Tobacco product brand packaging and propaganda strategy determination method based on data mining

Abstract

The invention discloses a tobacco product brand packaging and propaganda strategy determination method based on data mining. The method comprises the following steps: analyzing a first influence factor on tobacco product brand packaging according to an industry association announcement and news; predicting a second influence factor of tax and industry requirement change on the brand packaging and propaganda strategy of the tobacco product; analyzing acceptance degree information of consumers on brand packaging and health warning of tobacco products; mining preference degree information of consumers on different brand packaging design elements; determining diversified direction information of brand packaging of the tobacco products; analyzing a market brand packaging trend and a competition situation; determining a relation curve between the sales performance and different brand packages and different propaganda strategies; evaluating third influence factors of brand packaging and propaganda strategies of the tobacco products on market sales; and determining a target brand package and a target propaganda strategy of the tobacco product. The accuracy and reliability of tobacco product brand packaging and propaganda strategy making are improved, and the method can be applied to the technical field of information.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method for determining tobacco product brand packaging and publicity strategies based on data mining. Background Art

[0002] With the development and increasingly fierce competition in the tobacco industry, cross-departmental collaborative development and product branding, packaging, and promotional strategies are crucial to companies' market competitiveness. However, the current tobacco industry faces several issues with "industry-industry collaboration" and product branding, packaging, and promotion, requiring in-depth analysis and improvement. First, current product branding, packaging, and promotional strategies lack a deep understanding of consumer needs and preferences. Tobacco companies need to strengthen "industry-industry collaboration" to jointly promote development. This should also include understanding consumer preferences for different product branding, packaging, and design elements, as well as their acceptance of product branding, packaging, and health warnings. Only by gaining a deep understanding of consumer needs and preferences can companies develop product branding, packaging, and promotional strategies that better meet market demand. Second, current product branding, packaging, and promotional strategies lack diversity and personalization. Tobacco companies typically offer only a few flavors, qualities, and types of product branding, failing to meet the diverse needs of consumers. Furthermore, the lack of personalized product branding, packaging, and design also hinders product differentiation in the market. Therefore, tobacco companies need to optimize their product branding, packaging, and promotional strategies based on consumer needs and market trends, achieving diversification and personalization. Furthermore, current product branding, packaging, and promotional strategies lack in-depth analysis and understanding of competitors. Tobacco companies need to monitor competitors' product brand packaging, design, and promotional strategies, as well as analyze popular product brand packaging trends and competitive dynamics. Only by gaining a deep understanding of competitors' strategies can companies develop more competitive product brand packaging and promotional strategies. Finally, current product brand packaging and promotional strategies lack an assessment of their specific impact on market sales and overlook methods for designing tobacco product layouts at tobacco display counters to boost sales. This hinders tobacco companies from providing specific service guidance to cigarette retail customers and fostering a positive customer-customer relationship. Summary of the Invention

[0003] The purpose of the present invention is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0004] To this end, an object of an embodiment of the present invention is to provide a method for determining tobacco product brand packaging and promotional strategies based on data mining, which improves the accuracy and reliability of tobacco product brand packaging and promotional strategy formulation, thereby improving the market competitiveness of tobacco companies and tobacco products.

[0005] Another object of an embodiment of the present invention is to provide a tobacco product brand packaging and promotion strategy determination system based on data mining.

[0006] In order to achieve the above technical objectives, the technical solutions adopted by the embodiments of the present invention include:

[0007] In a first aspect, an embodiment of the present invention provides a method for determining tobacco product brand packaging and promotional strategies based on data mining, comprising the following steps:

[0008] Crawl negative reviews of the cigarette industry based on industry association announcements and news, extract key information related to brand packaging requirements based on the negative reviews, and analyze the primary influencing factors on tobacco product brand packaging based on the key information;

[0009] Analyze tobacco taxation and tobacco control regulations to predict the secondary impact of changes in taxation and industry requirements on tobacco product brand packaging and promotional strategies;

[0010] Use public opinion analysis tools to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites;

[0011] Based on online consumer feedback and comments on tobacco product brand packaging, sentiment analysis technology was used to explore consumer preferences for different brand packaging design elements;

[0012] Analyze consumer demand for brand packaging of different qualities and types based on market sales data, and determine the diversification direction of tobacco product brand packaging;

[0013] Crawl competitor brand packaging design data, use convolutional neural networks to identify brand packaging images, and analyze market brand packaging trends and competitive situations;

[0014] Based on advertising campaigns, tobacco product brand packaging designs, and tobacco product sales data, determine the relationship curve between sales performance and different brand packaging and different promotional strategies;

[0015] Based on the publicity effect and market feedback of tobacco product brand packaging, multiple regression analysis and prediction model were used to evaluate the third factor affecting tobacco product brand packaging and publicity strategy on market sales;

[0016] The target brand packaging and target promotion strategy of tobacco products are determined based on the acceptance level information, the preference level information, the diversification direction information, the market brand packaging trend and competition situation, the relationship curve, the first influencing factor, the second influencing factor and the third influencing factor.

[0017] Furthermore, in one embodiment of the present invention, crawling negative evaluation information of the cigarette industry based on industry association announcements and news, extracting key information related to brand packaging requirements based on the negative evaluation information, and analyzing the first influencing factor on tobacco product brand packaging based on the key information specifically includes:

[0018] crawling first text data of industry association announcements and news reports, performing denoising, word segmentation, and stop word removal on the first text data to obtain a word sequence;

[0019] Using a named entity recognition algorithm to identify the first entity related to tobacco products and tobacco companies in the word sequence;

[0020] extracting the key information related to industry regulation changes and brand packaging requirements from the first text data using keyword extraction or topic modeling based on the first entity;

[0021] Performing word frequency analysis on the key information to determine the importance and frequency of occurrence of each word;

[0022] Determine the relevance and context of each word based on its order and position in the text;

[0023] Performing sentiment analysis on the key information to obtain the ratio of positive sentiment, negative sentiment, and neutral sentiment, and determining the sentiment attitude and sentiment change;

[0024] Conducting topic distribution judgment on the key information, determining the subject words and subject weights, constructing a topic association graph, and analyzing the association relationships between topics;

[0025] Determine the specific content of the changes announced by the industry association and the first influencing factor on tobacco product brand packaging.

[0026] Furthermore, in one embodiment of the present invention, the analysis of tobacco taxation and tobacco control regulations to predict the second impact factor of changes in taxation and industry requirements on tobacco product brand packaging and promotional strategies specifically includes:

[0027] Crawl attribute data on product brand packaging specifications, brand packaging design features, printing information, brand packaging materials, promotional channels, and promotional content;

[0028] Based on the second text data of tobacco tax and tobacco control regulations, the impact on tobacco product brand packaging and promotion strategies is used as the result label;

[0029] Using a random forest algorithm, we constructed a predictive model for the impact of tax regulation changes on brand packaging promotions. This model determined how tobacco product brand packaging specifications, design features, materials, promotional channels, and promotional content would change under different tobacco tax and tobacco control regulations.

[0030] Based on the prediction results, the second influencing factor of tobacco tax and tobacco control regulations on tobacco product brand packaging and promotion strategies is determined.

[0031] Furthermore, in one embodiment of the present invention, the public opinion analysis tool is used to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites, which specifically includes:

[0032] Crawling comment text from social media platforms and professional health websites, and collecting consumer evaluation text on tobacco product brand packaging design, information delivery effectiveness, and health warning acceptance;

[0033] Using the TF-IDF algorithm, we extract keywords related to the attractiveness of tobacco product brand packaging, the fit with the brand image, and the degree to which the design style matches personal aesthetics.

[0034] Using the TextRank algorithm to determine consumer acceptance of different brand packaging and health warnings;

[0035] Using sentiment analysis algorithms, we analyze consumers’ perceptions of the impact of health warnings on smoking behavior and their level of concern for their own health.

[0036] Statistics are collected on the frequency of key words in consumers’ comments on the environmental friendliness of tobacco product brand packaging, and analysis is made of consumers’ recognition of the environmental friendliness of brand packaging materials and their acceptance of environmentally friendly brand packaging.

