A marketing activity optimization promotion method and system based on generative model

Through artificial intelligence algorithm based on generative models, analyzing the data of the pharmaceutical industry and generating personalized marketing solutions, solving the problems of low data analysis efficiency, low prediction accuracy and poor interpretability in the pharmaceutical industry marketing, and achieving more efficient marketing strategy formulation and Market insight.

CN116703469BActive Publication Date: 2024-11-12BEIJING ELEMENTS STARTING POINT TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310969517.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-08-03
Publication Date
2024-11-12
Estimated Expiration
2043-08-03

AI Technical Summary

Technical Problem

There are problems in marketing in the pharmaceutical industry with low data analysis efficiency, low prediction accuracy, poor model interpretability and insufficient customized services, resulting in insufficient efficiency and accuracy in marketing strategy formulation.

Method used

Using artificial intelligence algorithm based on generative models, by collecting and preprocessing market dynamic data, corporate competition data and consumer behavior data of the pharmaceutical industry, using generative models to analyze market trends and consumption rules, generate personalized marketing and promotion plans, and Explain the analysis results through the interpreter tool.

Benefits of technology

It improves the efficiency and accuracy of marketing activities, provides customized promotion solutions, helps enterprises better understand algorithm suggestions and market trends, and enhances market insights and decision-making capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116703469B_ABST
    Figure CN116703469B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of pharmaceutical marketing technology, and in particular to a marketing activity optimization promotion method and system based on a generative model, the method comprising the following steps: S1, collecting and summarizing market dynamic data, enterprise competition data and consumer behavior data in the pharmaceutical industry marketing field as marketing data, and preprocessing and classifying and storing the marketing data; S2, analyzing and summarizing the market trends and consumption patterns in the marketing data using an artificial intelligence algorithm model based on a generative model; S3, generating a personalized and customized exclusive promotion plan for the enterprise in combination with the prediction results, enterprise needs and market status; S4, using an interpreter to simplify the analysis and interpretation of the exclusive promotion plan and the evaluation and analysis results. The present invention greatly improves the efficiency of marketing activities by automatically processing and analyzing data and generating marketing plans and prediction results using an artificial intelligence algorithm model based on a generative model.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of pharmaceutical marketing technology, and in particular to a method and system for optimizing and promoting marketing activities based on a generation model. Background Art

[0002] The importance of marketing in the pharmaceutical industry lies in its ability to help companies achieve the following goals: First, through the correct positioning and promotion strategies, pharmaceutical products can be promoted to the right target market to meet the needs of patients and medical professionals. Second, pharmaceutical marketing helps to establish and consolidate the brand reputation of the company and increase the popularity and recognition of the product. In addition, through cooperation with medical institutions and decision makers, it can promote the rational allocation of medical resources and good treatment outcomes for patients.

[0003] In the marketing field of the pharmaceutical industry, there are currently a variety of relevant technical means to meet the growing demand for medical marketing, mainly including the following aspects:

[0004] 1. Data mining and analysis technology: This technology is often used to mine and analyze historical sales data, hospital diagnosis and treatment data, etc., to explore market demand, consumption trends, etc. These technologies can provide valuable information, but cannot automatically generate new promotion plans or predict results.

[0005] 2. Machine learning algorithms: In recent years, with the rapid development of machine learning technology, many pharmaceutical companies have begun to try to use machine learning algorithms for marketing analysis and forecasting. These algorithms can train models with historical data to predict future market changes and trends. However, these algorithms require professionals to train and optimize the models, and lack model interpretability.

[0006] 3. Marketing intelligence platform: Some companies are actively trying to build marketing intelligence platforms, integrating various data, analysis tools and algorithms to achieve intelligent management and optimization of marketing activities. These platforms usually require a lot of investment and time to develop and maintain, and lack personalized customization services.

[0007] Although the pharmaceutical industry has begun to use various technical means in the field of marketing, these technologies still have some shortcomings, as follows:

[0008] 1. Low data analysis efficiency: Traditional data analysis technology requires professionals to operate, and the analysis efficiency is low, and the results are difficult to guarantee effectiveness and accuracy.

[0009] 2. Low prediction accuracy: Machine learning algorithms require a large amount of historical data to train models, and it is difficult to achieve perfect prediction accuracy.

[0010] 3. Poor model interpretability: Many algorithms lack the ability to explain the models, and it is difficult for companies to understand how these algorithms make recommendations or predictions.

[0011] 4. Insufficient customized services: Existing marketing intelligence platforms often lack personalized customized services and cannot meet the different needs and special circumstances of enterprises. Summary of the invention

[0012] Based on this, it is necessary to provide a marketing activity optimization and promotion method and system based on a generative model to address the above technical issues.

[0013] In a first aspect, the present invention provides a method for optimizing and promoting marketing activities based on a generation model, the method comprising the following steps:

[0014] S1. Collect and summarize market dynamics data, enterprise competition data and consumer behavior data in the pharmaceutical industry marketing field as marketing data, and pre-process and classify and store the marketing data;

[0015] S2. Use artificial intelligence algorithm models based on generative models to analyze and summarize market trends and consumption patterns in marketing data, and predict potential promotion plans for pharmaceutical industry marketing activities;

[0016] S3. Generate a customized promotion plan for the enterprise based on the forecast results, enterprise needs and market status, and evaluate and analyze the marketing effect and competitive intelligence of the promotion plan;

[0017] S4. Use the interpreter to simplify the analysis and interpretation of the exclusive promotion plan and evaluation analysis results.

