity, consumer economic confidence, population size and per capita income of each region to form a historical data set, where sales data includes sales volume, extract consumer purchasing power index features and marketing delivery effect index features and add them to the historical data set, input the historical data set into the time training model for training, use sales volume as a label, and obtain the sales prediction model; The step 1 includes the following contents: Step 101: Obtain sales data of each region through the enterprise sales management system, the sales data including sales volume, marketing investment amount, average selling price and sales volume increment due to marketing, the each region is the sales region divided by the merchant according to the sales system, obtain market search heat through third-party search engine data, the market search heat is the daily search volume of all sales platforms where the product is listed, the sum of the search volumes of all sales platforms on the day is calibrated as the market search heat of the day, obtain the search volume on each platform from each sales platform, obtain the economic confidence, population size and per capita income of consumers in each region through the economic database, the sales data and market search heat are based on a day as a cycle, so as to obtain the daily sales data and market search heat of each region, according to the statistical cycle of economic confidence, population size and per capita income, divide the economic confidence, population size and per capita income within the statistical cycle by the number of days, form the daily economic confidence, population size and per capita income of each region, and summarize the daily sales data, market search heat, economic confidence, population size and per capita income of each region into a historical data set.

[0022] Step 102: Normalize the historical data set, extract features from the historical data set, and construct the consumer purchasing power index features. The formula is as follows: in, For the Cycle No. Purchasing power index of the region, For the Cycle No. The population of the region, For the Cycle No. The per capita income of the region, is the period index variable, Retrieve variable for region number, , , , ; The characteristics of marketing delivery effect indicators are constructed based on the following formula: in, For the Cycle No. Marketing effectiveness indicators for each region, For the Cycle No. The sales volume of each region increased due to marketing. For the Cycle No. The marketing investment amount for each region, is the period index variable, Retrieve variable for region number, , , , ; The characteristics of the consumer purchasing power index can reflect the local area's acceptance and willingness for the target product. Extracting it plays an important role in predicting sales. The characteristics of the marketing delivery effectiveness index can reflect the effectiveness of marketing activities in the region. Extracting it can reflect the potential relationship between marketing and sales.

[0023] The consumer purchasing power index characteristics and marketing delivery effect indicator characteristics are added to the historical data set. The historical data set is input into the time series model. Sales volume is used as the training label to obtain the daily predicted sales volume for each region in the next three months. The trained model is punctuated as the sales volume prediction model.

[0024] Step 2: Use the sales forecast model to obtain the daily forecast sales of each region in the next three months, and then continue to collect daily sales data, market search popularity, economic confidence, population size and per capita income of each region in the first month, obtain the change factors of each region between the first month and the previous month, and obtain the forecast correction coefficient through the change factors. Apply the correction coefficient to the forecast sales of the second and third months to obtain the daily optimized forecast sales of each region in the second and third months; The step 2 includes the following contents: Obtain daily sales data, market search heat, economic confidence, population size and per capita income of each region in the past time period and the current time, map the historical data set, input the historical data set into the sales forecast model to obtain the daily forecast sales of each region in the next three months, continue to collect the actual daily sales data, market search heat, economic confidence, population size and per capita income in each region in the first month, obtain the change factors of each region between the first month and the previous month, respectively, the change factors are the change rate of marketing investment amount, the change rate of average selling price, the change rate of sales increment due to marketing, the change rate of market search heat, the change rate of consumer economic confidence, the change rate of population size and the change rate of per capita income, judge the change factors with the same sign as the mean error between the forecast sales and actual sales of the first month, the same sign means that the change factor and the mean error between the forecast sales and actual sales of the first month are both greater than 0 or less than 0, and obtain the forecast correction coefficient, based on the following formula: in, is the prediction correction coefficient, and its specific value is Same number, For the The error between the predicted sales volume in the first month and the actual sales volume in the second month has the same sign as the change factor. is the average rate of change between the predicted sales and the actual sales in the first month, is the total number of changing factors with the same sign as the mean error between the predicted sales and the actual sales in the first month, Index variables for changing factors, , ; When there is a difference between the predicted sales and the actual sales, there must be variable factors involved. Find the variable factors that are consistent with the sales difference and compare their averages with the average rate of change between the predicted sales and actual sales in the first month. Use this as a direction to optimize the predicted sales for the second and third months.