[0037] Furthermore, in one embodiment of the present invention, sentiment analysis technology is used to mine information on consumers' preferences for different brand packaging design elements based on consumers' online feedback and comments on tobacco product brand packaging, which specifically includes:

[0038] Obtain data containing evaluations and sentiments about brand packaging colors based on consumers’ online feedback and reviews;

[0039] Use sentiment analysis algorithms to analyze consumers’ preferences and emotional responses to different colors, and determine consumers’ preferences for packaging designs of different colors;

[0040] Obtain data containing evaluations and sentiments about brand packaging images and graphics based on consumers’ online feedback and reviews;

[0041] Use sentiment analysis algorithms to analyze consumers’ preferences and emotional responses to different images and patterns, and determine consumers’ preferences for packaging designs with different images and patterns;

[0042] Extract data containing evaluations and sentiments about brand packaging fonts and typography based on consumers’ online feedback and reviews;

[0043] Use sentiment analysis algorithms to analyze consumers’ preferences and emotional responses to different fonts and layouts, and determine consumers’ preferences for packaging designs with different fonts and layouts;

[0044] Extract data containing evaluations and emotional expressions of brand packaging materials and textures based on consumers’ online feedback and comments;

[0045] Sentiment analysis algorithms are used to analyze consumers’ preferences and reactions to different materials and textures, and to determine consumers’ preference information for packaging designs with different materials and textures.

[0046] Furthermore, in one embodiment of the present invention, analyzing consumer demand for brand packaging of different qualities and types based on market sales data to determine information on the diversification direction of tobacco product brand packaging specifically includes:

[0047] Obtaining consumer purchase history data, preference data, and evaluation data on tobacco products;

[0048] Based on consumers’ purchase history and preference data, obtain consumers’ demand for brand packaging of different qualities and types;

[0049] Based on the tobacco products purchased by consumers and their corresponding brand packaging characteristics, obtain consumers' brand packaging needs for tobacco product characteristics and brand image;

[0050] Based on different consumers' purchase history and evaluation data, collaborative filtering algorithms are used to analyze consumers' preferences and relevance for different tobacco product brand packaging attributes, and to determine consumers' preferences and requirements for brand packaging attributes such as safety, sustainability, visual appeal, and convenience.

[0051] Based on consumers' preferences and correlations for the packaging attributes of various tobacco product brands, a collaborative filtering algorithm is used to analyze the tobacco product brand packaging attributes that consumers value most, and determine the diversified direction information of tobacco product brand packaging.

[0052] Furthermore, in one embodiment of the present invention, crawling competitor's brand packaging design data, using a convolutional neural network to identify brand packaging images, and analyzing market brand packaging trends and competitive situations specifically includes:

[0053] Obtain and label competitor brand packaging design image datasets to obtain label data containing brand packaging colors, shapes, patterns, and text;

[0054] Based on the acquired brand packaging design image dataset and label data, a convolutional neural network algorithm is used to build a brand packaging image recognition model to determine the brand packaging color, shape, pattern, and text attributes of new brand packaging images;

[0055] Competitors' brand packaging design images are divided into several clusters, each cluster representing a brand packaging trend or competitive situation. A K-means clustering algorithm is used to construct a brand packaging feature clustering analysis model.

[0056] Based on the characteristic attributes of the cluster, the brand packaging design trends in the market and the brand packaging strategies of competitors are determined.

[0057] Furthermore, in one embodiment of the present invention, determining a relationship curve between sales performance and different brand packaging and different promotional strategies based on advertising campaigns, tobacco product brand packaging designs, and tobacco product sales data specifically includes:

[0058] Obtain sales data and information on brand packaging types, including brand packaging types, sales performance, and brands;

[0059] Through data preprocessing, sales data and brand packaging types are encoded and association rules are mined;

[0060] Constructing frequent item sets of different brand packaging types and sales performance, wherein the frequent item sets represent the association relationship between different brand packaging and sales performance;

[0061] Construct association rules and extract association rules from frequent item sets based on confidence thresholds;

[0062] Evaluate the importance of association rules based on support and confidence, and select association rules with practical significance;

[0063] Based on the obtained association rules, the relationship curve between sales performance and different brand packaging and different promotional strategies is determined.

[0064] Furthermore, in one embodiment of the present invention, the third influencing factor of tobacco product brand packaging and promotional strategy on market sales is evaluated using a multiple regression analysis and prediction model based on the promotional effect of tobacco product brand packaging and market feedback, which specifically includes:

[0065] Obtain data related to tobacco product brand packaging design and color, including sales data, market research, and consumer feedback and comment text for different designs and color schemes;

[0066] Use multiple regression analysis to determine the primary influence coefficient of brand packaging design and color on market sales;

[0067] Obtain relevant data on the packaging materials and quality of different brands, including consumer evaluations of brand packaging materials and quality and product protection performance;

[0068] Use multiple regression analysis to determine the second impact coefficient of brand packaging material and quality on market sales volume;

[0069] Obtain relevant data on different promotional channels and content, including exposure of different channels and content and consumer response to the promotional content;

[0070] Use multiple regression analysis to determine the third influence coefficient of publicity channels and publicity content on market sales volume;

[0071] Obtain data on the target audience's response to brand packaging and promotional strategies, including consumer purchase intention and brand awareness;

[0072] Use multiple regression analysis to determine the fourth coefficient of influence of target audience response on market sales volume;

[0073] The third influencing factor of tobacco product brand packaging and promotional strategy on market sales is determined based on the first influencing coefficient, the second influencing coefficient, the third influencing coefficient, and the fourth influencing coefficient.

[0074] In a second aspect, an embodiment of the present invention provides a tobacco product brand packaging and promotion strategy determination system based on data mining, comprising:

[0075] A first impact factor determination module is used to crawl negative evaluation information of the cigarette industry based on industry association announcements and news, extract key information related to brand packaging requirements based on the negative evaluation information, and analyze the first impact factor on tobacco product brand packaging based on the key information;

[0076] The second impact factor determination module is used to analyze tobacco taxation and tobacco control regulations and predict the second impact factor of changes in taxation and industry requirements on tobacco product brand packaging and promotional strategies;

[0077] An acceptance information determination module, which uses public opinion analysis tools to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites;

[0078] The preference information determination module is used to mine consumer preferences for different brand packaging design elements based on online consumer feedback and comments on tobacco product brand packaging using sentiment analysis technology;

[0079] A diversification direction information determination module is used to analyze consumer demand for brand packaging of different qualities and types based on market sales data, and determine diversification direction information for tobacco product brand packaging;

[0080] The module for determining market brand packaging trends and competitive situations is used to crawl competitors' brand packaging design data, use convolutional neural networks to identify brand packaging images, and analyze market brand packaging trends and competitive situations.

[0081] A relationship curve determination module is used to determine the relationship curve between sales performance and different brand packaging and different promotional strategies based on advertising activities, tobacco product brand packaging design and tobacco product sales data;

[0082] The third influencing factor determination module is used to evaluate the third influencing factor of tobacco product brand packaging and promotional strategies on market sales based on the promotional effects of tobacco product brand packaging and market feedback, using multiple regression analysis and prediction models;

[0083] The brand packaging and promotion strategy determination module is used to determine the target brand packaging and target promotion strategy of tobacco products based on the acceptance level information, the preference level information, the diversification direction information, the market brand packaging trend and competitive situation, the relationship curve, the first influencing factor, the second influencing factor and the third influencing factor.

[0084] In a third aspect, an embodiment of the present invention provides a device for determining tobacco product brand packaging and promotional strategies based on data mining, comprising:

[0085] at least one processor;

[0086] at least one memory for storing at least one program;

[0087] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned method for determining tobacco product brand packaging and promotion strategies based on data mining.

[0088] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium storing a program executable by a processor, wherein the program executable by the processor is used to execute the above-mentioned method for determining tobacco product brand packaging and promotional strategies based on data mining when executed by the processor.

