A marketing campaign generation method and device based on sales data analysis

Abstract

A marketing activity generation method and device based on sales data analysis, wherein the method comprises: acquiring sales data of all pharmacies in a target area, disease graphs corresponding to all drugs in the pharmacies, and population health data and environmental climate monitoring data in the target area; establishing a regional health evaluation index set according to the population health data and the environmental climate monitoring data, and calculating a health situation index of the target area based on the regional health evaluation index set; extracting feature information of the first data to obtain a feature data set; performing cluster analysis on the feature data set to generate a category label of each pharmacy; generating a customer portrait according to the category label and the feature data set; and generating a marketing activity scheme corresponding to each category label through a preset model according to the category label, the customer portrait, and inventory data in the sales data. The application can attract target customer groups, thereby improving the return on investment of marketing.

Description

Technical Field

[0001] The present application relates to the field of data analysis, and specifically to a method and device for generating marketing activities based on sales data analysis. Background Art

[0002] With the development of big data and cloud computing technologies, companies can gain a deeper understanding of consumer behavior patterns by analyzing historical data, thereby guiding business decisions. This data analysis method can help companies more accurately grasp market demand and consumer behavior characteristics, thereby formulating more effective marketing strategies. Through this refined data-driven approach, companies not only improve marketing efficiency but also significantly increase consumer satisfaction and loyalty.

[0003] Currently, the pharmaceutical industry employs a variety of methods to attract customers in order to boost sales. For example, some pharmacies regularly run promotions, such as discounts and buy-one-get-one-free promotions. Others utilize membership systems, offering rewards like points and exclusive discounts to long-term customers. These traditional marketing methods can, to a certain extent, boost sales.

[0004] However, existing traditional marketing methods set unified promotional standards based on national average data, which makes marketing activities less targeted and difficult to effectively attract customer groups, thus affecting the conversion rate of marketing investment. Summary of the Invention

[0005] This application provides a method and device for generating marketing activities based on sales data analysis, which are used to attract target customer groups and thus improve the return on marketing investment.

[0006] In a first aspect of the present application, a method for generating a marketing campaign based on sales data analysis is provided, which is applied to a server. The method comprises:

[0007] Obtain sales data of all pharmacies in the target area, disease maps corresponding to all drugs in the pharmacies, and population health data and environmental climate monitoring data in the target area; establish a regional health assessment indicator set based on the population health data and environmental climate monitoring data, and calculate the health status index of the target area based on the regional health assessment indicator set; extract feature information of the first data to obtain a feature data set, which characterizes the sales characteristics, customer needs and health needs of the pharmacy, and the first data includes sales data, disease maps and health status index; perform cluster analysis on the feature data set to generate category labels for each pharmacy, and the category labels include the scale level and target customer group of the pharmacy; generate customer portraits based on the category labels and feature data sets, and the customer portraits include customer behavior preferences, disease demand characteristics and drug purchasing tendencies; generate marketing activity plans corresponding to each category label through a preset model based on the category labels, customer portraits and inventory data in the sales data.

[0008] Optionally, a regional health assessment indicator set is established based on population health data and environmental climate monitoring data, and the health status index of the target area is calculated based on the regional health assessment indicator set, specifically including:

[0009] Based on a preset statistical analysis method, health factors are extracted from population health data and environmental climate monitoring data; the health factors are stratified according to multiple time periods to obtain a health factor set, and the time periods include a first time period, a second time period, and a third time period, wherein the time range of the second time period is greater than the time range of the first time period, and the time range of the third time period is greater than the time range of the second time period; a health assessment indicator set is constructed based on the health factor set, and the health assessment indicator set includes multiple health assessment indicators for each time period; the multiple health assessment indicators for each time period are aggregated and calculated to obtain a health status index in the target area.

[0010] Optionally, a health assessment indicator set is constructed based on the health factor set, specifically including:

[0011] The health factors of each time period are extracted according to the target dimensions to obtain the basic feature vector, which includes the disease risk dimension, the environmental pressure dimension and the population health vulnerability dimension; the basic feature vector is subjected to time series correlation analysis to obtain the time series correlation feature; the basic feature vector is fused with the time series correlation feature to obtain the time series health feature vector, which includes the feature values ​​corresponding to multiple features; the feature values ​​are sorted by importance using a preset importance evaluation algorithm to obtain the importance sorting result; multiple features in the importance sorting result whose importance scores exceed the preset threshold are determined as health assessment indicators to obtain a health assessment indicator set.

[0012] Optionally, perform cluster analysis on the feature dataset to generate category labels for each pharmacy, including:

[0013] Based on the disease map, the disease demand characteristic data of the target area is extracted according to the regional health situation index; the clustering parameters are determined based on the preset clustering evaluation indicators, and the sales data and the disease demand characteristic data are clustered according to the clustering parameters to obtain the clustering results; the cluster number of each pharmacy is determined according to the clustering results; the characteristic distribution statistics of the pharmacies corresponding to the same cluster number in the preset dimensions are counted; and the category label corresponding to each cluster number is determined according to the characteristic distribution statistics.

[0014] Optionally, after determining the cluster number of each pharmacy according to the clustering results, the method further includes: calculating a clustering effectiveness index of the clustering results; if the clustering effectiveness index does not meet a preset condition, re-determining the clustering parameters.

[0015] Optionally, after generating a promotional activity plan for each category corresponding to each category label using a preset model based on the category label, customer profile, and inventory data in the sales data, the method further includes:

[0016] Calculate the similarity of the customer portraits of the first pharmacy and generate a similarity matrix, where the first pharmacy is the pharmacy corresponding to the same category label; calculate the difference value of the customer portraits of the first pharmacy based on the similarity matrix; extract the difference features of the target customer portraits, where the target customer portraits are customer portraits with a customer portrait difference value greater than a preset threshold; divide the first pharmacy into second pharmacies corresponding to multiple subcategory labels based on the difference features; adjust the promotion activity plan based on the sales data and customer portraits of each second pharmacy.

[0017] Optionally, a method for generating a marketing campaign based on sales data analysis, applied to a server, further includes:

[0018] Obtain a digital twin model of the pharmacy corresponding to each category label; based on sales data and customer profiles, use the digital twin model to simulate and evaluate the pharmacy's sales under the promotional plan to obtain simulation evaluation results; and adjust the promotional plan based on the simulation evaluation results.