[0018] In one embodiment, using an artificial intelligence algorithm model based on a generative model to analyze and summarize market trends and consumption patterns in marketing data and predict potential promotion plans for pharmaceutical industry marketing activities includes the following steps:

[0019] S21. Input the pre-processed marketing data into the artificial intelligence algorithm model for training, identify the pharmaceutical product information involved in the marketing data, and assign hierarchical marketing labels based on the identification results;

[0020] S22. extracting parts of the marketing data with the same marketing tag as product data, setting a marketing analysis interval, and analyzing the market trend and consumption rules of the product data;

[0021] S23. Determine the advantages and disadvantages of the pharmaceutical product corresponding to the product data in the marketing and sales process based on the analysis results, and predict the potential promotion plan for the pharmaceutical product.

[0022] In one embodiment, the pre-processed marketing data is input into an artificial intelligence algorithm model for training, the pharmaceutical product information involved in the marketing data is identified, and hierarchical marketing labels are assigned according to the identification results, including the following steps:

[0023] S211. Input the marketing data into the artificial intelligence algorithm model to identify and extract the pharmaceutical product information involved in the marketing data;

[0024] S212: The product type of pharmaceutical products is taken as the first level and given a first-level label;

[0025] S213, the therapeutic area of ​​pharmaceutical products is regarded as the second level and given a secondary label;

[0026] S214. The target audience of pharmaceutical products is regarded as the third level and given a third-level label;

[0027] S214. The price range of pharmaceutical products is regarded as the fourth level and is assigned a four-level label.

[0028] In one embodiment, extracting a portion of the marketing data with the same marketing tag as commodity data, setting a marketing analysis interval, and analyzing the market trend and consumption pattern of the commodity data includes the following steps:

[0029] S221. In the marketing data, the parts with the same marketing tags at four levels are preferentially selected as product data, which are used as the data basis for market trends and consumption patterns;

[0030] S222, setting a time length and an analysis period as a time-based marketing analysis interval;

[0031] S223, extracting market dynamic data from the product data, constructing a market time series that combines the analysis cycle time nodes and the market dynamic data of the pharmaceutical product, using the ratio of the market dynamic data in the current analysis cycle minus the market dynamic data in the previous analysis cycle to the analysis cycle duration as the node score of the market trend of the pharmaceutical product, taking all continuous node scores in the marketing analysis interval as the market trend of the pharmaceutical product, and drawing a change curve;

[0032] S224. Extract consumer behavior records from product data, calculate the brand share of different brands of pharmaceutical products and the channel share of different sales channels in each analysis period in turn based on consumers' purchasing characteristics, and obtain the new occupancy rate of new consumers in the total number of consumers in a single analysis period, and use the change patterns of the above three occupancy rates over time as the consumption pattern of the pharmaceutical product.

[0033] In one embodiment, judging the advantages and disadvantages of the pharmaceutical product corresponding to the product data in the marketing and sales process according to the analysis results, and predicting the potential promotion plan of the pharmaceutical product includes the following steps:

[0034] S231. Select the parts of the marketing data that have the same marketing labels in the first three levels as competitive product data, and calculate the market trends and consumption patterns of pharmaceutical products in different price ranges according to the four-level labels;

[0035] S232. Compare and analyze the market trends and consumption patterns of product data and competitor data to determine the market performance of pharmaceutical products in different price ranges;

[0036] S233. Based on the market trends and consumption patterns of the product data within the marketing analysis period, use a deep learning algorithm to predict the trends and patterns of the pharmaceutical product within the next analysis period;

[0037] S234. Generate potential promotion plans based on the market performance and trend prediction results of pharmaceutical products in the corresponding price range of comprehensive commodity data.

[0038] In one embodiment, comparing and analyzing the market trends and consumption patterns of product data and competitor product data to determine the market performance of pharmaceutical products in different price ranges includes the following steps:

[0039] S2321. Compare the market trend of the product data with the market trend of the competitor data, calculate the difference between the actual value of the market dynamic data and the average value of the market trend change curve in the marketing analysis interval, if the difference between the two is less than the preset threshold, then it is determined that the market performance of the pharmaceutical product corresponding to the product data has a benign competition phenomenon, if the difference between the two is greater than the preset threshold and the price range of the competitor data is less than the price range of the product data, then it is determined that the market performance of the pharmaceutical product corresponding to the product data has a low-price replacement phenomenon, if the difference between the two is greater than the preset threshold and the price range of the competitor data is greater than the price range of the product data, then it is determined that the market performance of the pharmaceutical product corresponding to the product data has a high-price suppression phenomenon;

[0040] S2322. Compare the market trend of the commodity data with the consumption pattern of the competitor data, and calculate the difference between the actual values ​​of each market share in the two and the average values ​​in the marketing analysis interval. If the market share of a certain brand in the commodity data is greater than the sum of the market shares of other brands, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a brand monopoly phenomenon. If the market share of a certain channel in the commodity data is greater than the sum of the market shares of other channels, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a channel monopoly phenomenon. If the new share in the commodity data is lower than the new share of the competitor data and the difference is greater than the preset threshold, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a shortage of new users.