[0089] The advantages and benefits of the present invention will be described in part in the following description and will become apparent from the following description or learned through practice of the present invention:

[0090] The embodiment of the present invention crawls negative evaluation information of the cigarette industry based on industry association announcements and news, extracts key information related to brand packaging requirements based on the negative evaluation information, and analyzes the first influencing factor of tobacco product brand packaging based on the key information, analyzes tobacco tax and tobacco control regulations, and predicts the second influencing factor of tax and industry requirement changes on tobacco product brand packaging and promotion strategies. Public opinion analysis tools are used to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites. Sentiment analysis technology is used to explore consumer preference information for different brand packaging design elements based on consumers' online feedback and comments on tobacco product brand packaging. Consumer demand for brand packaging of different qualities and types is analyzed based on market sales data. , determine the diversification direction information of tobacco product brand packaging, crawl competitors' brand packaging design data, use convolutional neural networks to recognize brand packaging images, analyze market brand packaging trends and competitive situations, and determine the relationship curve between sales performance and different brand packaging and different promotional strategies based on advertising activities, tobacco product brand packaging design, and tobacco product sales data. Based on the promotional effect of tobacco product brand packaging and market feedback, use multiple regression analysis and prediction models to evaluate the third influencing factor of tobacco product brand packaging and promotional strategy on market sales. Based on the acceptance level information, preference level information, diversification direction information, market brand packaging trends and competitive situations, the relationship curve, the first influencing factor, the second influencing factor, and the third influencing factor, the target brand packaging and target promotional strategy of the tobacco product are determined. The embodiments of the present invention improve the accuracy and reliability of the formulation of tobacco product brand packaging and promotional strategies, thereby improving the market competitiveness of tobacco companies and tobacco products. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following introduction is made to the drawings required for use in the embodiments of the present invention. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative work.

[0092] Figure 1 A flowchart of a method for determining tobacco product brand packaging and promotional strategies based on data mining provided by an embodiment of the present invention;

[0093] Figure 2A structural diagram of a tobacco product brand packaging and promotion strategy determination system based on data mining provided by an embodiment of the present invention;

[0094] Figure 3 This is a structural block diagram of a device for determining tobacco product brand packaging and promotional strategies based on data mining provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0095] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and are not to be construed as limiting the present invention. The step numbers in the following embodiments are provided for ease of explanation only and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0096] In the description of the present invention, "a plurality" means two or more. The terms "first" and "second" are used solely to distinguish technical features and are not to be construed as indicating or implying relative importance, or as implicitly indicating the number of the indicated technical features, or as implicitly indicating the order of the indicated technical features. Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art.

[0097] Reference Figure 1 The embodiment of the present invention provides a method for determining tobacco product brand packaging and promotion strategy based on data mining, which specifically includes the following steps:

[0098] S101. Crawl negative evaluation information of the cigarette industry based on industry association announcements and news, extract key information related to brand packaging requirements based on the negative evaluation information, and analyze the first influencing factors on tobacco product brand packaging based on the key information.

[0099] As an optional implementation method, based on industry association announcements and news, negative evaluation information about the cigarette industry is crawled, key information related to brand packaging requirements is extracted based on the negative evaluation information, and the primary influencing factors on tobacco product brand packaging are analyzed based on the key information, which specifically includes:

[0100] S1011, crawling first text data of industry association announcements and news reports, performing denoising, word segmentation, and stop word removal on the first text data to obtain a word sequence;

[0101] S1012. Identify the first entity related to tobacco products and tobacco companies in the word sequence using a named entity recognition algorithm;

[0102] S1013. Based on the first entity, extract key information related to changes in industry regulations and brand packaging requirements from the first text data using keyword extraction or topic modeling.

[0103] S1014. Perform word frequency analysis on key information to determine the importance and frequency of each word;

[0104] S1015, determining the relevance and context of each word based on the order of the words and their positions in the text;

[0105] S1016. Perform sentiment analysis on key information to obtain the ratio of positive sentiment, negative sentiment, and neutral sentiment, and determine emotional attitudes and sentiment changes;

[0106] S1017. Determine the topic distribution of key information, determine the key words and topic weights, construct a topic association diagram, and analyze the association relationships between topics;

[0107] S1018. Determine the specific content of the changes announced by the industry association and the primary influencing factor on tobacco product brand packaging.

[0108] Specifically, the crawler crawls text data from industry association announcements and news reports and performs text preprocessing, including noise removal, word segmentation, and stop word removal. Named entity recognition algorithms are used to identify entities related to the cigarette industry, including tobacco companies and cigarette brands. After identifying relevant entities, keyword extraction or topic modeling is used to extract key information from the text related to changes in industry regulations and product brand and packaging requirements. A word frequency analysis is performed on the extracted key information, using word frequency statistics to obtain word frequency information and determine word importance and frequency of occurrence. Word relevance and context are determined based on word ranking and position in the text. Sentiment analysis is performed on the extracted key information to determine the proportion of positive, negative, and neutral sentiment, and to determine emotional attitudes and changes in sentiment. The extracted key information is then subjected to topic distribution analysis to determine the key words and their weights. A topic association graph is constructed to analyze the relationships between topics. The specific content and timeliness of the association announcement changes, adjustments to product brand and packaging requirements, and their impact on market demand are determined.

[0109] For example, crawl 1000 pieces of text data of industry announcements and news reports. Remove HTML tags, special characters, and non-Chinese characters to obtain clean text data. Use the Jieba word segmentation tool to segment each text to obtain a list of words. Use a stop word list to filter the word segmentation results and remove common words without practical meaning, such as "de" (的), "shi" (是), etc. For example, utilize the named entity recognition algorithm and use tools such as LTP to perform named entity recognition on the word segmentation results to identify entities related to the cigarette industry, such as tobacco enterprises and cigarette brands. Use the TF-IDF algorithm to extract keywords from the text to obtain key information related to changes in industry regulations and product brand packaging requirements, such as "product brand packaging requirements". Word frequency analysis, for example, count the frequency of keywords in the text to obtain word frequency information. For example, the announcement change appears 200 times, and the product brand packaging requirement appears 150 times. Sentiment analysis, for example, use a sentiment analysis algorithm to perform sentiment analysis on the key information. Obtain the proportions of positive sentiment, negative sentiment, and neutral sentiment, such as the positive sentiment accounting for 30%, the negative sentiment accounting for 10%, and the neutral sentiment accounting for 60%. Topic distribution judgment, for example, use a topic model algorithm to perform topic distribution judgment on the key information. Determine the topic words and topic weights, construct an association graph of the topics, and analyze the association relationships between the topics. Based on the word frequency information and the results of the topic distribution analysis, determine the specific content of the changes in industry regulations and the adjustments to product brand packaging requirements, and analyze the association relationships between the topics to judge the impact on timeliness and market demand. For example, based on the keyword "prohibited advertising" and the topic "health warning", infer that the content of the change in industry regulations is about the prohibition of cigarette advertising, which will have a certain negative impact on the market demand of tobacco enterprises in terms of business, such as a sharp drop in sales and product sluggishness.

[0110] S102. Analyze the tobacco tax and tobacco control regulations, and predict the second impact factor of the changes in tax and industry requirements on the brand packaging and publicity strategies of tobacco products.

[0111] Further as an optional implementation manner, analyze the tobacco tax and tobacco control regulations, and predict the second impact factor of the changes in tax and industry requirements on the brand packaging and publicity strategies of tobacco products, which specifically includes:

[0112] S1021. Crawl the attribute data of product brand packaging specifications, brand packaging design features, printing information, brand packaging materials, publicity channels, and publicity content;

[0113] S1022. Regarding the impact on the brand packaging and publicity strategies of tobacco products as the result label according to the second text data of the tobacco tax and tobacco control regulations;

[0114] S1023. Use a random forest algorithm to construct a predictive model for the impact of tax regulation changes on brand packaging and promotional content, and determine how tobacco product brand packaging specifications, design features, materials, promotional channels, and promotional content will change under different tobacco tax and tobacco control regulations.

[0115] S1024. Based on the prediction results, determine the second influencing factor of tobacco taxes and tobacco control regulations on tobacco product brand packaging and promotion strategies.