[0019] In a second aspect of the present application, a marketing activity generation system based on sales data analysis is provided, comprising:

[0020] The acquisition module is used to obtain the sales data of all pharmacies in the target area, the disease maps corresponding to all drugs in the pharmacies, and the population health data and environmental climate monitoring data in the target area;

[0021] A calculation module is used to establish a regional health assessment indicator set based on population health data and environmental climate monitoring data, and calculate the health status index of the target area based on the regional health assessment indicator set;

[0022] an extraction module, configured to extract characteristic information of the first data to obtain a characteristic data set, wherein the characteristic data set represents the sales characteristics, customer needs, and health needs of the pharmacy, the first data including sales data, disease maps, and health status index;

[0023] The analysis module is used to perform cluster analysis on the feature dataset and generate a category label for each pharmacy. The category label includes the pharmacy's scale and target customer group.

[0024] The first generation module is used to generate customer profiles based on category labels and feature datasets. The customer profiles include customer behavior preferences, disease demand characteristics, and drug purchase tendencies.

[0025] The second generation module is used to generate a marketing activity plan corresponding to each category label through a preset model based on the category label, customer portrait and inventory data in the sales data.

[0026] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device performs any of the methods described above.

[0027] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions. When the instructions are executed, any one of the methods described above is executed.

[0028] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0029] 1. This solution integrates sales data, disease maps, population health data, and environmental climate monitoring data to establish a regional health assessment indicator set and calculate a health status index, accurately identifying the health needs of the target area. It also generates pharmacy category labels by extracting feature data sets and performing cluster analysis to clarify pharmacy size and customer groups. It combines category labels to construct customer profiles, providing in-depth insights into customer behavioral preferences, disease needs, and purchasing tendencies. It also integrates inventory data to design marketing campaigns and optimize resource allocation. Through a comprehensive analysis of the entire process, from health needs to customer profiles to inventory dynamics, it matches different marketing campaign plans to different customer groups, enhancing the targeted nature of marketing activities and improving the conversion rate of marketing investment.

[0030] 2. Health factors are extracted based on population health data and environmental climate monitoring data. The dynamic changing trends of health factors in the short, medium and long term are captured through stratified processing of multiple time periods. Combined with time series correlation analysis and feature importance ranking, the ability to mine the complex correlations between health factors is enhanced, ensuring the scientific nature and accuracy of health assessment indicators. Through the calculation of the health status index, the health status and changing trends of the target area can be comprehensively and dynamically reflected, providing efficient and reliable data support for the accurate assessment of regional health needs and the formulation of subsequent marketing strategies.

[0031] 3. By integrating disease demand characteristic data and sales data, it is possible to accurately capture the multidimensional characteristics of pharmacies in terms of disease demand and sales characteristics. Through clustering parameter optimization and clustering operations, it is possible to scientifically group pharmacies according to their characteristics, ensuring the rationality and accuracy of the clustering results. Through statistical analysis of the characteristic distribution of cluster numbers, the category labels of pharmacies are accurately matched with information such as their scale level, target customer group, or business direction, further refining the pharmacy classification. This technical method achieves accurate identification of pharmacy characteristics and clear classification of business characteristics through cluster analysis, providing an important foundation for the subsequent construction of customer portraits and the formulation of marketing plans, significantly improving the efficiency and accuracy of pharmacy classification and analysis, and effectively supporting the implementation of precision marketing strategies.

[0032] 4. By conducting in-depth analysis of the similarities and differences in customer profiles of pharmacies under the same category label, we refined the pharmacy classification and further captured the potential differences in customer needs within pharmacies of the same category. By extracting the differentiated characteristics of target customer profiles, we provided a basis for precise promotional strategies for sub-category pharmacies. At the same time, through dynamic adjustments to promotional campaign plans, we ensured that marketing activities were highly aligned with the customer needs of pharmacies in each sub-category. This technical approach not only enabled the refined adjustment and optimization of promotional activities, significantly improving the accuracy and effectiveness of marketing plans, but also effectively covered diverse customer needs through differentiated strategies, further enhancing the market competitiveness and customer satisfaction of pharmacies.

[0033] 5. By introducing digital twin technology, the actual operating characteristics of pharmacies are digitized and modeled, enabling the prediction and verification of promotional campaign effectiveness in a virtual environment, thereby reducing the trial-and-error costs of real-world promotional activities. Simulation evaluation results can accurately identify potential problems with promotional plans and make timely adjustments, ensuring that promotional activities are more aligned with the pharmacy's actual operating needs and customer profiles. Furthermore, the high simulation capabilities of digital twin models can capture the pharmacy's dynamic sales performance under different promotional strategies, providing data support and deep insights for optimizing marketing plans. This technical approach significantly improves the scientific nature and accuracy of promotional campaign design, reduces the waste of marketing resources, enhances the intelligence of marketing decision-making, and improves the profitability and market competitiveness of pharmacies in actual operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flowchart of a method for generating a marketing activity based on sales data analysis in an embodiment of the present application;

[0035] Figure 2 This is a structural diagram of a marketing activity generation system based on sales data analysis in an embodiment of the present application;

[0036] Figure 3 It is a structural diagram of an electronic device in an embodiment of the present application.

[0037] Explanation of the accompanying drawings: 201, acquisition module; 202, calculation module; 203, extraction module; 204, analysis module; 205, first generation module; 206, second generation module; 207, adjustment module; 208, simulation module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0038] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0039] In the description of the embodiments of this application, words such as "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "for example" or "for instance" in the embodiments of this application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "for example" or "for instance" is intended to present the relevant concepts in a concrete manner.

[0040] In the description of the embodiments of the present application, the term "multiple" means two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0041] Figure 1 This is a flow chart of a method for generating a marketing activity based on sales data analysis in an embodiment of the present application.

[0042] See also Figure 1 In an embodiment of the present application, a method for generating a marketing activity based on sales data analysis is applied to a server, and the method includes:

[0043] S101. Obtain sales data of all pharmacies in the target area, disease maps corresponding to all drugs in the pharmacies, and population health data and environmental climate monitoring data in the target area;

[0044] The system can obtain sales data of the target area through the pharmacy's sales system or the database of the chain pharmaceutical company. The sales data may include drug sales volume, sales time, customer purchase frequency and inventory data, etc., which can reflect the pharmacy's operating conditions and drug demand.

[0045] Disease maps, established networks of relationships between drugs and diseases, can be directly obtained from the National Medical Products Administration's drug database, official information from drug manufacturers, or medical literature. These maps provide a clear basis for analyzing the scope of a drug's application, such as whether a certain class of antibiotics is suitable for treating a specific infectious disease.

[0046] For population health data, public health systems and medical insurance systems can be connected to the data system of the present invention to obtain population health data, or health survey data from the target area can be used to collect population health data. Population health data can include chronic disease prevalence, disease distribution, and health behavior characteristics (such as smoking and drinking rates) to assess the overall health of regional residents.