[0041] In one embodiment, based on the market trend and consumption pattern of the product data in the marketing analysis period, using a deep learning algorithm to predict the trend and pattern of the pharmaceutical product in the next analysis period includes the following steps:

[0042] S2331. Node scores in the product data, brand occupancy rates of various brands, channel occupancy rates of various sales channels, and newly added occupancy rates are sequentially used as data to be predicted, and a set of data to be predicted is formed according to N analysis cycles within the marketing analysis interval;

[0043] S2332, performing differential preprocessing on the data set to be predicted to eliminate the trend between adjacent data, wherein the differential preprocessing calculation formula is: In the formula, y ( t ) represents the data to be predicted after differential preprocessing; x ( t ) represents the data to be predicted; Indicates that the data to be predicted is t - p The value of time; t Indicates the analysis period; p Indicates the delay time of the data set to be predicted; S2333, taking the data set to be predicted after differential preprocessing as the input of the deep learning network model to perform feature learning, and extracting the data features of the data to be predicted;

[0044] S2334. Using the data features as input, the support vector regression prediction model is used to predict the data to be predicted in the next analysis period, and the reverse difference is used to calculate and obtain the value of the data to be predicted, wherein the reverse difference calculation formula is: In the formula, It represents the prediction result of the next analysis cycle after the reverse difference processing. represents the predicted value for the next analysis period, Indicated in t - p+1 predicted value.

[0045] In one embodiment, generating a potential promotion plan based on the market performance and trend law prediction results of the pharmaceutical products in the price range corresponding to the product data includes the following steps:

[0046] S2341. If there is a phenomenon of low-price substitution in the market performance of commodity data, price adjustment marketing shall be carried out for the pharmaceutical products corresponding to the commodity data in combination with the trend law prediction results. If there is a phenomenon of brand exclusivity at the same time, the price adjustment marketing efforts of other brand pharmaceutical products shall be increased. If there is a phenomenon of channel exclusivity at the same time, the price adjustment marketing efforts of other channel pharmaceutical products shall be increased and the channel business scope shall be adjusted. If there is a phenomenon of lack of new users at the same time, the promotion coverage of pharmaceutical products shall be increased.

[0047] S2342. If there is a high price suppression phenomenon in the market performance of commodity data, then in combination with the trend law prediction results, the pharmaceutical products corresponding to the commodity data will be subject to augmented publicity and marketing. If there is a brand monopoly phenomenon at the same time, the augmented publicity and marketing efforts of the monopoly brand pharmaceutical products will be increased. If there is a channel monopoly phenomenon at the same time, the augmented publicity and marketing efforts of the monopoly channel pharmaceutical products will be increased and the depth of channel promotion will be increased.

[0048] S2343. If there is healthy competition in the market performance of commodity data, fine-tune and augment the marketing of the pharmaceutical products corresponding to the commodity data in combination with the trend law prediction results. If there is also brand exclusivity, increase the price adjustment marketing efforts of the remaining brand pharmaceutical products. If there is also channel exclusivity, increase the price adjustment marketing efforts of the remaining channel pharmaceutical products and expand the channel business scope. If there is also a shortage of new users, increase the promotion coverage of the pharmaceutical products and conduct expanded publicity and promotion based on the target audience of the pharmaceutical products.

[0049] In one embodiment, combining the prediction results, enterprise needs and market status, generating an enterprise-customized exclusive promotion plan, and evaluating and analyzing the marketing effect and competitive intelligence of the exclusive promotion plan include the following steps:

[0050] S31. Obtain the expected market data and strategic positioning of pharmaceutical companies as corporate needs, and make adaptive adjustments to the marketing efforts of pharmaceutical products based on the current market status of pharmaceutical products;

[0051] S32. Select market channels and target audiences that match the company's strategic positioning, and generate customized exclusive promotion plans based on potential promotion plans;

[0052] S33. Collect the market actions, product strategies and marketing activities of competing enterprises, and make targeted adjustments to the exclusive promotion plan based on the market performance of the enterprise itself;

[0053] S34. Promote the exclusive promotion plan on the service platform, monitor market dynamics data in real time, and conduct marketing effect analysis and evaluation based on feedback content and forecast results.

[0054] In a second aspect, the present invention further provides a marketing activity optimization and promotion system based on a generation model, the system comprising the following modules:

[0055] The data collection module is used to collect and summarize market dynamic data, enterprise competition data and consumer behavior data in the pharmaceutical industry marketing field as marketing data, and pre-process and classify and store the marketing data;

[0056] Model prediction module, which is used to analyze and summarize the market trends and consumption patterns in marketing data using artificial intelligence algorithm models based on generative models, and to predict potential promotion plans for marketing activities in the pharmaceutical industry;

[0057] The enterprise customization module is used to combine the forecast results, enterprise needs and market status to generate a personalized and exclusive promotion plan for the enterprise, and evaluate and analyze the marketing effect and competitive intelligence of the exclusive promotion plan;

[0058] The interpreter module is used to simplify the analysis and interpretation of the exclusive promotion plan and evaluation analysis results using the interpreter.

[0059] Beneficial effects of the present invention:

[0060] 1. The marketing activity optimization and promotion method and system based on the generative model can automatically process and analyze data, and generate marketing plans and prediction results by utilizing the artificial intelligence algorithm model based on the generative model, thereby greatly improving the efficiency of marketing activities; at the same time, customized services can be further optimized according to enterprise needs, thereby more effectively improving marketing efficiency; at the same time, through deep learning, market trends and rules can be automatically identified to generate more accurate prediction results and promotion plans, which can more effectively help enterprises better grasp market dynamics and consumer needs and make more informed decisions.