[0116] Specifically, attribute data on product brand packaging specifications, product brand packaging design features, printing information, product brand packaging materials, promotional channels, and promotional content was crawled. Based on the text of tax and tobacco control regulations, the impact of cigarette product brand packaging and promotion was used as the result label. Using a random forest algorithm, a predictive model was constructed to determine the impact of tax regulation changes on product brand packaging and promotion. This model determined how the design features, materials, promotional channels, and content of cigarette product brand packaging and promotion would change under different tax and tobacco control regulations. Based on the predicted results, the specific impact of tax and tobacco control regulations on cigarette product brand packaging and promotion was determined, including changes in design features, materials, promotional channels, and content.

[0117] For example, suppose we crawled the product brand, packaging, and specification attributes of 100 cigarette brands, yielding the following data. Brand A's attributes are length: 10 cm, width: 5 cm, height: 2 cm; Brand B's attributes are length: 8 cm, width: 4 cm, height: 2 cm; and Brand Z's attributes are length: 12 cm, width: 6 cm, height: 3 cm. Based on industry regulations, the impact of cigarette product brand, packaging, and promotion is used as the result label. For example, suppose the impact of tax rate changes on product brand, packaging, and promotion can be categorized into three levels: high, medium, and low. We randomly selected 10 brands and manually labeled their impact levels. Brand A was ranked medium, Brand B was ranked low, and Brand Z was ranked high. Using the random forest algorithm, we constructed a predictive model for the impact of industry regulation changes on product brand, packaging, and promotion. The features used were product brand, packaging, design features, printing information, product brand, packaging material, promotional channels, and promotional content, and the impact level was used as the label. By training the model, we can predict the impact of product brand, packaging, and promotion on other brands under different tax and industry regulations. For example, using a random forest algorithm to train a model and perform predictions yielded the following impact predictions: Brand K is high, Brand L is medium, and Brand Z is low. Based on these predictions, we can determine how the design features, materials, promotional channels, and content of cigarette brand packaging and promotions will change under different tax and industry requirements. For example, in the industrial and commercial sectors, if changes in tax and industry regulations lead to an increase in brands with a "high" prediction, we can infer that in this scenario, cigarette brand packaging design features may be more unique, materials may be more upscale, promotional channels may use more high-end media, and promotional content may emphasize quality and uniqueness.

[0118] The random forest algorithm is used to determine the impact of tobacco tax rates and tobacco tax revenue on cigarette product brand packaging, and to judge the changing trends of manufacturers' product brand packaging cost pressure and investment under different tax rates.

[0119] Based on tobacco tax regulations and cigarette brand and packaging attributes, obtain tobacco tax rate and cigarette brand and packaging attribute data. Clean and process the data, including addressing missing values, outliers, and duplicate values, to ensure data quality and accuracy. Perform data visualization, perform descriptive statistics and exploratory analysis, and qualitatively analyze the relationships and trends between attributes. Use the random forest algorithm to construct a tobacco tax revenue prediction model based on tobacco tax rate and cigarette brand and packaging attribute data. Predict tobacco tax revenue based on cigarette brand and packaging cost data. Predict the impact of cigarette brand and packaging attributes on tobacco tax revenue based on cigarette sales volume and cigarette brand data. Predict the changing trends of tobacco tax revenue under different tax rates based on cigarette brand and packaging cost data. Based on the model results, determine the changing trends in manufacturers' product brand and packaging cost pressures and investment under different tax rates.

[0120] For example, based on tobacco tax regulations and cigarette brand and packaging attributes, tobacco tax rates and data on cigarette brand and packaging attributes were obtained. Data on 100 cigarette brand and packaging attributes, along with corresponding tobacco tax rates, were obtained from relevant government departments. Data was cleaned and processed, including handling missing values, outliers, and duplicates, to ensure data quality and accuracy. Missing values ​​were found for 10 of these brands, and interpolation was used to fill these missing values. Data visualization was performed, and descriptive statistics and exploratory analysis were performed to qualitatively analyze the relationships and trends between attributes. Histograms and boxplots of tobacco tax rates were plotted to analyze the distribution of tax rates and the presence of outliers. A random forest algorithm was used to construct a tobacco tax revenue prediction model. Using cigarette brand and packaging attributes as feature variables and tobacco tax revenue as the target variable, a random forest model was trained to generate a prediction model. Tobacco tax revenue was predicted based on cigarette brand and packaging cost data. Based on the cigarette brand and packaging cost data and the established prediction model, different product brand and packaging cost data were input to obtain the corresponding predicted tax revenue. Based on cigarette sales volume and brand data, the impact of cigarette brand and packaging attributes on tobacco tax revenue is predicted. By analyzing the relationship between cigarette sales volume and brand data and tobacco tax revenue, the impact of different brand and packaging attributes on tax revenue can be assessed. Based on cigarette brand and packaging cost data, the changing trends of tobacco tax revenue under different tax rates are predicted. For example, by using a model to predict tobacco tax revenue under different tax rates, a trend curve is generated, which can be used to observe the impact of increasing or decreasing tax rates on tax revenue. Based on the model results, the changing trends of manufacturers' product brand and packaging cost pressures and investment under different tax rates are determined. Analysis of the model results reveals that manufacturers face greater cost pressures on product brand and packaging under high tax rates, requiring increased investment to reduce costs.

[0121] S103. Use public opinion analysis tools to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites.

[0122] As an optional implementation method, public opinion analysis tools can be used to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites, including:

[0123] S1031. Crawl comment texts on social media platforms and professional health websites, and collect consumer evaluation texts on tobacco product brand packaging design, information delivery effectiveness, and health warning acceptance;

[0124] S1032. Use the TF-IDF algorithm to extract keywords related to the attractiveness of tobacco product brand packaging, the fit with the brand image, and the degree to which the design style conforms to personal aesthetics;

[0125] S1033, using the TextRank algorithm to determine consumer acceptance of different brands of packaging and health warnings;

[0126] S1034. Use sentiment analysis algorithms to analyze consumers' perceptions of the impact of health warnings on smoking behavior and their level of concern for their own health.

[0127] S1035. Count the frequency of key words in consumers' comments on the environmental friendliness of tobacco product brand packaging, and analyze consumers' recognition of the environmental friendliness of brand packaging materials and their acceptance of environmentally friendly brand packaging.

[0128] Specifically, we crawled social media commentary to collect consumer evaluations of cigarette brand packaging design, messaging effectiveness, and health warning acceptance. We used the TF-IDF algorithm to extract keywords related to consumers' perceived appeal of cigarette brand packaging, its alignment with the brand image, and its design style's alignment with personal aesthetics. We also crawled social media comments about different cigarette brand packaging and used the TextRank algorithm to determine consumer preferences and evaluations of each product, including design, messaging effectiveness, and health warning acceptance. We also crawled discussions on social media and professional health websites to assess consumer acceptance of health warnings. Using sentiment analysis, we analyzed consumers' perceptions of the impact of health warnings on smoking behavior and their perceived concern for their own health. We also counted the frequency of keywords in consumer comments about the environmental friendliness of cigarette brand packaging and analyzed consumers' acceptance of the environmentally friendly packaging materials and their acceptance of environmentally friendly packaging.