[0047] In addition, the system can obtain real-time and historical environmental climate monitoring data from the meteorological monitoring platform through an API interface, and can also organize and extract data from publicly released environmental quality reports. Environmental climate monitoring data involves environmental factors such as air quality (such as PM2.5 concentration), temperature, humidity, and precipitation. These data are closely related to health status. For example, air pollution can lead to a high incidence of respiratory diseases.

[0048] To ensure that sales data, disease maps, population health data, and environmental climate data can be linked, they need to be standardized, including in terms of format, identifiers, temporal and spatial dimensions, units, and metrics. First, data from different sources should be converted into a unified structured format, such as a database table or standardized file format. Sales data, disease maps, population health data, and environmental climate data should be matched using unified identifiers (such as drug ID, region ID, and time), ensuring consistent representation of drugs, regions, and time across all datasets. Second, the temporal and spatial dimensions of the data should be aligned. For example, hourly climate data should be aggregated to the daily level to align with the temporal dimensions of sales and health data, and pharmacy addresses should be mapped to administrative divisions to ensure consistent spatial scope. Units and metrics should also be standardized, for example, by standardizing the currency to RMB and converting health metrics to the same statistical units (such as percentages or per thousand people).

[0049] S102. Extract health factors from population health data and environmental climate monitoring data based on preset statistical analysis methods; extract key factors that affect health status from population health data and environmental climate monitoring data through data mining and statistical analysis (such as multiple regression analysis, factor analysis, etc.). These key factors are called health factors. Population health data may include the prevalence of people of different age groups in the region, the incidence of chronic diseases, the spread of infectious diseases, etc.; environmental climate monitoring data may include temperature, humidity, air quality (such as PM2.5, PM10), tap water quality, and seasonal climate changes. For example, in winter, temperature and air humidity may lead to a high incidence of respiratory diseases and are therefore identified as health factors, while in summer, tap water quality and temperature may be closely related to the spread of intestinal diseases and are therefore identified as health factors.

[0050] S103. Hierarchically process the health factors according to multiple time periods to obtain a health factor set, where the time periods include a first time period, a second time period, and a third time period, wherein the time range of the second time period is greater than the time range of the first time period, and the time range of the third time period is greater than the time range of the second time period;

[0051] The extracted health factors are layered according to multiple time periods to generate a set of health factors. The time period is divided into three levels: the first time period (short-term, such as weeks or months), the second time period (medium-term, such as quarters or half years), and the third time period (long-term, such as annual or multi-year trends). The first time period focuses on reflecting short-term changes in health factors, such as the impact of air quality fluctuations on respiratory diseases within a week. The second time period reflects medium-term trends, such as the correlation between the spread of seasonal influenza and temperature fluctuations within a quarter. The third time period is used to capture long-term impacts, such as the impact of year-round air pollution levels on the incidence of chronic diseases (such as asthma or lung disease). Through layered processing, health factors are integrated into different time scales, which can more comprehensively reflect the temporal dynamic characteristics of health factors and their potential impact on health trends.

[0052] S104: Construct a health assessment indicator set based on the health factor set, where the health assessment indicator set includes multiple health assessment indicators for each time period;

[0053] Specifically, the health factors of each time period are feature extracted according to the target dimension to obtain a basic feature vector, which includes the disease risk dimension, the environmental pressure dimension and the population health vulnerability dimension; the basic feature vector is subjected to time series correlation analysis to obtain time series correlation features; the basic feature vector and the time series correlation features are feature fused to obtain a time series health feature vector, which includes feature values ​​corresponding to multiple features; the feature values ​​are ranked by importance using a preset importance evaluation algorithm to obtain an importance ranking result; multiple features in the importance ranking result whose importance scores exceed a preset threshold are determined as health assessment indicators to obtain a health assessment indicator set.

[0054] Among them, the construction of the health assessment indicator set is completed by feature extraction, analysis and screening of health factors in each time period. Specifically, first of all, the core features reflecting the health status of the region can be extracted from the health factor set through statistical methods, machine learning techniques and other methods, and the health situation of the region can be described from aspects such as disease risk, environmental pressure and population health status (population health vulnerability). For example, by analyzing data such as disease incidence, air quality index, temperature fluctuations and the proportion of the elderly population in a certain time period, basic features that can quantify the regional health level are extracted and constructed into a basic feature vector. The basic feature vector is a static description of health factors in different time periods, which is used to characterize the direct effect of health factors on regional health.

[0055] On this basis, further analysis of the dynamic characteristics of health factors in the time dimension is also needed. Through time series correlation analysis, the mutual relationship and potential time dependence of health factors in different time periods can be obtained. For example, fluctuations in air temperature may have a certain lag correlation with the incidence of influenza, while changes in air quality may have a periodic impact on the transmission trend of respiratory diseases. Through analysis of these correlations, the dynamic characteristics of health factors in the time dimension can be extracted, and combined with the basic feature vector, a time series health feature vector is generated. The time series health feature vector contains not only the static characteristics of health factors themselves, but also the dynamic correlation characteristics between health factors.

[0056] To further screen out the most significant features affecting the health situation of the target area, the importance of the feature values in the time series health feature vector needs to be sorted. Each feature value can be quantitatively scored by a pre-set importance evaluation algorithm to evaluate its impact on the health of the region. For example, by analyzing disease incidence, air quality index, and the proportion of the elderly population, the importance score of the feature to the health of the region is calculated based on the pre-set importance evaluation algorithm to identify the key features in the health factors. Among them, the importance evaluation algorithm can quantitatively score each feature value based on the statistical characteristics of the data, combined with the field experience of experts, or based on the feature importance evaluation of the machine learning model. Finally, by sorting the importance scores of all features, the features with scores exceeding the pre-set threshold are screened out to determine the health evaluation indicators, and a health evaluation indicator set is constructed, and for each screened health evaluation indicator, its specific data value (i.e. feature indicator value) in the target time period is matched, which is obtained by extracting, aggregating, calculating or time aligning the population health data and environmental and climate monitoring data, and can accurately reflect the feature state in the target time period. For example, the feature indicator value of the air quality index can be the daily average or monthly average in the time period, the feature indicator value of the influenza incidence can be the weekly case number per thousand people, and the feature indicator value of the proportion of the elderly population can be directly derived from statistical data. For example, in a certain target area and time period, the features obtained by analysis include air quality index (AQI), influenza incidence, proportion of the elderly population, and temperature fluctuation amplitude, etc. After importance evaluation, the importance score of the air quality index is 0.85, the influenza incidence is 0.78, the proportion of the elderly population is 0.65, and the temperature fluctuation amplitude is 0.45. If the pre-set threshold is 0.5, the air quality index, influenza incidence, and proportion of the elderly population are selected as health evaluation indicators. At this time, the health evaluation indicator set will include these indicators and their corresponding feature indicator values, such as air quality index 120 (indicating light pollution), influenza incidence 15‰ (i.e. 15 people per thousand people infected with influenza), and proportion of the elderly population 18% (i.e. the proportion of the elderly population in the region is 18%).