[0061] 2. This marketing activity optimization and promotion method and system based on the generative model can realize the intelligent management and optimization of big data related to pharmaceutical products by automatically identifying the patterns and trends in the data. At the same time, the interpreter tool can also help companies better understand the logic behind the algorithm suggestions and prediction results, better guide the company's decision-making, help pharmaceutical companies quickly respond to market changes, better grasp market opportunities, and thus improve the company's competitiveness and market share. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0063] Figure 1 A flow chart of a marketing activity optimization and promotion method based on a generative model of the present invention;

[0064] Figure 2 This is a schematic diagram of the structure of a marketing activity optimization and promotion system based on a generation model of the present invention.

[0065] Figure numbers: 1. Data acquisition module; 2. Model prediction module; 3. Enterprise customization module; 4. Interpreter module. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0067] See also Figure 1 , provides a marketing activity optimization promotion method based on a generation model, the method comprising the following steps:

[0068] S1. Collect and summarize market dynamics data, enterprise competition data and consumer behavior data in the pharmaceutical industry marketing field as marketing data, and pre-process and classify and store the marketing data.

[0069] In the description of the present invention, the process of collecting market dynamics data, enterprise competition data and consumer behavior data is achieved through multiple channels, such as market research, consumer surveys, enterprise public reports, social media monitoring, etc. Ensure that the data source is reliable, representative and fully covers different dimensions.

[0070] 1. Market dynamic data includes but is not limited to the following aspects:

[0071] Market Size and Growth Rate: Understand the size and growth trends of specific pharmaceutical markets.

[0072] Market share: Assess each competitor's share and position in the market.

[0073] Market segments and trends: Understand the different market segments and changing trends.

[0074] Market Demand and Consumer Preferences: Understand consumer needs, preferences, and behavior patterns.

[0075] 2. Enterprise competition data includes but is not limited to the following aspects:

[0076] Competitor information: Collect information on competitors’ company profile, product lines, market positioning, etc.

[0077] Product Features and Benefits: Understand your competitors’ product features, benefits, and unique selling points.

[0078] Pricing strategy: Analyze competitors' pricing strategies and pricing levels.

[0079] Market Share and Growth Rate: Evaluate competitors' market share and growth rate.

[0080] 3. Consumer behavior data includes but is not limited to the following:

[0081] Purchase behavior: Understand consumers’ purchase decision process, purchase channels and frequency.

[0082] Preferences and Needs: Analyze consumer preferences, needs, and purchasing motivations.

[0083] Feedback and Reviews: Collect consumer feedback, reviews, and word-of-mouth information.

[0084] Marketing channel preference: Understand consumers’ preference and usage of different marketing channels.

[0085] Summarize and organize the collected data, and pre-process the collected data to clean and organize the data to ensure the accuracy and consistency of the data. This includes removing duplicate data, processing missing values, correcting erroneous data, etc. At the same time, standardize and convert the data format for subsequent analysis and classification. Finally, classify and store the pre-processed data according to the characteristics of the data and classification requirements. Classify according to different data types, themes or dimensions, such as market size data, competitor data, consumer purchasing behavior data, etc. Use appropriate data management tools and systems to ensure data security and easy retrieval.

[0086] S2. Use artificial intelligence algorithm models based on generative models to analyze and summarize market trends and consumption patterns in marketing data, and predict potential promotion plans for marketing activities in the pharmaceutical industry.

[0087] In the description of the present invention, the artificial intelligence algorithm model adopts the PharmGPT model, which is an artificial intelligence algorithm model based on a generative model. GPT (Generative Pre-trained Transformer) is a generative model based on the Transformer architecture, which can generate natural language text after large-scale unsupervised training. PharmGPT is a GPT model that has been specifically trained and adjusted for the medical field. It uses relevant data in the medical field for pre-training, so that it has the ability to understand and generate text in the medical field. By pre-training on a large amount of medical literature, drug databases, clinical data, etc., PharmGPT can generate medical-related analysis, recommendations and predictions.

[0088] Generative models are a type of artificial intelligence model that aims to generate new samples similar to training data. Compared with discriminative models, generative models have the ability to generate new samples and can handle open-ended tasks and unknown data. In the pharmaceutical field, PharmGPT can generate medical texts, diagnostic recommendations, drug properties and other information to assist pharmaceutical research and clinical decision-making.

[0089] In the description of the present invention, the artificial intelligence algorithm model based on the generative model is used to analyze and summarize the market trends and consumption patterns in the marketing data, and to predict the potential promotion plans for the marketing activities of the pharmaceutical industry, including the following steps:

[0090] S21. Input the pre-processed marketing data into the artificial intelligence algorithm model for training, identify the pharmaceutical product information involved in the marketing data, and assign hierarchical marketing labels based on the identification results.

[0091] In the description of the present invention, the pre-processed marketing data is input into the artificial intelligence algorithm model for training, the pharmaceutical product information involved in the marketing data is identified, and hierarchical marketing labels are assigned according to the identification results, including the following steps:

[0092] S211. Input the marketing data into the artificial intelligence algorithm model to identify and extract the pharmaceutical product information involved in the marketing data.

[0093] S212. The product type of pharmaceutical products is taken as the first level and assigned a first-level label.