[0129] For example, 1,000 reviews of cigarette brand packaging were crawled from Weibo. 500 of them found the packaging design of a certain brand attractive. 400 of them found the packaging to be consistent with the brand image, and 100 of them found the packaging design style to be in line with their personal aesthetics. For example, the TF-IDF algorithm was used to extract keywords related to consumers' perceived appeal of cigarette brand packaging, its consistency with the brand image, and its consistency with their personal aesthetics. Keywords such as "exquisite," "fashionable," and "unique" were extracted from the comments. 1,000 comments were crawled from platforms like Weibo and Tieba. 500 of them found the packaging design of a certain brand attractive. 300 of them found the packaging to be effective in conveying brand information, and 200 of them found the health warnings on the packaging to be highly receptive. For example, the TextRank algorithm was used to determine consumers' preferences and evaluations of different product brand packaging. Keywords such as "beautiful," "attractive," and "concise and clear" were extracted from comment text. Discussion text from social media and professional health websites was crawled, for example, 1,000 texts were crawled from Weibo and the Health Home website. Of these, 400 comments indicated that health warnings had an impact on smoking behavior, and 600 indicated that health warnings had a high level of concern for their health. Sentiment analysis algorithms, such as sentiment lexicon methods, were used to analyze consumers' perceived impact of health warnings on smoking behavior and their perceived level of concern for their health. For example, sentiment analysis revealed that 60% of comments indicated that health warnings had a positive impact on smoking behavior. Keyword frequency analysis was performed to analyze consumer comments on the environmental friendliness of cigarette brand packaging. For example, the frequency of keywords such as "environmentally friendly," "degradable," and "green product brand packaging" was counted from comment text, with "environmentally friendly" appearing 300 times. Consumers' recognition of the environmental friendliness of product brand packaging materials and their acceptance of environmentally friendly product brand packaging were analyzed. For example, word frequency analysis revealed that 30% of comments indicated that product brand packaging materials were environmentally friendly.

[0130] S104. Based on consumers’ online feedback and comments on tobacco product brand packaging, sentiment analysis technology is used to explore consumers’ preferences for different brand packaging design elements.

[0131] As an optional implementation, sentiment analysis technology is used to mine consumer preferences for different brand packaging design elements based on online consumer feedback and comments on tobacco product brand packaging, including:

[0132] S1041. Obtain data containing evaluations and emotional expressions of brand packaging colors based on consumers' online feedback and comments;

[0133] S1042. Analyze consumers' preferences and emotional responses to different colors using a sentiment analysis algorithm to determine consumers' preferences for packaging designs of different colors.

[0134] S1043. Based on consumers' online feedback and comments, obtain data containing evaluations and emotional expressions of brand packaging images and patterns;

[0135] S1044. Analyze consumers' preferences and emotional responses to different images and patterns using a sentiment analysis algorithm to determine consumers' preferences for packaging designs with different images and patterns.

[0136] S1045. Extract data containing evaluations and sentiments about brand packaging fonts and typography based on consumers’ online feedback and comments.

[0137] S1046. Use sentiment analysis algorithms to analyze consumers' preferences and emotional responses to different fonts and layouts, and determine consumers' preferences for packaging designs with different fonts and layouts;

[0138] S1047. Extract data containing evaluations and emotional expressions of brand packaging materials and textures based on consumers' online feedback and comments;

[0139] S1048. Use sentiment analysis algorithms to analyze consumers' preferences and reactions to different materials and textures, and determine consumers' preference information for packaging designs with different materials and textures.

[0140] Specifically, we crawl online consumer feedback and reviews to obtain data containing evaluations and sentimental expressions of product brand packaging colors. Using sentiment analysis algorithms, we determine consumer preferences and emotional responses to different colors, identifying which colors elicit pleasure and excitement, and which colors bring a sense of calm or comfort. We also determine consumer preferences for product brand packaging images and patterns. Based on online user feedback and reviews, we obtain data containing evaluations and sentimental expressions of product brand packaging images and patterns. Using sentiment analysis algorithms, we determine consumer preferences for different images and patterns, identifying which images and patterns capture consumer interest and attention, and which images and patterns provide a positive emotional experience. We crawl online consumer feedback and reviews to extract data containing evaluations and sentimental expressions of product brand packaging fonts and typography. Using sentiment analysis, we determine consumer preferences for different fonts and typography designs, identifying which fonts and typography designs convey specific emotions and styles, and which fonts and typography designs cause distress or discomfort to consumers. Based on online consumer feedback and reviews, we extract data containing evaluations and sentimental expressions of product brand packaging materials and textures. Sentiment analysis algorithms are used to determine consumers' preferences and reactions to different materials and textures, and to determine which materials and textures can bring consumers a sense of comfort and quality, and which materials and textures will make consumers feel cheap or uncomfortable.

[0141] For example, based on crawled online consumer feedback and review data, we obtained the following values. Regarding product brand packaging color preferences and emotional responses, 30% of consumers prefer red product brand packaging, believing that red arouses excitement and joy. 40% of consumers prefer blue product brand packaging, believing that blue brings a sense of calm and comfort. 30% of consumers prefer green product brand packaging, believing that green brings a sense of relaxation and nature. Regarding product brand packaging imagery and patterns, 55% of consumers prefer product brand packaging with animal prints, believing that such patterns attract their interest and attention. 30% of consumers prefer product brand packaging with floral patterns, believing that such patterns create a positive emotional experience. 15% of consumers prefer simple geometric patterns, believing that such patterns convey a specific emotion and style. Regarding product brand packaging font and typography preferences, 55% of consumers prefer rounded and readable fonts, believing that such fonts convey a friendly and professional feeling. 30% of consumers prefer bold fonts, believing that such fonts convey strength and determination. 15% of consumers prefer handwritten fonts, believing they convey creativity and uniqueness. Regarding preferences and reactions to product brand packaging materials and textures, 60% prefer smooth and textured product brand packaging, believing such materials provide a sense of comfort and quality. 40% prefer metallic textures, believing they enhance the product's premium and luxurious feel. By analyzing and analyzing this data, we can understand consumer preferences and emotional reactions to different colors, images, fonts, and materials in the commercial context, helping us design product brand packaging that better meets consumer needs and expectations.

[0142] S105. Analyze consumer demand for brand packaging of different qualities and types based on market sales data, and determine information on the diversification direction of tobacco product brand packaging.

[0143] As an optional implementation method, based on market sales data, consumer demand for brand packaging of different qualities and types is analyzed to determine the diversification direction of tobacco product brand packaging, which specifically includes:

[0144] S1051. Obtaining purchase history data, preference data, and evaluation data of consumers purchasing tobacco products;

[0145] S1052. Obtain consumers' demand for brand packaging of different qualities and types based on their purchase history data and preference data;

[0146] S1053. Based on the tobacco products purchased by consumers and their corresponding brand packaging characteristics, obtain consumers' brand packaging needs regarding tobacco product characteristics and brand image;

[0147] S1054. Based on different consumers' purchase history and evaluation data, use collaborative filtering algorithms to analyze consumers' preferences and relevance for different tobacco product brand packaging attributes, and determine consumers' preferences and requirements for brand packaging attributes such as safety, sustainability, visual appeal, and convenience;

[0148] S1055. Based on consumers' preferences and correlations for the packaging attributes of various tobacco product brands, a collaborative filtering algorithm is used to analyze the tobacco product brand packaging attributes that consumers value most, and determine the diversified direction information for tobacco product brand packaging.

[0149] Specifically, obtain data on consumers' purchase history, preferences, and evaluations of tobacco products. Based on consumers' purchase history and preference data, obtain data on consumers' demand for tobacco product brand packaging of different qualities and types. Based on the tobacco products purchased by consumers and the corresponding product brand packaging characteristics, obtain consumers' demand for tobacco product brand packaging characteristics and brand image. Based on the purchase history and evaluation data of different consumers, use collaborative filtering algorithms to analyze consumers' preferences and correlations for different tobacco product brand packaging attributes, and determine consumers' preferences and requirements for product brand packaging attributes such as safety, sustainability, visual appeal, and convenience. Based on consumers' preferences and correlations for each tobacco product brand packaging attribute, use collaborative filtering algorithms to determine the tobacco product brand packaging attributes that consumers value most, and determine the focus and characteristics of tobacco product brand packaging.