[0057] S105: Aggregate and calculate multiple health assessment indicators in each time period to obtain a health status index in the target area.

[0058] The health assessment index set constructed in step S104 includes health assessment indicators and their characteristic index values ​​for the target area and the corresponding time period. Health assessment indicators are important features that have been screened, such as the air quality index (AQI), influenza incidence, proportion of the elderly population, and temperature fluctuation amplitude; while characteristic index values ​​are specific data of these features within the time period. For example, in a certain time period, the air quality index is 120, the influenza incidence is 15‰, and the proportion of the elderly population is 18%. These characteristic index values ​​will be used as input data for subsequent standardization processing and aggregation calculations.

[0059] Since different health assessment indicators have different dimensions and value ranges (for example, the air quality index may be 0-500, while the influenza incidence rate is 0-100‰), directly using these characteristic indicator values ​​for calculations may result in the results being biased towards indicators with a larger value range. Therefore, it is necessary to standardize the values ​​of each characteristic indicator so that they are within the same scale range. The standardization formula is:

[0060]

[0061] Among them, Z i is the characteristic index value after standardization of the i-th health assessment index, x i is the characteristic index value of the i-th health assessment index, max(x) and min(x) are the upper and lower limits of the value range of the i-th health assessment index, respectively. The standardized characteristic index value Z i The air quality index will fall within the range of [0,1]. For example, if the air quality index ranges from 0 to 500 and the air quality index is 120 during a certain period of time, the normalized value is: Through standardization, all health assessment indicator values ​​are adjusted to a unified range to facilitate subsequent calculations.

[0062] Different health assessment indicators have different degrees of impact on the overall health situation, so it is necessary to assign weights to each indicator. The allocation of weights can be based on expert experience or data analysis results, such as assigning weights through expert scoring, or quantifying the importance of each indicator through statistical analysis methods (such as regression analysis). Suppose the weight distribution results for a certain time period are: the weight of the air quality index (AQI) is 0.4, the weight of the influenza incidence rate is 0.35, and the weight of the proportion of the elderly population is 0.25. This means that air quality has the greatest impact on the regional health situation, followed by the influenza incidence rate, and the impact of the proportion of the elderly population is relatively small. The sum of the weights must meet the following constraints: Among them, ω i is the weight of the i-th health assessment indicator, and n is the total number of health assessment indicators.

[0063] The health status index in the target area is calculated using the aggregation formula:

[0064]

[0065] Among them, H is the health status index of the target area; Z i is the characteristic index value after standardization of the i-th health assessment index; ω i is the weight of the i-th health assessment indicator, reflecting the impact of a single indicator on health status; f(Z i ) is a nonlinear function used to perform nonlinear transformation (e.g., exponential, logarithmic or other forms) on the effect of a single health assessment indicator; ω ij is the interaction weight between the i-th and j-th health assessment indicators, reflecting the impact of the interaction between the indicators on the health situation; g(Z i ,Z j ) is the interaction effect function, which is used to describe the interaction relationship between two indicators; α is the overall contribution adjustment coefficient of a single health assessment indicator; β is the overall contribution adjustment coefficient of the interaction effect; γ is the adjustment coefficient of the bias term; Bias is the bias term, which is usually a constant or set according to prior knowledge in a specific field.

[0066] is the contribution of a single health assessment indicator to the health status index H, f(Z i ) The standardized index value Z i Perform nonlinear transformations; The contribution of the interaction effects between indicators to the health status index was calculated.

[0067] S106. Perform cluster analysis on the feature data set to generate a category label for each pharmacy. The category label includes the pharmacy's scale and target customer group.

[0068] Specifically, based on the disease map, the disease demand characteristic data of the target area is extracted according to the regional health situation index; the clustering parameters are determined based on the preset clustering evaluation indicators, and according to the clustering parameters, the sales data and the disease demand characteristic data are clustered to obtain the clustering results; the cluster number of each pharmacy is determined according to the clustering results; the characteristic distribution statistics of the pharmacies corresponding to the same cluster number in the preset dimensions are counted; and the category label corresponding to each cluster number is determined according to the characteristic distribution statistics.

[0069] The disease map contains information linking drugs and corresponding diseases, describing the therapeutic or preventive relationship between the two. For example, an antiviral drug is primarily used to treat influenza, while a blood pressure medication is primarily targeted at patients with hypertension. Extracting disease demand characteristics data for the target region is accomplished by combining the regional health status index and the disease map. First, based on the disease risk analysis within the regional health status index, major health issues within the target region, such as influenza, respiratory infections, or chronic diseases, are identified. These risks are directly reflected by specific indicators within the health status index (such as the influenza risk index, chronic disease risk index, or respiratory disease risk index). Next, using the disease map, the disease risks identified in the health status index are associated with corresponding drugs or treatment options. For example, if the health status index indicates a high risk of influenza, the disease map can be used to match demand for influenza-related drugs (such as antivirals, fever reducers, and cough suppressants). These disease risks are then compared and analyzed with actual drug sales data to verify the match between the health status index and pharmacy drug demand, ensuring the accuracy of the extracted results. For example, if drug sales data shows a significant increase in antiviral drug sales in winter, this further confirms that influenza is the primary disease demand characteristic in that region. Finally, by comprehensively analyzing the health trend index, disease maps, and sales data, we can extract disease demand characteristics data for the target region, such as "influenza is prevalent in winter, leading to high demand for antiviral drugs" or "chronic diseases dominate the market for a long time, leading to stable demand for antihypertensive and hypoglycemic drugs."

[0070] After extracting the disease demand feature data, it is necessary to perform cluster analysis on the pharmacies in the target area so that the pharmacies can be divided into different categories according to sales characteristics and customer needs. Before performing the clustering operation, it is necessary to first determine the parameters of the cluster analysis, such as selecting a suitable clustering algorithm and setting the number of clusters (K value). Commonly used clustering algorithms include K-Means, DBSCAN, and hierarchical clustering. The determination of the number of clusters K value can be completed by a preset clustering evaluation index. Common methods include determining the K value by finding the inflection point of the sum of squared errors within the cluster or directly setting the initial value based on field experience. For example, in the target area, if it is known that pharmacies can be roughly divided into three categories: large comprehensive pharmacies, small community pharmacies, and specialty pharmacies, K can be initially set to 3.