[0094] Among them, primary labels include prescription drugs, over-the-counter drugs (OTC drugs), biological products and health products;

[0095] S213. The therapeutic area of ​​pharmaceutical products is regarded as the second level and is assigned a secondary label.

[0096] Among them, the secondary labels include cardiovascular drugs, nervous system drugs, digestive system drugs, oncology drugs and immune system drugs.

[0097] S214. The target audience of pharmaceutical products is regarded as the third level and is given a third-level label.

[0098] Among them, the third-level labels include medicines for children, medicines for adults, medicines for the elderly, etc.

[0099] S214. The price range of pharmaceutical products is regarded as the fourth level and is assigned a four-level label.

[0100] Among them, the four-level labels include high-end products, mid-range products and low-end products, or the price range is divided according to the pricing information of the goods to assign four-level labels.

[0101] S22. Extract the parts with the same marketing tags from the marketing data as commodity data, set the marketing analysis interval, and analyze the market trends and consumption patterns of the commodity data.

[0102] In the description of the present invention, extracting a portion of the marketing data with the same marketing tag as commodity data, setting a marketing analysis interval, and analyzing the market trend and consumption rules of the commodity data includes the following steps:

[0103] S221. In the marketing data, the parts with the same marketing labels at four levels are preferentially selected as product data, which are used as the data basis for market trends and consumption patterns.

[0104] S222. Set the time length and analysis period as the time-based marketing analysis interval.

[0105] S223. Extract market dynamic data from commodity data, construct a market time series that combines analysis cycle time nodes with market dynamic data of pharmaceutical commodities, use the ratio of the market dynamic data in the current analysis cycle minus the market dynamic data in the previous analysis cycle to the length of the analysis cycle as the node score of the market trend of the pharmaceutical commodity, use all continuous node scores in the marketing analysis interval as the market trend of the pharmaceutical commodity, and draw a change curve.

[0106] S224. Extract consumer behavior records from product data, calculate the brand share of different brands of pharmaceutical products and the channel share of different sales channels in each analysis period in turn based on consumers' purchasing characteristics, and obtain the new occupancy rate of new consumers in the total number of consumers in a single analysis period, and use the change patterns of the above three occupancy rates over time as the consumption pattern of the pharmaceutical product.

[0107] S23. Determine the advantages and disadvantages of the pharmaceutical product corresponding to the product data in the marketing and sales process based on the analysis results, and predict the potential promotion plan for the pharmaceutical product.

[0108] In the description of the present invention, judging the advantages and disadvantages of the pharmaceutical product corresponding to the product data in the marketing and sales process according to the analysis results, and predicting the potential promotion plan of the pharmaceutical product includes the following steps:

[0109] S231. Select the parts of the marketing data with the same marketing labels in the first three levels as competitive product data, and use the four-level labels as a distinction to calculate the market trends and consumption patterns of pharmaceutical products in different price ranges.

[0110] S232. Compare and analyze the market trends and consumption patterns of product data and competitor data to determine the market performance of pharmaceutical products in different price ranges.

[0111] In the description of the present invention, comparing and analyzing the market trends and consumption patterns of product data and competitor data to determine the market performance of pharmaceutical products in different price ranges includes the following steps:

[0112] S2321. Compare the market trend of the commodity data with the market trend of the competitor data, calculate the difference between the actual value of the market dynamic data in the two and the average value of the market trend change curve in the marketing analysis interval; if the difference between the two is less than the preset threshold, then it is determined that the market performance of the pharmaceutical commodity corresponding to the commodity data has a benign competition phenomenon; if the difference between the two is greater than the preset threshold and the selling price range of the competitor data is smaller than the selling price range of the commodity data, then it is determined that the market performance of the pharmaceutical commodity corresponding to the commodity data has a low-price replacement phenomenon; if the difference between the two is greater than the preset threshold and the selling price range of the competitor data is greater than the selling price range of the commodity data, then it is determined that the market performance of the pharmaceutical commodity corresponding to the commodity data has a high-price suppression phenomenon.

[0113] S2322. Compare the market trend of the commodity data with the consumption pattern of the competitor data, and calculate the difference between the actual values ​​of each market share in the two and the average values ​​in the marketing analysis interval. If the market share of a certain brand in the commodity data is greater than the sum of the market shares of other brands, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a brand monopoly phenomenon. If the market share of a certain channel in the commodity data is greater than the sum of the market shares of other channels, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a channel monopoly phenomenon. If the new share in the commodity data is lower than the new share of the competitor data and the difference is greater than the preset threshold, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a shortage of new users.

[0114] S233. Based on the market trends and consumption patterns of the product data within the marketing analysis period, use deep learning algorithms to predict the trends and patterns of the pharmaceutical product in the next analysis cycle.

[0115] In the description of the present invention, based on the market trends and consumption patterns of the commodity data in the marketing analysis interval, the trend and pattern of the pharmaceutical commodity in the next analysis period are predicted by using a deep learning algorithm, including the following steps:

[0116] S2331. The node scores in the product data, the brand share of each brand, the channel occupancy rate of each sales channel and the newly added occupancy rate are sequentially used as the data to be predicted, and a set of data to be predicted is formed according to N analysis cycles within the marketing analysis interval.