[0150] For example, suppose a consumer purchases the following tobacco products. Brand A cigarettes have a purchase history of 1 pack in January, 2 packs in February, and 3 packs in March. Brand B cigarettes have a purchase history of 2 packs in January, 2 packs in February, and 2 packs in March. Brand C cigarettes have a purchase history of 3 packs in January, 3 packs in February, and 3 packs in March. Based on this consumer's purchase history, their preference for different tobacco brands can be calculated. By calculating the average purchase volume for each brand, the following results are obtained: The average purchase volume for Brand A is 2 packs, the average purchase volume for Brand B is 2 packs, and the average purchase volume for Brand C is 3 packs. The average purchase volume indicates that this consumer has a higher preference for Brand C. Furthermore, suppose a survey is conducted asking this consumer about their preferences for different qualities and types of tobacco product packaging. The following data is obtained from the survey results. On a scale of 10, this consumer gives an industrial safety rating of 8. On an industrial and commercial sustainability rating of 6, this consumer gives an industrial safety rating of 6. In terms of commercial visual appeal, the consumer gave it a rating of 9. In terms of commercial convenience, the consumer gave it a rating of 7. Based on the consumer's ratings of product brand packaging attributes, it can be determined that the consumer has a high preference for visual appeal and safety. Furthermore, a collaborative filtering algorithm, combined with other consumers' purchase history and review data, can be used to analyze consumer preferences and correlations for different tobacco product brand packaging attributes. If other consumers also like Brand C's tobacco products and give it a high rating for visual appeal, it can be inferred that the consumer also has a high preference for visual appeal. Based on the consumer's preferences and correlations for various tobacco product brand packaging attributes, a collaborative filtering algorithm can be used to determine the tobacco product brand packaging attributes that consumers value most. If the consumer has a high preference for both safety and visual appeal, it can be determined that tobacco product brand packaging should prioritize safety and visual appeal.

[0151] Using the K-means clustering algorithm, the brand packaging design of cigarette products at different prices is determined based on the consumption levels of different consumer groups.

[0152] Consumption level data and attribute data related to cigarette brand packaging design are obtained for different consumer groups. The consumption level data for different consumer groups includes consumption amount and purchase frequency, and the attribute data related to cigarette brand packaging design includes color, shape, text, and pattern. The acquired consumption level data and attribute data related to cigarette brand packaging design are cleaned to remove outliers and missing values, and feature selection and scaling are performed on the selected features to ensure that differences between different attributes do not affect the clustering results. A K-means clustering algorithm is used to perform cluster analysis on the consumption level data and attribute data related to cigarette brand packaging design for different consumer groups to determine the number of clusters. Through iterative calculations, the clusters to which each data point belongs are determined. The clustering results are analyzed to determine the consumption level of different consumer groups and their preferences for cigarette brand packaging design based on the cluster centers and member distribution. The reliability of the clustering results is verified by comparing them with actual market data and consumer feedback. If the clustering results are not as expected, adjust the feature selection or K value parameters and rerun the clustering algorithm for analysis.

[0153] For example, consider three different consumer groups, A, B, and C. Data on their spending levels and attributes related to cigarette brand packaging design has been collected. Consumption level data shows that consumer group A spends 200 yuan and purchases cigarettes once a week. Consumer group B spends 500 yuan and purchases cigarettes three times a week. Consumer group C spends 1,000 yuan and purchases cigarettes five times a week. Attribute data related to cigarette brand packaging design shows that consumer group A prefers light-colored packaging, square shapes, and concise text. Consumer group B prefers medium-colored packaging, rectangular shapes, and descriptive text. Consumer group C prefers dark-colored packaging, circular shapes, and trendy text. During data cleaning, check the spending data for outliers, such as negative or excessively large amounts. Also, check the purchase frequency data for missing values, gaps, or unreasonable frequencies. During feature selection, you can select the most relevant features for cluster analysis based on correlation, variance, or other statistical metrics. You can select spending amount and purchase frequency as features for consumption level data, and color, shape, text, and pattern as features for product brand packaging design attributes. Feature scaling normalizes the values ​​of different attributes to ensure that differences between them do not affect the clustering results. You can scale spending amount and purchase frequency values ​​to a range of 0 to 1. When using the K-means clustering algorithm, you need to select the number of clusters, K. You can manually select the K value and run the algorithm multiple times to determine the optimal K value based on the stability and effectiveness of the clustering results. You can try K=2, K=3, and K=4. Then, based on the distribution of cluster centers and membership, analyze the consumption levels and preferences for product brand packaging design among different consumer groups. Finally, you can compare the clustering results with actual market data and consumer feedback for verification. If the clustering results do not meet your expectations, adjust the feature selection or K value parameters and rerun the clustering algorithm for analysis. If the clustering results are unclear or inaccurate, you may need to adjust the feature selection, such as adding more features relevant to consumer behavior or adjusting the K value to better capture the differences between different consumer groups.

[0154] S106. Crawl competitors’ brand packaging design data, use convolutional neural networks to identify brand packaging images, and analyze market brand packaging trends and competitive situations.

[0155] As an optional implementation, we can crawl competitor brand packaging design data, use convolutional neural networks to identify brand packaging images, and analyze market brand packaging trends and competitive situations. This includes:

[0156] S1061. Obtain and label a competitor's brand packaging design image dataset to obtain label data including brand packaging color, shape, pattern, and text;

[0157] S1062. Based on the acquired brand packaging design image dataset and label data, a convolutional neural network algorithm is used to construct a brand packaging image recognition model to determine the brand packaging color, shape, pattern, and text attributes of the new brand packaging image.

[0158] S1063. Divide competitor brand packaging design images into several clusters, with each cluster representing a brand packaging trend or competitive situation. Use the K-means clustering algorithm to construct a brand packaging feature clustering analysis model.

[0159] S1064. Based on the characteristic attributes of the cluster, determine the brand packaging design trends in the market and the brand packaging strategies of competitors.

[0160] Specifically, a dataset of competitor product brand packaging design images is obtained and labeled, including attributes such as product brand packaging color, shape, pattern, and text. Based on this dataset and labels, a convolutional neural network algorithm is used to construct a product brand packaging image recognition model to identify the product brand packaging color, shape, pattern, and text attributes of new product brand packaging images. Competitor product brand packaging design images are grouped into clusters, each representing a product brand packaging trend or competitive situation. A K-means clustering algorithm is then used to construct a product brand packaging feature clustering analysis model. Based on the cluster feature attributes, market product brand packaging design trends and competitor product brand packaging strategies are identified. Model features include the following: Product brand packaging color, including color type, brightness, and saturation. Product brand packaging shape, including square, circular, and rectangular shapes. Pattern, including floral, icon, and striped patterns. Text, including brand name, product features, and instructions. The convolutional layers of the convolutional neural network model extract image features to identify specific image shape and texture features. Adjust the weights of the convolutional neural network model and adjust the structure of the convolutional neural network model to optimize the feature recognition accuracy.

[0161] For example, suppose you have acquired a dataset of 1,000 images of competitor product brand packaging designs and labeled them with attributes such as color, shape, pattern, and text. First, you can use a convolutional neural network algorithm to build a product brand packaging image recognition model. The dataset is divided into a training set and a test set. The model is trained with the training set, and the model's accuracy is evaluated with the test set. Assume that the training results in a product brand packaging image recognition model with an accuracy of 90%. Next, you can group the competitor product brand packaging design images into several clusters, each representing a product brand packaging trend or competitive landscape. You can use the K-means clustering algorithm to group the images into 10 clusters. Each cluster represents a different product brand packaging design trend or competitor's product brand packaging strategy. For example, suppose Cluster 1's characteristic attributes include red, square, floral pattern, and brand name text; while Cluster 2's characteristic attributes include blue, circular, icon pattern, and product feature text. By analyzing the characteristic attributes of each cluster, you can identify market product brand packaging design trends and competitor product brand packaging strategies. Furthermore, you can optimize feature recognition accuracy by adjusting the weights and structure of the convolutional neural network model. For example, you can increase the number of convolutional layers or adjust the size of the convolution kernel to improve the accuracy of the model. Suppose that by adjusting the structure and weights of the model, the accuracy of the product brand packaging image recognition model is increased to 95%.

[0162] Based on competitor product brand packaging design images, a convolutional neural network is used to identify product brand packaging design styles, color and pattern preferences, as well as fonts and logos.