[0071] After the parameters are determined, cluster analysis is performed by combining pharmacy sales data and disease demand profile data. Sales data includes information such as pharmacy sales volume, sales revenue, and inventory levels, while disease demand profile data reflects the primary demand for specific disease-related medications among customers in the target area. Clustering operations analyze this multidimensional feature data to classify pharmacies into different categories. For example, cluster analysis might categorize pharmacies in the target area into the following categories:

[0072] Category A: Large comprehensive pharmacies, covering a wide range of medicines, with the target customer group mainly being patients with chronic diseases. The sales characteristics are reflected in the high proportion of chronic disease medicines.

[0073] Category B: Small community pharmacies, targeting the daily medication needs of ordinary residents. Their sales characteristics are reflected in the high sales of common medicines such as cold medicines and anti-inflammatory drugs.

[0074] Category C: Specialized pharmacies, targeting specific disease groups (such as diabetic patients), with sales characteristics reflected in the sales volume of specific drugs accounting for a significantly higher proportion than other categories of pharmacies.

[0075] After completing the clustering operation, we need to perform feature distribution statistics for the pharmacies corresponding to each cluster ID to further analyze the common characteristics of pharmacies in each category. Feature distribution statistics include key indicators such as average sales volume, inventory levels, and drug category coverage. For example, for a cluster ID (such as Category A), we can calculate information such as the average monthly sales volume, average number of drug categories covered, and the sales share of chronic disease drugs for pharmacies in that category.

[0076] By analyzing the statistical values ​​of the feature distribution, a category label can be generated for each cluster number. The category label is a brief description of the pharmacy's characteristics, including its scale (such as large pharmacy or small pharmacy) and the characteristics of its target customer group (such as chronic disease patients, ordinary residents, etc.). For example, the category label for category A could be described as "a large-scale comprehensive pharmacy targeting the middle-aged and elderly population with a high demand for chronic disease medications."

[0077] In order to ensure the accuracy and rationality of the clustering results, the clustering effectiveness index needs to be calculated after determining the cluster number of each pharmacy based on the clustering results.

[0078] Specifically, a clustering effectiveness index of the clustering result is calculated; if the clustering effectiveness index does not meet the preset conditions, the clustering parameters are re-determined.

[0079] The clustering effectiveness metric is a quantitative assessment of the quality of the clustering results. Common effectiveness metrics include the silhouette coefficient, Calinski-Harabasz index, and Davies-Bouldin index. For example, if the silhouette coefficient of the current clustering result is 0.4, which is lower than the preset threshold of 0.5, this indicates that there are problems with the clustering results and that further optimization of the clustering parameters is necessary. The clustering operation can be repeated by increasing or decreasing the K value. For example, adjusting the initial K value from 3 to 5 may make the clustering results more consistent with reality. Alternatively, by filtering or weighting the features of the input data, the impact of key features on the clustering results can be more significant. For example, increasing the weight of the feature "drug category coverage" may make the clustering results more accurate. After adjusting the parameters, the clustering operation needs to be repeated and the clustering effectiveness metric recalculated until the metric meets the preset conditions.

[0080] S107. Generate a customer profile based on the category labels and feature dataset. The customer profile includes customer behavior preferences, disease demand characteristics, and drug purchase tendencies.

[0081] Classification labels are derived through cluster analysis and include the pharmacy's size (e.g., large pharmacy, small community pharmacy) and its primary target customer base (e.g., elderly individuals, chronic disease patients, or young customers). Combined with these classification labels, we can initially identify the characteristics of each pharmacy's customer base. For example, a small community pharmacy may primarily serve nearby residents, while a large pharmacy may cover a wider customer base with more diverse needs.

[0082] By analyzing feature datasets, customer behavior preferences, disease demand characteristics, and drug purchasing trends are generated. Customer behavior preference analysis is based on pharmacy sales data, identifying customer purchasing habits, such as purchase frequency, preferred time of day (e.g., whether purchases are concentrated during peak hours in the morning and evening), and purchase channels (online or offline). For example, customers of large pharmacies may be more inclined to make large, one-time purchases, while customers of small pharmacies prioritize daily convenience and smaller purchases. Disease demand characteristics are derived by combining disease maps and regional health trend indices. The disease map provides correlations between drugs and diseases, while the health trend index reflects health risk characteristics in the target area, such as influenza risk and chronic disease risk. This data allows the identification of key disease needs by customer groups. For example, a high demand for antihypertensive drugs indicates that the pharmacy's customer base is primarily composed of patients with chronic diseases, while high sales of cold and antiviral drugs may indicate a high proportion of young customers. Finally, by analyzing drug sales and inventory data, customer drug purchasing trends are identified. This includes customer brand preferences (e.g., preference for well-known brands), price sensitivity (e.g., sales distribution of low-priced versus high-priced drugs), and repeat purchase trends for specific drugs. For example, during peak flu season, customers' purchasing propensity for antiviral drugs increases significantly, while middle-aged and elderly customers may be more inclined to purchase medications for chronic disease management. Ultimately, a customer profile is formed that encompasses their behavioral preferences, disease needs, and medication purchasing tendencies.

[0083] S108. Generate a marketing campaign plan corresponding to each category label using a preset model based on the category label, customer profile, and inventory data in the sales data;

[0084] In step S108, inventory data is first analyzed to determine the pharmacy's best-selling and slow-selling drugs. For example, in a certain category of pharmacy, antihypertensive drugs may be in sufficient inventory and have high sales, while vitamins may be in overstock. In this case, promotional activities should be designed based on the sales characteristics and inventory status of the drugs to improve inventory turnover and sales efficiency. At the same time, inventory analysis also needs to consider potential disease demand. For example, if the health trend index predicts that the peak influenza season is approaching, but the inventory of antiviral drugs is insufficient, a replenishment plan and related promotional activities should be developed in advance.

[0085] Secondly, match customer profiles with inventory data to identify key drugs and target customers for marketing campaigns. For example, pharmacies whose customer profiles show a predominance of chronic disease patients can prioritize promotions for medications like antihypertensives and hypoglycemics, and design targeted promotions such as "discounts on chronic disease medications" or "free health checkup services." For slow-selling or seasonal medications, design dedicated inventory clearance campaigns, such as "buy an antiviral and get a free fever reducer" or "discounts on vitamins."