[0025] Add 1 to the forecast correction coefficient of each region and multiply it by the forecast sales of each region in the second and third months. The result is the optimized forecast sales of each region on a daily basis in the second and third months. If the forecast correction coefficient is 0, retrain the model.

[0026] Since the sales forecast obtained by the sales forecast model is not necessarily accurate, the actual sales results of the first month can reveal this change. By linking the change analysis to specific change factors, we can more accurately determine the reasons for the difference between the predicted sales and the actual sales. Based on the specific change factors, we can optimize the sales forecasts for the second and third months to obtain more accurate sales forecasts. Step 3: Perform volatility analysis and market potential assessment on the optimized forecast sales volume in the second and third months, generate investment scores for each region based on the volatility value and market potential, obtain the marketing investment quota for each region in the second and third months based on the investment score and the total marketing investment budget, and determine whether to issue the quota based on the increase rate of the marketing investment quota in the first month; The step 3 includes the following contents: Step 301: Perform volatility analysis on the optimized predicted sales volume of each region in the second month and the third month, obtain the optimized predicted sales volume of each region in the second month and the third month, and obtain the volatility of the optimized predicted sales volume of each region in the second month and the third month, respectively, according to the following formula: in, For the number The volatility value of the region, For the number Region No. Optimized sales forecast for the day, For the number The average sales volume of regional optimization forecast, Retrieve variable for region number, is the number of days in this month, is the day index variable, , ; The volatility of the optimized sales forecast is reflected in the form of mean square error.

[0027] Evaluate the market potential of each region, obtain the average sales volume of the last three months through the historical data set, and obtain the market potential of each region based on the following formula: in, For Region market potential, For Region Optimized sales forecast, For Region Average sales volume in the last three months, Retrieve variables for regions, .

[0028] Step 302: Based on the volatility analysis and market potential assessment of each region, the investment score of each region is formed according to the following formula: in, For Region Investment ratings, For Region The volatility of For Region market potential, and is the weight coefficient, is the region index variable, ; The weight coefficient is obtained by fitting the historical data set. The sales data and marketing investment quota data of the last three months in the historical data set are extracted, and the volatility analysis and market potential evaluation are performed on the sales data. The volatility value and market potential are input into the linear regression model, and the marketing investment quota is used as a label to obtain the weight coefficient through fitting. and , the weight coefficient obtained and Substitute this into the investment score formula; The actual sales situation in each region may be different. By obtaining the weight coefficient of the corresponding region through the historical data of different regions, we can obtain a market potential assessment result that is more in line with the region.

[0029] The total budget for marketing investment in the second and third months is obtained, and the marketing investment amount is divided for each region according to the investment score to form the marketing investment amount for each region in the second and third months. The formula is as follows: in, For Region No. Monthly marketing investment amount, For Region No. Monthly investment rating, It is Total budget for monthly marketing investment, is the total number of regions, is the region index variable, , , Retrieve a variable for the month, , ; According to the different score ratios, the marketing investment quotas for different regions can be divided to adapt to local conditions.

[0030] The marketing investment quota obtained for each region in the second month will be compared with the marketing investment quota in the first month. For regions where the absolute value of the increase rate is less than 5%, no marketing investment quota will be actually issued. Instead, it will be issued according to the marketing investment quota of the first month. The part of the marketing investment quota that exceeds that of the first month will be marked as marketing investment reserve funds. For regions where the absolute value of the increase rate is greater than 5%, the marketing investment quota will be issued according to the marketing investment quota already obtained.