[0163] Obtain a dataset of competitor product brand packaging design images. Based on this dataset, train a convolutional neural network model to annotate different product brand packaging design styles. Using the trained model, predict new product brand packaging design images to determine their product brand packaging design style. Based on the prediction results, identify the attributes of the product brand packaging design style, including minimalist, modern, and retro. Based on the dataset, use a convolutional neural network model to annotate the dataset's different color and pattern preferences. Based on the trained model, predict new product brand packaging design images to determine their color and pattern preferences. Based on the prediction results, identify the color and pattern attributes of the product brand packaging design, including bright colors, geometric patterns, and floral patterns. Based on the dataset, use a convolutional neural network model to annotate different fonts and logos. Based on the trained model, predict new product brand packaging design images to determine their fonts and logos, including different font styles, trademarks, or logos. Based on the recognition results, identify the competitor's product brand packaging design strategy.

[0164] For example, a dataset of 1,000 images of a competitor's product brand packaging design is obtained. This dataset is used to train a convolutional neural network model to predict the style of a new product brand packaging design image. Before training, the dataset needs to be annotated, meaning each image is assigned a label. Suppose product brand packaging design styles are categorized into three categories: minimalist, modern, and retro. Each image is assigned a label; for example, minimalist images are labeled 0, modern images are labeled 1, and retro images are labeled 2. After annotation, a convolutional neural network model is trained. A classic convolutional neural network architecture with convolutional, pooling, and fully connected layers is selected. Assume that 80% of the image data is used for training, and the remaining 20% ​​is used for model validation. After multiple iterations of training, a model with an accuracy of 85% is obtained. This model is then used to predict the style of a new product brand packaging design image. For example, a new product brand packaging design image is predicted to be modern in style by the model. Based on this prediction, the stylistic attributes of this product brand packaging design are determined to be modern. Similarly, the same dataset can be used to train a convolutional neural network model to predict the color and pattern preferences of product brand packaging designs. Assume that color and pattern preferences are categorized into three categories: bright colors, geometric patterns, and floral patterns. Each image in the dataset is assigned a corresponding label. For example, images with bright colors are labeled 0, images with geometric patterns are labeled 1, and images with floral patterns are labeled 2. Similarly, 80% of the data is used for training, and the remaining 20% ​​is used to validate the model's accuracy. After multiple iterations of training, a model with an accuracy of 90% is obtained. This model is used to predict the color and pattern preferences of new product brand packaging images. Assume that the prediction results indicate a bright color preference and a geometric pattern preference. Based on this prediction, the color attributes of this product brand packaging design are determined to be bright, and the pattern attributes are determined to be geometric. Finally, the font and logo choices of competitors can be analyzed based on their product brand packaging design strategies. Similarly, a convolutional neural network model can be used for training and prediction to determine the font and logo used in product brand packaging designs. Suppose fonts and logos are categorized into three categories: different font styles, trademarks, or logos. Each image in the dataset is assigned a corresponding label, for example, 0 for font style and 1 for trademarks or logos. Similarly, 80% of the data is used for training, and the remaining 20% ​​is used to validate the model's accuracy. After multiple iterations of training, a model with 95% accuracy is obtained. Now, when presented with a new image of a product brand packaging design, this model is used to predict its font and logo.Assume that after prediction, the font style is obtained as 0 and the trademark or logo is 1. Based on the prediction results, it is determined that the competitor has selected a font style with the number 0 and a trademark or logo with the number 1 in the product brand packaging design.

[0165] The K-means clustering algorithm is used to cluster the identified attributes and determine the popular product brand packaging trends in the market, and to judge the mainstream styles and characteristics of product brand packaging designs in the current market.

[0166] Based on the color, pattern, font, and material attributes of the product brand packaging design, K cluster centers are selected as initial values. The initial cluster centers are determined by random selection or using a genetic algorithm. Based on the distance between each sample and each cluster center, the sample is assigned to the cluster closest to it. The Euclidean distance or Manhattan distance between the sample and the cluster center is calculated to determine the cluster assignment method for the sample. For each cluster, the average value of all samples is calculated as the new cluster center. Based on the attribute values ​​of the samples within the cluster, the average value is calculated to update the cluster center. The genetic algorithm is used to optimize the cluster center until the cluster center and cluster division no longer change or the preset number of iterations is reached. The samples are reallocated to the clusters based on the new cluster centers, and the cluster centers are updated. Based on the color, pattern, font, and material attributes of the product brand packaging design, the final cluster analysis results are determined, including the specific cluster division and the cluster center of each cluster. Based on the common characteristics of the samples in the cluster, the popular product brand packaging trends in the market are identified, including popular colors, prominent patterns, and popular fonts.

[0167] For example, suppose there are 100 product brand packaging samples, each with color, pattern, font, and material attributes. We want to divide these samples into K clusters and determine the cluster centers for each cluster. First, we need to select K initial cluster centers. We can use random selection or a genetic algorithm to determine these initial values. Assume that three initial cluster centers are selected. Then, we calculate the distance between each sample and these cluster centers. Suppose that the Euclidean distance between sample 1 and cluster center 1 is 5, the Euclidean distance between sample 2 and cluster center 3 is 3, and the Euclidean distance between sample 1 and cluster center 3 is 6. Based on the closest cluster center, sample 1 is assigned to cluster 2. Next, we calculate the average value of all samples in each cluster as the new cluster center. Suppose that cluster 2 contains samples 1, 2, and 3, and their color attributes are red, green, and blue, respectively. The new cluster centers are (red + green + blue) / 3, or (1 + 2 + 3) / 3 = 2. We then use a genetic algorithm to optimize the cluster centers. We iterate until the cluster centers and the division into clusters no longer change, or until a preset number of iterations is reached. Finally, samples are reassigned to clusters based on the new cluster centers and the cluster centers are updated, resulting in a specific cluster division and cluster centers for each cluster. Based on the common characteristics of the samples within the clusters, popular product brand packaging trends in the market are determined. If the samples in Cluster 1 all feature bright colors, bold patterns, and stylish fonts, then we can conclude that popular product brand packaging trends in the market are bright colors, bold patterns, and stylish fonts.

[0168] S107. Based on advertising campaigns, tobacco product brand packaging designs, and tobacco product sales data, determine the relationship curve between sales performance and different brand packaging and different promotional strategies.

[0169] As a further optional implementation, based on advertising campaigns, tobacco product brand packaging designs, and tobacco product sales data, a relationship curve between sales performance and different brand packaging and different promotional strategies is determined, which specifically includes:

[0170] S1071. Obtain sales data and brand packaging type information, including brand packaging type, sales performance, and brand;

[0171] S1072. Through data preprocessing, encode the sales data and brand packaging types and perform association rule mining;

[0172] S1073. Construct frequent item sets of different brand packaging types and sales performance. The frequent item sets represent the association between different brand packaging and sales performance.

[0173] S1074. Construct association rules and extract association rules from the frequent item set according to the confidence threshold.

[0174] S1075. Evaluate the importance of association rules based on support and confidence, and select association rules with practical significance;

[0175] S1076. Determine the relationship curve between sales performance and different brand packaging and different promotional strategies based on the obtained association rules.

[0176] Specifically, sales data and information on product brand and packaging types are obtained, including product brand and packaging type, sales performance, and brand. Through data preprocessing, the sales data and product brand and packaging type are encoded and then association rule mining is performed. Frequent item sets are constructed, and based on a support threshold, frequent item sets for different product brand and packaging types and sales performance are calculated. Frequent item sets represent the associations between different product brand and packaging types and sales performance. Association rules are constructed, and based on a confidence threshold, association rules are extracted from the frequent item sets. The importance of association rules is evaluated based on support and confidence, and meaningful association rules are screened out. Based on the resulting association rules, the relationship between different product brand and packaging types and sales performance is determined.