[0117] S2332, performing differential preprocessing on the data set to be predicted to eliminate the trend between adjacent data, wherein the differential preprocessing calculation formula is: In the formula, y ( t ) represents the data to be predicted after differential preprocessing, x ( t ) represents the data to be predicted, Indicates that the data to be predicted is t - p The value of time, t Represents the analysis period, p S2333: The data set to be predicted after differential preprocessing is used as the input of the deep learning network model to perform feature learning and extract data features of the data to be predicted.

[0118] S2334. Using the data features as input, the support vector regression prediction model is used to predict the data to be predicted in the next analysis period, and the reverse difference is used to calculate and obtain the value of the data to be predicted, wherein the reverse difference calculation formula is: In the formula, It represents the prediction result of the next analysis cycle after the reverse difference processing. represents the predicted value for the next analysis period, Indicated in t - p +1 when the predicted value. S234. Generate potential promotion plans based on the market performance and trend prediction results of pharmaceutical products in the corresponding price range of the comprehensive commodity data.

[0119] In the description of the present invention, generating a potential promotion plan based on the market performance and trend law prediction results of the pharmaceutical products in the price range corresponding to the comprehensive commodity data includes the following steps:

[0120] S2341. If there is a low-price replacement phenomenon in the market performance of commodity data, price adjustment marketing will be implemented for the pharmaceutical products corresponding to the commodity data in combination with the trend law prediction results. If there is also a brand exclusivity phenomenon, the price adjustment marketing efforts of other brand pharmaceutical products will be increased. If there is also a channel exclusivity phenomenon, the price adjustment marketing efforts of other channel pharmaceutical products will be increased and the channel business scope will be adjusted. If there is also a shortage of new users, the promotion coverage of pharmaceutical products will be increased.

[0121] S2342. If there is a phenomenon of high price suppression in the market performance of commodity data, then the augmented publicity and marketing of the pharmaceutical products corresponding to the commodity data shall be carried out in combination with the trend law prediction results. If there is also a brand exclusivity phenomenon, then the augmented publicity and marketing of the monopoly brand pharmaceutical products shall be increased. If there is also a channel exclusivity phenomenon, then the augmented publicity and marketing of the monopoly channel pharmaceutical products shall be increased and the depth of channel promotion shall be increased.

[0122] S2343. If there is healthy competition in the market performance of commodity data, fine-tune and augment the marketing of the pharmaceutical products corresponding to the commodity data in combination with the trend law prediction results. If there is also brand exclusivity, increase the price adjustment marketing efforts of the remaining brand pharmaceutical products. If there is also channel exclusivity, increase the price adjustment marketing efforts of the remaining channel pharmaceutical products and expand the channel business scope. If there is also a shortage of new users, increase the promotion coverage of the pharmaceutical products and conduct expanded publicity and promotion based on the target audience of the pharmaceutical products.

[0123] S3. Combine the forecast results, enterprise needs and market status to generate a personalized and exclusive promotion plan for the enterprise, and evaluate and analyze the marketing effect and competitive intelligence of the exclusive promotion plan.

[0124] In the description of the present invention, combining the prediction results, enterprise needs and market status, generating an enterprise-customized exclusive promotion plan, and evaluating and analyzing the marketing effect and competitive intelligence of the exclusive promotion plan include the following steps:

[0125] S31. Obtain the expected market data and strategic positioning of pharmaceutical companies as corporate needs, and make adaptive adjustments to the marketing efforts of pharmaceutical products based on the current market status of pharmaceutical products.

[0126] S32. Select market channels and target audiences that are consistent with the company's strategic positioning, and combine potential promotion plans to generate a customized and exclusive promotion plan for the company.

[0127] S33. Collect market actions, product strategies and marketing activities of competing companies, and make targeted adjustments to the exclusive promotion plan based on the company's own market performance.

[0128] S34. Promote the exclusive promotion plan on the service platform, monitor market dynamics data in real time, and conduct marketing effect analysis and evaluation based on feedback content and forecast results.

[0129] S4. Use the interpreter to simplify the analysis and interpretation of the exclusive promotion plan and evaluation analysis results.

[0130] In the description of the present invention, an interpreter tool is provided. The interpreter can analyze and interpret the results of the PharmGPT algorithm, allowing companies to better understand the logic behind the model and use it to develop more effective marketing strategies.

[0131] See also Figure 2 , provides a marketing activity optimization and promotion system based on a generative model, which includes the following modules:

[0132] The data collection module 1 is used to collect and summarize market dynamic data, enterprise competition data and consumer behavior data in the pharmaceutical industry marketing field as marketing data, and pre-process and classify and store the marketing data.

[0133] Model prediction module 2 is used to analyze and summarize the market trends and consumption patterns in the marketing data using an artificial intelligence algorithm model based on a generative model, and to predict potential promotion plans for marketing activities in the pharmaceutical industry.

[0134] Enterprise customization module 3 is used to combine the forecast results, enterprise needs and market status to generate an exclusive promotion plan customized for the enterprise, and evaluate and analyze the marketing effect and competitive intelligence of the exclusive promotion plan.

[0135] The interpreter module 4 is used to use the interpreter to simplify the analysis and interpretation of the exclusive promotion plan and the evaluation and analysis results.