[0031] The impact of marketing investment amount on sales volume is not intuitive. A slight change in marketing investment amount will not drastically affect sales volume. However, for places with large changes in marketing investment amount, it means that sales volume in this area may change drastically. For places where sales volume may increase significantly, increasing marketing investment can achieve greater sales volume.

[0032] Step 4: Obtain the predicted sales volume through the sales forecast model every month, and adjust the marketing investment amount for the second month of the second period based on the marketing investment amount for the third month of the first period.

[0033] The step 4 includes the following contents: Taking a month as a cycle, at the end of each month, the sales forecast model is used to obtain the daily forecast sales of each region for the next three months, and the marketing investment quotas for the second and third months are obtained. Among the marketing investment quotas for the second and third months obtained in the adjacent two-month cycles, the marketing investment quotas for the second and third months obtained in the previous month are marked as the marketing investment quotas for the first second and third months, and the marketing investment quotas for the second and third months obtained in the next month are marked as the marketing investment quotas for the second and third months of the second time. When the marketing investment quota for the second month of the second time is obtained, the marketing investment quota for the first second month has been issued, and the marketing investment quota for the first third month has not yet been issued, and is aligned in time with the marketing investment quota for the second month of the second time.

[0034] By forecasting sales every month, we can obtain more accurate sales forecasts based on market changes in real time, thereby adjusting marketing strategies more flexibly.

[0035] Furthermore, the marketing investment amount in the second month of the second time is compared with the marketing investment amount in the third month of the first time to determine the change rate of the marketing investment amount. The formula is as follows: in, is the change rate of marketing investment amount, This is the marketing investment amount for the first three months. The marketing investment amount for the second month of the second session; For areas where the absolute value of the change rate of marketing investment quota is less than 5%, the marketing investment quota of the second month of the second period will be replaced by the marketing investment quota of the third month of the first period; for areas where the change rate of marketing investment quota is negative and the absolute value is greater than or equal to 5%, the marketing investment quota of the second month of the second period will be replaced by the average of the marketing investment quotas of the second month of the second period and the marketing investment quotas of the third month of the first period; for areas where the change rate of marketing investment quota is greater than or equal to 5%, the marketing investment quota of the second month of the second period shall prevail.

[0036] The rate of change of marketing spending has changed significantly in the decreasing direction, which means that the sales volume in this area may experience drastic fluctuations, but a certain sales volume and market share must still be guaranteed, so special adjustments to the marketing investment amount are needed to ensure future sales and development.

[0037] When the second month's marketing investment quota is first issued, if there is no first third month's marketing investment quota data to determine the marketing investment quota change rate, the marketing investment quota change rate will not be determined and step 6 will be directly followed.

[0038] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0039] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0040] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0041] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A marketing data analysis and processing method based on big data, characterized in that: The specific steps include: Step 1: Collect daily sales data, market search popularity, consumer economic confidence, population size and per capita income of each region to form a historical data set, where sales data includes sales volume. Extract consumer purchasing power index features and marketing delivery effect indicator features and add them to the historical data set. Input the historical data set into the time training model for training, use sales volume as a label, and obtain the sales prediction model. Step 2: Use the sales forecast model to obtain the daily forecast sales of each region in the next three months, and then continue to collect daily sales data, market search popularity, economic confidence, population size and per capita income of each region in the first month, obtain the change factors of each region between the first month and the previous month, and obtain the forecast correction coefficient through the change factors. Apply the correction coefficient to the forecast sales of the second and third months to obtain the daily optimized forecast sales of each region in the second and third months; Step 3: Perform volatility analysis and market potential assessment on the optimized forecast sales volume in the second and third months, generate investment scores for each region based on the volatility value and market potential, obtain the marketing investment quota for each region in the second and third months based on the investment score and the total marketing investment budget, and determine whether to issue the quota based on the increase rate of the marketing investment quota in the first month; Step 4: Obtain the predicted sales volume through the sales forecast model every month, and adjust the marketing investment amount for the second month of the second period based on the marketing investment amount for the third month of the first period.