[0086] Then, a specific marketing campaign plan is generated using a pre-set model. This model can be based on a rule-based decision-making model or a machine learning model, combining behavioral preferences, disease demand characteristics, and drug purchasing tendencies within the customer profile to recommend the most appropriate promotional campaign for each customer. For example, for customers who tend to purchase online, online coupon campaigns can be designed; for customers who prefer high-end branded drugs, promotional campaigns with high-end drug bundles can be designed. Furthermore, the marketing plan should also include health service activities, such as free health lectures or blood sugar and blood pressure testing services for patients with chronic diseases, or promotional vaccination promotions during the peak flu season.

[0087] Finally, we optimize the timing and format of campaign content by comprehensively considering customer behavior preferences, both in terms of timing and channel. For example, we can focus promotional information during peak purchasing periods and reach target customers through multiple channels, such as WeChat groups and app notifications. Through these steps, we ultimately generate marketing campaign plans for each category tag, ensuring that campaigns precisely match customer needs.

[0088] Optional, in Figure 1 After step S108 of the illustrated embodiment, the following steps may be performed:

[0089] Calculate the similarity of the customer portraits of the first pharmacy and generate a similarity matrix, where the first pharmacy is the pharmacy corresponding to the same category label; calculate the difference value of the customer portraits of the first pharmacy based on the similarity matrix; extract the difference features of the target customer portraits, where the target customer portraits are customer portraits with a customer portrait difference value greater than a preset threshold; divide the first pharmacy into second pharmacies corresponding to multiple subcategory labels based on the difference features; adjust the promotion activity plan based on the sales data and customer portraits of each second pharmacy.

[0090] For pharmacies with the same category label (i.e., the first pharmacy), the similarity of customer profiles between pharmacies is calculated based on their customer profiles (including behavioral preferences, disease demand characteristics, and drug purchasing tendencies), generating a similarity matrix. The similarity matrix quantifies the degree of difference between customer profiles of different pharmacies by measuring the similarity. The customer profile difference value of each pharmacy is calculated based on the similarity matrix, and target customer profiles with a customer profile difference value greater than a preset threshold are identified. These profiles reflect significantly different customer needs or behavioral patterns.

[0091] Extract the different characteristics of the target customer profile, such as the significant differences in purchasing preferences, disease needs or consumption habits among different pharmacy customer groups, and further divide the first pharmacy originally belonging to the same category label into second pharmacies corresponding to multiple subcategory labels. This process can more accurately refine the pharmacy classification and identify subtle differences between pharmacies. For example, pharmacies in the same "small community pharmacy" category may be divided into two subcategories: "pharmacy mainly for chronic disease patients" and "pharmacy mainly for seasonal needs of young people" due to different customer age groups or health needs.

[0092] Based on each second-tier pharmacy's sales data and customer profile, promotional campaigns are adjusted to more precisely match the characteristics of each subcategory. For example, for pharmacies primarily catering to chronic disease patients, promotions could focus on medications for chronic conditions like blood pressure and diabetes, with additional discounts on health testing services. Meanwhile, for pharmacies primarily catering to younger customers, promotions could focus on cold medicines and vitamins, and design promotional campaigns on online social platforms.

[0093] Optional, in Figure 1 In the illustrated embodiment, the method further comprises:

[0094] Obtain a digital twin model of the pharmacy corresponding to each category label; based on sales data and customer profiles, use the digital twin model to simulate and evaluate the pharmacy's sales under the promotional plan to obtain simulation evaluation results; and adjust the promotional plan based on the simulation evaluation results.

[0095] The system retrieves a digital twin model of the pharmacy corresponding to each category label from an industry database or internal enterprise technology platform. This digital twin model is pre-built based on pharmacy business data (such as sales records, inventory information, and customer behavior) and market characteristics, accurately reflecting the pharmacy's operational characteristics. After obtaining the digital twin model, the pharmacy's sales data, customer profiles, and current marketing campaign plans are used as inputs. The digital twin model is then used to perform simulation evaluations and predict the potential effectiveness of promotional activities.

[0096] Based on the above input data, the digital twin model simulates the potential impact of promotional activities on pharmacy sales and may generate the following simulation results:

[0097] Sales volume forecasting: Predict sales trends during promotional activities, such as the increase in drug sales and the lasting impact after the activity ends.

[0098] Customer response analysis: Evaluate the target customers' response to the promotion, such as whether the discount is strong enough to attract the target customer group or whether there are any potential customers that are not covered.

[0099] Inventory pressure assessment: Simulate the impact of promotional activities on drug inventory, evaluate whether the inventory can meet the promotional demand, and whether there will be insufficient or excessive inventory problems.

[0100] Based on the simulation and evaluation results of the digital twin model, promotional plans can be optimized and adjusted to address any issues or deficiencies in the campaign. For example, if simulation results indicate that current discounts are insufficient to attract target customers, discounts can be increased or the minimum purchase threshold can be lowered. If the evaluation indicates that inventory of a specific medication may be low, pre-stocking is necessary. If certain customer groups are less responsive, promotional content can be adjusted to more precisely address their needs. Furthermore, promotional timing and channels can be optimized based on customer behavioral preferences, such as adding nighttime coupons for online customers or offering in-store discounts during specific hours for offline customers. These adjustments make promotional activities more tailored to customer needs and improve efficiency.

[0101] See also Figure 2 , is a schematic diagram of the structure of a marketing activity generation system based on sales data analysis provided in an embodiment of the present application. A marketing activity generation system based on sales data analysis 200 specifically includes:

[0102] Acquisition module 201 is used to obtain sales data of all pharmacies in the target area, disease maps corresponding to all drugs in the pharmacies, and population health data and environmental climate monitoring data in the target area;

[0103] A calculation module 202 is used to establish a regional health assessment indicator set based on population health data and environmental climate monitoring data, and calculate the health status index of the target area based on the regional health assessment indicator set;

[0104] Extraction module 203, configured to extract feature information of the first data to obtain a feature data set, wherein the feature data set represents the sales characteristics, customer needs, and health needs of the pharmacy, and the first data includes sales data, disease maps, and health status index;

[0105] Analysis module 204, configured to perform cluster analysis on the feature data set to generate a category label for each pharmacy, the category label including the pharmacy's scale and target customer group;

[0106] The first generation module 205 is used to generate a customer profile based on the category label and feature data set, where the customer profile includes customer behavior preferences, disease demand characteristics, and drug purchase tendencies;

[0107] The second generating module 206 is used to generate a marketing activity plan corresponding to each category tag through a preset model according to the category tag, customer portrait and inventory data in the sales data.