[0136] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

Claims

1. A marketing activity optimization and promotion method based on generative model, It is characterized in that The method comprises the following steps: S1. Collect and summarize market dynamics data, enterprise competition data and consumer behavior data in the pharmaceutical industry marketing field as marketing data, and pre-process and classify and store the marketing data; The market dynamics data include: market size and growth rate, market share, market segmentation and trends, market demand and consumer preferences; the enterprise competition data include: competitor information, product features and advantages, pricing strategy, market share and growth rate; the consumer behavior data include: purchasing behavior, preferences and needs, feedback and evaluation, marketing channel preferences; S2. Use artificial intelligence algorithm models based on generative models to analyze and summarize market trends and consumption patterns in marketing data, and predict potential promotion plans for pharmaceutical industry marketing activities; S3. Generate a customized promotion plan for the enterprise based on the forecast results, enterprise needs and market status, and evaluate and analyze the marketing effect and competitive intelligence of the promotion plan; S4. Use the interpreter to simplify the analysis and interpretation of the exclusive promotion plan and evaluation analysis results; The method of using an artificial intelligence algorithm model based on a generative model to analyze and summarize the market trends and consumption patterns in the marketing data and predicting potential promotion plans for marketing activities in the pharmaceutical industry includes the following steps: S21. Input the pre-processed marketing data into the artificial intelligence algorithm model for training, identify the pharmaceutical product information involved in the marketing data, and assign hierarchical marketing labels based on the identification results; S22. extracting parts of the marketing data with the same marketing tag as product data, setting a marketing analysis interval, and analyzing the market trend and consumption rules of the product data; S23. Determine the advantages and disadvantages of the pharmaceutical product corresponding to the product data in the marketing and sales process based on the analysis results, and predict the potential promotion plan for the pharmaceutical product; The step of inputting the pre-processed marketing data into the artificial intelligence algorithm model for training, identifying the pharmaceutical product information involved in the marketing data, and assigning hierarchical marketing labels according to the identification results includes the following steps: S211. Input the marketing data into the artificial intelligence algorithm model to identify and extract the pharmaceutical product information involved in the marketing data; S212: The product type of pharmaceutical products is taken as the first level and given a first-level label; S213, the therapeutic area of ​​pharmaceutical products is regarded as the second level and given a secondary label; S214. The target audience of pharmaceutical products is regarded as the third level and given a third-level label; S214. The price range of pharmaceutical products is taken as the fourth level and given a four-level label; The extracting of the parts with the same marketing tags from the marketing data as commodity data, setting the marketing analysis interval, and analyzing the market trend and consumption rules of the commodity data includes the following steps: S221. In the marketing data, the parts with the same marketing tags at four levels are preferentially selected as product data, which are used as the data basis for market trends and consumption patterns; S222, setting a time length and an analysis period as a time-based marketing analysis interval; S223, extracting market dynamic data from the product data, constructing a market time series that combines the analysis cycle time nodes and the market dynamic data of the pharmaceutical product, using the ratio of the market dynamic data in the current analysis cycle minus the market dynamic data in the previous analysis cycle to the analysis cycle duration as the node score of the market trend of the pharmaceutical product, taking all continuous node scores in the marketing analysis interval as the market trend of the pharmaceutical product, and drawing a change curve; S224, extracting consumer behavior records from the product data, calculating the brand share of different brands of pharmaceutical products and the channel share of different sales channels in each analysis period in turn according to the purchasing characteristics of consumers, and obtaining the new occupancy rate of new consumers in all consumers in a single analysis period, and taking the change patterns of the above three occupancy rates over time as the consumption pattern of the pharmaceutical product; Judging the advantages and disadvantages of the pharmaceutical product corresponding to the product data in the marketing and sales process based on the analysis results, and predicting the potential promotion plan of the pharmaceutical product includes the following steps: S231. Select the parts of the marketing data that have the same marketing labels in the first three levels as competitive product data, and calculate the market trends and consumption patterns of pharmaceutical products in different price ranges according to the four-level labels; S232. Compare and analyze the market trends and consumption patterns of product data and competitor data to determine the market performance of pharmaceutical products in different price ranges; S233. Based on the market trends and consumption patterns of the product data within the marketing analysis period, use a deep learning algorithm to predict the trends and patterns of the pharmaceutical product within the next analysis period; S234. Generate potential promotion plans based on the market performance and trend prediction results of pharmaceutical products in the corresponding price range of the comprehensive commodity data; The method of predicting the trend and consumption pattern of the pharmaceutical product in the next analysis period by using a deep learning algorithm based on the market trend and consumption pattern of the product data in the marketing analysis period includes the following steps: S2331. Node scores in the product data, brand occupancy rates of various brands, channel occupancy rates of various sales channels, and newly added occupancy rates are sequentially used as data to be predicted, and a set of data to be predicted is formed according to N analysis cycles within the marketing analysis interval; S2332, performing differential preprocessing on the data set to be predicted to eliminate the trend between adjacent data, wherein the differential preprocessing calculation formula is: y(t)=x(t)-x(tp) In the formula, y(t) represents the data to be predicted after differential preprocessing; x(t) represents the data to be predicted; x(tp) represents the value of the data to be predicted at tp; t represents the analysis period; p represents the delay time of the data set to be predicted; S2333, taking the data set to be predicted after the differential preprocessing as the input of the deep learning network model to perform feature learning, and extracting data features of the data to be predicted; S2334. Using the data features as input, the support vector regression prediction model is used to predict the data to be predicted in the next analysis period, and the reverse difference is used to calculate and obtain the value of the data to be predicted, wherein the reverse difference calculation formula is: In the formula, It represents the prediction result of the next analysis cycle after the reverse difference processing. represents the predicted value for the next analysis period, represents the predicted value at t-p+1; The method of combining the forecast results, enterprise needs and market status to generate an exclusive promotion plan customized for the enterprise, and evaluating and analyzing the marketing effect and competitive intelligence of the exclusive promotion plan includes the following steps: S31. Obtain the expected market data and strategic positioning of pharmaceutical companies as corporate needs, and make adaptive adjustments to the marketing efforts of pharmaceutical products based on the current market status of pharmaceutical products; S32. Select market channels and target audiences that match the enterprise's strategic positioning, and generate a customized exclusive promotion plan for the enterprise based on the potential promotion plan; S33. Collect the market actions, product strategies and marketing activities of competing enterprises, and make targeted adjustments to the exclusive promotion plan in combination with the market performance of the enterprise itself; S34. The exclusive promotion plan is promoted on the service platform, and market dynamic data is monitored in real time. Marketing effect analysis and evaluation are performed based on feedback content and forecast results.