2. The marketing data analysis and processing method based on big data according to claim 1, characterized in that: The sales data of each region is obtained through the enterprise sales management system, and the sales data includes sales volume, marketing investment amount, average selling price and sales volume increase due to marketing. The each region is the sales region divided by the merchant according to the sales system. The market search heat is obtained through the third-party search engine data, and the consumer economic confidence, population size and per capita income of each region are obtained through the economic database. The sales data and market search heat are based on a daily cycle, so as to obtain the daily sales data and market search heat of each region. According to the statistical cycle of economic confidence, population size and per capita income, the economic confidence, population size and per capita income within the statistical cycle are evenly divided according to the number of days to form the daily economic confidence, population size and per capita income of each region. The daily sales data, market search heat, economic confidence, population size and per capita income of each region are summarized into a historical data set.

3. The marketing data analysis and processing method based on big data according to claim 1, characterized in that: Normalize the historical data set, extract features from the historical data set, and construct the consumer purchasing power index features. The formula is as follows: in, For the Cycle No. Purchasing power index of the region, For the Cycle No. The population of the region, For the Cycle No. The per capita income of the region, is the period index variable, Retrieve variable for region number, , , , ; The characteristics of marketing delivery effect indicators are constructed based on the following formula: in, For the Cycle No. Marketing effectiveness indicators for each region, For the Cycle No. The sales volume of each region increased due to marketing. For the Cycle No. The marketing investment amount for each region, is the period index variable, Retrieve variable for region number, , , , ; The consumer purchasing power index characteristics and marketing delivery effect indicator characteristics are added to the historical data set. The historical data set is input into the time series model. Sales volume is used as the training label to obtain the daily predicted sales volume for each region in the next three months. The trained model is punctuated as the sales volume prediction model.

4. The marketing data analysis and processing method based on big data according to claim 1, characterized in that: Obtain daily sales data, market search heat, economic confidence, population size and per capita income of each region in the past time period and the current time, map the historical data set, input the historical data set into the sales forecast model to obtain the daily forecast sales of each region in the next three months, continue to collect the actual daily sales data, market search heat, economic confidence, population size and per capita income in each region in the first month, obtain the change factors of each region between the first month and the previous month, respectively, the change factors are the change rate of marketing investment amount, the change rate of average selling price, the change rate of sales increment due to marketing, the change rate of market search heat, the change rate of consumer economic confidence, the change rate of population size and the change rate of per capita income, judge the change factors with the same sign as the mean error between the forecast sales and actual sales of the first month, the same sign means that the change factor and the mean error between the forecast sales and actual sales of the first month are both greater than 0 or less than 0, and obtain the forecast correction coefficient, based on the following formula: in, is the prediction correction coefficient, and its specific value is Same number, For the The error between the predicted sales volume in the first month and the actual sales volume in the second month has the same sign as the change factor. is the average rate of change between the predicted sales and the actual sales in the first month, is the total number of changing factors with the same sign as the mean error between the predicted sales and the actual sales in the first month, Index variables for changing factors, , ; Add 1 to the forecast correction coefficient of each region and multiply it by the forecast sales of each region in the second and third months. The result is the optimized forecast sales of each region on a daily basis in the second and third months. If the forecast correction coefficient is 0, retrain the model.

5. The marketing data analysis and processing method based on big data according to claim 4 is characterized by: Perform a volatility analysis on the optimized forecast sales volume of each region in the second and third months, obtain the daily optimized forecast sales volume of each region in the second and third months, and obtain the volatility of the daily optimized forecast sales volume of each region in the second and third months, respectively, based on the following formula: in, For the number The volatility value of the region, For the number Region No. Optimized sales forecast for the day, For the number The average sales volume of regional optimization forecast, Retrieve variable for region number, is the number of days in this month, is the day index variable, , ; Evaluate the market potential of each region, obtain the average sales volume of the last three months through the historical data set, and obtain the market potential of each region based on the following formula: in, For Region market potential, For Region Optimized sales forecast, For Region Average sales volume in the last three months, Retrieve variables for regions, .