[0108] Optionally, the calculation module 202 is specifically configured to:

[0109] Based on a preset statistical analysis method, health factors are extracted from population health data and environmental climate monitoring data; the health factors are stratified according to multiple time periods to obtain a health factor set, and the time periods include a first time period, a second time period, and a third time period, wherein the time range of the second time period is greater than the time range of the first time period, and the time range of the third time period is greater than the time range of the second time period; a health assessment indicator set is constructed based on the health factor set, and the health assessment indicator set includes multiple health assessment indicators for each time period; the multiple health assessment indicators for each time period are aggregated and calculated to obtain a health status index in the target area.

[0110] Optionally, the calculation module 202 is further configured to:

[0111] The health factors of each time period are extracted according to the target dimensions to obtain the basic feature vector, which includes the disease risk dimension, the environmental pressure dimension and the population health vulnerability dimension; the basic feature vector is subjected to time series correlation analysis to obtain the time series correlation feature; the basic feature vector is fused with the time series correlation feature to obtain the time series health feature vector, which includes the feature values ​​corresponding to multiple features; the feature values ​​are sorted by importance using a preset importance evaluation algorithm to obtain the importance sorting result; multiple features in the importance sorting result whose importance scores exceed the preset threshold are determined as health assessment indicators to obtain a health assessment indicator set.

[0112] Optionally, the analysis module 204 is specifically configured to:

[0113] Based on the disease map, the disease demand characteristic data of the target area is extracted according to the regional health situation index; the clustering parameters are determined based on the preset clustering evaluation indicators, and the sales data and the disease demand characteristic data are clustered according to the clustering parameters to obtain the clustering results; the cluster number of each pharmacy is determined according to the clustering results; the characteristic distribution statistics of the pharmacies corresponding to the same cluster number in the preset dimensions are counted; and the category label corresponding to each cluster number is determined according to the characteristic distribution statistics.

[0114] Optionally, the analysis module 204 is further configured to:

[0115] Calculate clustering effectiveness indicators of clustering results;

[0116] If the clustering effectiveness index does not meet the preset conditions, the clustering parameters are re-determined.

[0117] Optionally, the system further includes an adjustment module 207, specifically configured to:

[0118] Calculate the similarity of the customer portraits of the first pharmacy and generate a similarity matrix, where the first pharmacy is the pharmacy corresponding to the same category label; calculate the difference value of the customer portraits of the first pharmacy based on the similarity matrix; extract the difference features of the target customer portraits, where the target customer portraits are customer portraits with a customer portrait difference value greater than a preset threshold; divide the first pharmacy into second pharmacies corresponding to multiple subcategory labels based on the difference features; adjust the promotion activity plan based on the sales data and customer portraits of each second pharmacy.

[0119] Optionally, the system further includes a simulation module 208, specifically configured to:

[0120] Obtain a digital twin model of the pharmacy corresponding to each category label; based on sales data and customer profiles, use the digital twin model to simulate and evaluate the pharmacy's sales under the promotional plan to obtain simulation evaluation results; and adjust the promotional plan based on the simulation evaluation results.

[0121] It should be noted that the apparatus provided in the above examples is only exemplified by the above division of functional modules when realizing its functions. In actual applications, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above described functions. In addition, the apparatus and method embodiments provided in the above examples belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be described here.

[0122] The embodiment also discloses an electronic device, which refers to Figure 3 The electronic device can include at least one processor 301, at least one communication bus 302, a user interface 303, a network interface 304, and at least one memory 305.

[0123] The communication bus 302 is used to realize the connection and communication between the components.

[0124] The user interface 303 can include a display screen (Display) and a camera (Camera), and the optional user interface 303 can further include a standard wired interface and a wireless interface.

[0125] The network interface 304 can optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0126] The processor 301 can include one or more processing cores. The processor 301 connects various parts of the server through various interfaces and lines, executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be realized in at least one of the hardware forms of a digital signal processing (Digital Signal Processing, DSP), a field-programmable gate array (Field-Programmable Gate Array, FPGA), and a programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate a combination of one or several of a central processing unit (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU), and a modem. Among them, the CPU is mainly used to process the operating system, user interface, and application programs; the GPU is used to render and draw the content to be displayed on the display screen; and the modem is used to process wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 301, but can be realized by a separate chip.

[0127] The memory 305 can include a random access memory (RAM) and can also include a read-only memory (ROM). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets, or instruction sets. The memory 305 can include a program storage area and a data storage area, where the program storage area can store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the various method embodiments described above, etc.; the data storage area can store data involved in the various method embodiments described above, etc. The memory 305 can optionally be at least one storage device located away from the aforementioned processor 301. As shown, the memory 305 as a computer storage medium can include an operating system, a network communication module, a user interface module, and an application program of a marketing activity generation method based on sales data analysis. Figure 3

[0128] In the electronic device shown, the user interface 303 is mainly used to provide an interface for user input and obtain data input by the user; and the processor 301 can be used to call the application program of the marketing activity generation method based on sales data analysis stored in the memory 305, and when executed by one or more processors 301, the electronic device performs the method of one or more of the above embodiments. Figure 3

[0129] It should be noted that, for the above-mentioned method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0130] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0131] ​​In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic, such as the division of units, which is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interface, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0132] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0133] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0134] If the integrated unit 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 memory 305. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory 305 and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory 305 includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a magnetic disk, or an optical disk.

[0135] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any variations, uses or adaptive changes of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for generating marketing activities based on sales data analysis, characterized in that: Applied to a server, the method includes: Obtain sales data of all pharmacies in the target area, disease maps corresponding to all drugs in the pharmacies, and population health data and environmental climate monitoring data in the target area; A regional health assessment indicator set is established based on the population health data and environmental climate monitoring data, and a health status index of the target area is calculated based on the regional health assessment indicator set. The regional health assessment indicator set is used to characterize the health status of the target area. The calculation formula of the health status index is: ， Wherein, H is the health status index, is the characteristic index value after standardization of the i-th health assessment index, is the weight of the i-th health assessment indicator, is a nonlinear function, is the interaction weight between the i-th health assessment indicator and the j-th health assessment indicator, is the interaction effect function between the i-th health assessment index and the j-th health assessment index, is the overall contribution adjustment coefficient of a single health assessment indicator, is the overall contribution adjustment coefficient of the interaction effect, is the adjustment coefficient of the bias term, Bias is the bias term; Extracting feature information of the first data to obtain a feature data set, wherein the feature data set represents sales characteristics, customer needs, and health needs of the pharmacy, the first data including the sales data, the disease map, and the health status index; Performing cluster analysis on the feature data set to generate a category label for each of the pharmacies, the category label including the scale level and target customer group of the pharmacy; Generate a customer profile based on the category label and the feature data set, the customer profile including customer behavior preferences, disease demand characteristics, and drug purchasing tendencies; Generate a marketing campaign plan corresponding to each category tag through a preset model based on the category tag, the customer profile, and the inventory data in the sales data; The cluster analysis of the feature data set to generate a category label for each of the pharmacies specifically includes: Based on the disease map, extracting disease demand characteristic data of the target area according to the regional health situation index; Determining clustering parameters based on preset clustering evaluation indicators, and performing clustering operations on the sales data and the disease demand feature data according to the clustering parameters to obtain clustering results; Determine a cluster number for each of the pharmacies according to the clustering result; Counting the characteristic distribution statistics of the pharmacies corresponding to the same cluster number in the preset dimension; Determine the category label corresponding to each cluster number according to the feature distribution statistical value; After determining the cluster number of each pharmacy according to the clustering result, the method further includes: Calculating a clustering effectiveness index of the clustering result; If the clustering effectiveness index does not meet the preset conditions, the clustering parameters are re-determined; The method further comprises: Obtain a digital twin model of the pharmacy corresponding to each category label; Based on the sales data and the customer profile, a simulation evaluation is performed on the sales situation of the pharmacy under the marketing activity plan through the digital twin model to obtain a simulation evaluation result; Adjust the marketing campaign plan based on the simulation evaluation results.