2. A marketing activity optimization and promotion method based on a generative model according to claim 1, Features: The comparison and analysis of the market trends and consumption patterns of the product data and the competing product data to determine the market performance of pharmaceutical products in different price ranges includes the following steps: S2321. Compare the market trend of the product data with the market trend of the competitor data, calculate the difference between the actual value of the market dynamic data and the average value of the market trend change curve in the marketing analysis interval, if the difference between the two is less than the preset threshold, then it is determined that the market performance of the pharmaceutical product corresponding to the product data has a benign competition phenomenon, if the difference between the two is greater than the preset threshold and the price range of the competitor data is less than the price range of the product data, then it is determined that the market performance of the pharmaceutical product corresponding to the product data has a low-price replacement phenomenon, if the difference between the two is greater than the preset threshold and the price range of the competitor data is greater than the price range of the product data, then it is determined that the market performance of the pharmaceutical product corresponding to the product data has a high-price suppression phenomenon; S2322. Compare the market trend of the commodity data with the consumption pattern of the competitor data, and calculate the difference between the actual values ​​of each market share in the two and the average values ​​in the marketing analysis interval. If the market share of a certain brand in the commodity data is greater than the sum of the market shares of other brands, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a brand monopoly phenomenon. If the market share of a certain channel in the commodity data is greater than the sum of the market shares of other channels, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a channel monopoly phenomenon. If the new share in the commodity data is lower than the new share of the competitor data and the difference is greater than the preset threshold, then it is determined that the market performance of the pharmaceutical product corresponding to the commodity data has a shortage of new users.

3. A marketing activity optimization and promotion method based on a generative model according to claim 2, Features: The generation of potential promotion plans based on the market performance and trend law prediction results of the pharmaceutical products in the price range corresponding to the comprehensive product data includes the following steps: S2341. If there is a low-price replacement phenomenon in the market performance of the commodity data, then price adjustment marketing is performed on the pharmaceutical commodity corresponding to the commodity data in combination with the trend law prediction result; if there is a brand exclusivity phenomenon at the same time, the price adjustment marketing efforts of the remaining brand pharmaceutical commodities are increased; if there is a channel exclusivity phenomenon at the same time, the price adjustment marketing efforts of the remaining channel pharmaceutical commodities are increased and the channel business scope is adjusted; if there is a shortage of new users at the same time, the promotion coverage of the pharmaceutical commodities is increased; S2342. If there is a high price suppression phenomenon in the market performance of the commodity data, then, in combination with the trend law prediction results, perform augmented publicity and marketing for the pharmaceutical products corresponding to the commodity data; if there is a brand monopoly phenomenon at the same time, then increase the augmented publicity and marketing efforts of the monopoly brand pharmaceutical products; if there is a channel monopoly phenomenon at the same time, then increase the augmented publicity and marketing efforts of the monopoly channel pharmaceutical products and increase the depth of channel promotion; S2343. If there is healthy competition in the market performance of the commodity data, fine-tuning and augmented marketing will be performed on the pharmaceutical products corresponding to the commodity data in combination with the trend law prediction results. If there is also brand exclusivity, the price adjustment marketing efforts of other brand pharmaceutical products will be increased. If there is also channel exclusivity, the price adjustment marketing efforts of other channel pharmaceutical products will be increased and the channel business scope will be expanded. If there is also a shortage of new users, the promotion coverage of pharmaceutical products will be increased, and expanded publicity and promotion will be carried out according to the target audience of the pharmaceutical products.

4. A marketing activity optimization and promotion system based on a generative model, used to implement the marketing activity optimization and promotion method based on a generative model as described in any one of claims 1 to 3, It is characterized in that The system includes the following modules: The data collection module (1) is used to collect and summarize market dynamic data, enterprise competition data and consumer behavior data in the pharmaceutical industry marketing field as marketing data, and pre-process and classify and store the marketing data; A model prediction module (2), which is used to analyze and summarize the market trends and consumption patterns in the marketing data using an artificial intelligence algorithm model based on a generative model, and to predict potential promotion plans for marketing activities in the pharmaceutical industry; Enterprise customization module (3), used to generate an enterprise-customized exclusive promotion plan based on the forecast results, enterprise needs and market status, and to evaluate and analyze the marketing effect and competitive intelligence of the exclusive promotion plan; The interpreter module (4) is used to use the interpreter to simplify the analysis and interpretation of the exclusive promotion plan and the evaluation and analysis results.