6. The marketing data analysis and processing method based on big data according to claim 1, characterized in that: Based on the volatility analysis and market potential assessment of each region, the investment score of each region is formed according to the following formula: in, For Region Investment ratings, For Region The volatility of For Region market potential, and is the weight coefficient, is the region index variable, ; The weight coefficient is obtained by fitting the historical data set. The sales data and marketing investment quota data of the last three months in the historical data set are extracted, and the volatility analysis and market potential evaluation are performed on the sales data. The volatility value and market potential are input into the linear regression model, and the marketing investment quota is used as a label to obtain the weight coefficient through fitting. and , the weight coefficient obtained and Substitute this into the investment score formula; The total budget for marketing investment in the second and third months is obtained, and the marketing investment amount is divided for each region according to the investment score to form the marketing investment amount for each region in the second and third months. The formula is as follows: in, For Region No. Monthly marketing investment amount, For Region No. Monthly investment rating, It is Total budget for monthly marketing investment, is the total number of regions, is the region index variable, , , Retrieve a variable for the month, , ; The marketing investment quota obtained for each region in the second month will be compared with the marketing investment quota in the first month. For regions where the absolute value of the increase rate is less than 5%, no marketing investment quota will be actually issued. Instead, it will be issued according to the marketing investment quota of the first month. The part of the marketing investment quota that exceeds that of the first month will be marked as marketing investment reserve funds. For regions where the absolute value of the increase rate is greater than 5%, the marketing investment quota will be issued according to the marketing investment quota already obtained.

7. The marketing data analysis and processing method based on big data according to claim 1, characterized in that: Taking a month as a cycle, at the end of each month, the sales forecast model is used to obtain the daily forecast sales of each region for the next three months, and the marketing investment quotas for the second and third months are obtained. Among the marketing investment quotas for the second and third months obtained in the adjacent two-month cycles, the marketing investment quotas for the second and third months obtained in the previous month are marked as the marketing investment quotas for the first second and third months, and the marketing investment quotas for the second and third months obtained in the next month are marked as the marketing investment quotas for the second and third months of the second time. When the marketing investment quota for the second month of the second time is obtained, the marketing investment quota for the first second month has been issued, and the marketing investment quota for the first third month has not yet been issued, and is aligned in time with the marketing investment quota for the second month of the second time.

8. The marketing data analysis and processing method based on big data according to claim 1, characterized in that: Compare the marketing investment amount in the second month of the second time with the marketing investment amount in the third month of the first time to determine the change rate of the marketing investment amount. The formula is as follows: in, is the change rate of marketing investment amount, This is the marketing investment amount for the first three months. The marketing investment amount for the second month of the second session; For areas where the absolute value of the change rate of marketing investment quota is less than 5%, the marketing investment quota of the second month of the second period will be replaced by the marketing investment quota of the third month of the first period; for areas where the change rate of marketing investment quota is negative and the absolute value is greater than or equal to 5%, the marketing investment quota of the second month of the second period will be replaced by the average of the marketing investment quotas of the second month of the second period and the marketing investment quotas of the third month of the first period; for areas where the change rate of marketing investment quota is greater than or equal to 5%, the marketing investment quota of the second month of the second period shall prevail.

9. The marketing data analysis and processing method based on big data according to claim 1, characterized in that: When the second month's marketing investment quota is first issued, if there is no first third month's marketing investment quota data to determine the marketing investment quota change rate, the marketing investment quota change rate will not be determined and step 6 will be directly followed.

Images