2. The method according to claim 1, characterized in that The step of establishing a regional health assessment indicator set based on the population health data and the environmental climate monitoring data, and calculating the health status index of the target area based on the regional health assessment indicator set, specifically includes: extracting health factors from the population health data and the environmental climate monitoring data based on a preset statistical analysis method; hierarchically processing the health factors according to a plurality of time periods to obtain a health factor set, the time periods comprising a first time period, a second time period, and a third time period, wherein a time range of the second time period is greater than a time range of the first time period, and a time range of the third time period is greater than a time range of the second time period; Constructing a health assessment indicator set according to the health factor set, wherein the health assessment indicator set includes a plurality of health assessment indicators for each of the time periods; Aggregate and calculate the multiple health assessment indicators in each time period to obtain a health status index in the target area.

3. The method according to claim 2, characterized in that The step of constructing a health assessment indicator set based on the health factor set specifically includes: Extracting features of the health factors of each time period according to target dimensions to obtain a basic feature vector, wherein the target dimensions include disease risk dimension, environmental pressure dimension, and population health vulnerability dimension; Performing time series correlation analysis on the basic feature vector to obtain a time series correlation feature; Performing feature fusion on the basic feature vector and the time series correlation feature to obtain a time series health feature vector, where the time series health feature vector includes feature values ​​corresponding to multiple features; The importance of the characteristic values ​​is sorted by using a preset importance evaluation algorithm to obtain an importance sorting result; The plurality of features whose importance scores in the importance ranking result exceed a preset threshold are determined as health assessment indicators to obtain a health assessment indicator set.

4. The method according to claim 1, wherein After generating a marketing campaign plan for each category corresponding to the category label using a preset model based on the category label, the customer profile, and the inventory data in the sales data, the method further includes: Calculating the similarity of customer portraits of the first pharmacy to generate a similarity matrix, where the first pharmacy is the pharmacy corresponding to the same category label; Calculating a customer portrait difference value of the first pharmacy according to the similarity matrix; Extracting difference features of the target customer portrait, wherein the target customer portrait is a customer portrait whose difference value of the customer portrait is greater than a preset threshold; Dividing the first pharmacy into second pharmacies corresponding to a plurality of subcategory labels according to the difference characteristics; Adjust the marketing campaign plan based on the sales data of each second pharmacy and the customer portrait.

5. A marketing activity generation system based on sales data analysis, characterized in that: include: An acquisition module is used to obtain sales data of all pharmacies in the target area, disease maps corresponding to all drugs in the pharmacies, and population health data and environmental climate monitoring data in the target area; A calculation module is used to establish a regional health assessment indicator set based on the population health data and environmental climate monitoring data, and calculate the health status index of the target area based on the regional health assessment indicator set. The regional health assessment indicator set is used to characterize the health status of the target area. The calculation formula of the health status index is: ， Wherein, H is the health status index, is the characteristic index value after standardization of the i-th health assessment index, is the weight of the i-th health assessment indicator, is a nonlinear function, is the interaction weight between the i-th health assessment indicator and the j-th health assessment indicator, is the interaction effect function between the i-th health assessment index and the j-th health assessment index, is the overall contribution adjustment coefficient of a single health assessment indicator, is the overall contribution adjustment coefficient of the interaction effect, is the adjustment coefficient of the bias term, Bias is the bias term; an extraction module, configured to extract feature information of the first data to obtain a feature data set, wherein the feature data set represents the sales characteristics, customer needs, and health needs of the pharmacy, the first data including the sales data, the disease map, and the health status index; An analysis module, configured to perform cluster analysis on the feature data set to generate a category label for each of the pharmacies, the category label including the scale level and target customer group of the pharmacy; A first generating module is configured to generate a customer profile based on the category label and the feature data set, wherein the customer profile includes customer behavior preferences, disease demand characteristics, and drug purchase tendencies; A second generating module is configured to generate a marketing activity plan corresponding to each category label through a preset model based on the category label, the customer profile, and the inventory data in the sales data; The analysis module is specifically configured to extract disease demand characteristic data of the target area based on the disease atlas and the regional health status index; Determining clustering parameters based on preset clustering evaluation indicators, and performing clustering operations on the sales data and the disease demand feature data according to the clustering parameters to obtain clustering results; Determine a cluster number for each of the pharmacies according to the clustering result; Counting the characteristic distribution statistics of the pharmacies corresponding to the same cluster number in the preset dimension; Determine the category label corresponding to each cluster number according to the feature distribution statistical value; The analysis module is further specifically used to calculate the clustering effectiveness index of the clustering result; If the clustering effectiveness index does not meet the preset conditions, the clustering parameters are re-determined; A simulation module, specifically used to obtain a digital twin model of the pharmacy corresponding to each of the category labels; Based on the sales data and the customer profile, a simulation evaluation is performed on the sales situation of the pharmacy under the marketing activity plan through the digital twin model to obtain a simulation evaluation result; Adjust the marketing campaign plan based on the simulation evaluation results.

6. A marketing campaign generation device based on sales data analysis, characterized in that: include: one or more processors and memory; The memory is coupled to the one or more processors, and is used to store computer program code, where the computer program code includes computer instructions. The one or more processors call the computer instructions to enable the device for generating marketing activities based on sales data analysis to execute the method according to any one of claims 1 to 4.

7. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a device for generating a marketing campaign based on sales data analysis, the device is caused to execute the method according to any one of claims 1 to 4.

Images