[0177] For example, the sales data of a cigarette company includes different brands of cigarette products, different product brand packaging types, and corresponding sales performance. Association rule mining is performed according to the following steps: Determine the relevant attributes: product brand packaging type, sales performance, and brand. Crawl the sales data of cigarette products, including indicators such as sales volume and sales revenue, as well as the product brand packaging type and brand corresponding to each product. Encode the sales data and product brand packaging type and convert them into a format suitable for association rule mining. For example, use one-hot encoding to convert the product brand packaging type into a binary feature vector. Construct frequent item sets and calculate the frequent item sets of different product brand packaging types and sales performance based on the set support threshold. Frequent item sets are combinations of product brand packaging types and sales performance that frequently appear in sales data, reflecting the association relationship between them. Construct association rules: Extract association rules from the frequent item sets based on the set confidence threshold. Association rules are a formal expression that describes the relationship between product brand packaging type and sales performance. For example, if a product uses a high-end product brand packaging type, then sales performance is more likely to improve. Evaluate the importance of association rules: Evaluate the importance of association rules through support and confidence. Support reflects the frequency of the association rule in the entire sales data, while confidence indicates the likelihood of sales performance when a given product brand packaging type is used. Filter out association rules with practical significance: Based on the evaluation results, filter out association rules with practical significance. For example, select rules with high support and confidence, or rules that are related to business objectives. In business, if the association rules reveal a strong correlation between high-end product brand packaging types and sales performance, consider using high-end product brand packaging in your products to improve sales performance.

[0178] In some optional embodiments, association rule mining is used to determine whether there is a certain synergistic effect based on the correlation between advertising activities and product brand packaging design, thereby jointly promoting the improvement of sales performance.

[0179] Specifically, data on advertising campaigns and product brand packaging designs are obtained, including advertising channels, advertising types, advertising content, advertising duration, advertising investment, product brand packaging appearance, product brand packaging materials, product brand packaging logos, product brand packaging dimensions, and product brand packaging characteristics. The data is cleaned, deduplicated, and missing values ​​are processed. The association rule mining Apriori algorithm is used to mine association rules on the preprocessed data. Appropriate minimum support and minimum confidence thresholds are set to filter out eligible association rules. The mined association rules are evaluated, and the confidence and support evaluation indicators of each rule are calculated. Association rules with high confidence and the greatest support are selected as the rules of focus. Based on the mining results and evaluation indicators, text is generated to explain the meaning of the association rules and the underlying correlation relationships.

[0180] For example, suppose a company's marketing department has obtained data on advertising campaigns and product brand packaging designs over a period of time. The following data is obtained: Advertising channels include television, radio, outdoor advertising, the internet, magazines, newspapers, and other advertising channels. Advertising types include product promotion, brand promotion, promotional activities, and public service announcements. Advertising content includes specific advertising copy or visual descriptions. Ad duration is expressed in seconds, such as 30 seconds or 60 seconds. Advertising investment is expressed in 10,000 yuan, representing the company's investment in the advertisement. Product brand packaging appearance refers to the form of the product brand packaging, such as boxed, bottled, or bagged. Product brand packaging material refers to the material used for the product brand packaging, such as paper boxes, plastic bottles, or cans. Product brand packaging logo refers to the relevant information indicated on the product brand packaging, such as the product name, brand logo, and production date. Product brand packaging dimensions refer to the dimensions of the product brand packaging, such as the specific values ​​for length, width, and height. Product brand packaging attributes refer to the characteristics or properties of the product brand packaging, such as moisture resistance, shock resistance, and ease of opening. Data cleaning is performed to remove duplicate data and address missing values. For example, for the advertising investment item, if a data entry is missing the investment amount, the data entry can be filled with the average or discarded. The Apriori algorithm for association rule mining is used to analyze the data. Assume that the minimum support is set to 3 and the minimum confidence is set to 8. The goal is to mine frequent itemsets and association rules to understand the relationship between advertising activities and product brand packaging design. After the algorithm runs, several association rules are obtained. For example, the rule "High Advertising Investment -> Advertising Channel: Television" has a confidence of 9 and a support of 4. This means that when advertising investment is high, television advertising is more likely to be chosen. The rule "Product Brand Packaging Material: Paper Box" -> Product Brand Packaging Appearance: Boxed" has a confidence of 8 and a support of 3. This indicates that when paper box is selected as the packaging material, the product brand is more likely to choose a boxed appearance. Analysis of the mining results and evaluation metrics shows that when advertising investment is high, television advertising is more likely to be chosen, while when paper box is selected as the packaging material, the product brand is more likely to choose a boxed appearance.

[0181] In some optional embodiments, an Apriori algorithm is used to design permutations of tobacco products of different prices on a tobacco display counter to maximize tobacco sales.

[0182] Specifically, obtain tobacco product sales data, including product price, brand, type, product brand packaging, and promotional information. Ensure data accuracy and consistency through data cleaning and deduplication. Analyze sales data using the Apriori algorithm to identify frequent item sets and association rules. Set thresholds based on support and confidence levels to determine the importance of frequent item sets and association rules. Based on the goal of maximizing sales, select several association rules with the highest sales. Based on the selected association rules, design permutations and combinations of tobacco products at different prices. Evaluate the effectiveness of the designed tobacco display counters and make adjustments and optimizations based on consumer feedback and changes in sales data.

[0183] For example, the following sales data was obtained from a tobacco shop. The product prices are [10 yuan, 20 yuan, 15 yuan, 25 yuan, 30 yuan, 20 yuan], the brands are [Brand A, Brand B, Brand C, Brand A, Brand B, Brand C], the categories are [Cigarettes, Cigarettes, Cigarettes, Cigarettes, Cigarettes], the product brand packaging is [Boxed, Boxed, Bagged, Bagged, Boxed, Boxed], and the promotions are [No promotion, No promotion, Discount, Discount, No promotion, Discount]. First, the data needs to be cleaned and deduplicated. The cleaned data is as follows: the product prices are [10 yuan, 20 yuan, 15 yuan, 25 yuan, 30 yuan], the brands are [Brand A, Brand B, Brand C], the categories are [Cigarettes, Cigarettes], the product brand packaging is [Boxed, Bagged], and the promotions are [No promotion, Discount]. Next, the Apriori algorithm is used for analysis to identify frequent itemsets and association rules. Setting the support threshold to 2 and the confidence threshold to 5, the resulting frequent itemsets and association rules are as follows: the frequent itemsets are {Product Price: 20 Yuan, Brand: Brand A, Category: Cigarettes, Product Brand Packaging: Boxed}, {Product Price: 20 Yuan, Category: Cigarettes, Product Brand Packaging: Boxed, Promotion: No}, and the association rules are {Product Price: 20 Yuan, Brand: Brand A} -> {Cigarettes, Product Brand Packaging: Boxed}, and {Product Price: 20 Yuan, Category: Cigarettes} -> {Product Brand Packaging: Boxed, Promotion: No}. Based on the goal of maximizing sales, we can filter out the association rules that yield the highest sales. The formula for calculating sales is: Sales = Product Price * Sales Quantity. Based on the sales data and association rules, we can calculate the following sales: Sales for Association Rule 1 = 20 Yuan * 2 = 40 Yuan, and Sales for Association Rule 2 = 20 Yuan * 2 = 40 Yuan. Therefore, both Association Rules 1 and 2 yield the highest sales. Based on these filtered association rules, we can design permutations and combinations of tobacco products at different prices. The price of the product in association rule 1 can be set to 25 yuan, and the price of the product in association rule 2 can be set to 30 yuan to increase sales. Finally, the effectiveness of the designed tobacco display counter can be evaluated. Based on consumer feedback and changes in sales data, the price, brand, variety, product brand packaging, and promotional activities of the product can be adjusted and optimized to increase sales and meet consumer needs.

[0184] S108. Based on the publicity effect and market feedback of tobacco product brand packaging, use multiple regression analysis and prediction models to evaluate the third factor affecting tobacco product brand packaging and publicity strategies on market sales.

[0185] As an optional implementation method, based on the promotional effect of tobacco product brand packaging and market feedback, a multivariate regression analysis and prediction model is used to evaluate the third factor affecting tobacco product brand packaging and promotional strategy on market sales, which specifically includes:

[0186] S1081. Obtain data related to tobacco product brand packaging design and color, including sales data, market research, and consumer feedback and comment text for different designs and color schemes;

[0187] S1082. Use multiple regression analysis to determine the primary influence coefficient of brand packaging design and color on market sales;

[0188] S1083. Obtain relevant data on packaging materials and quality of different brands, including consumer evaluations of brand packaging materials and quality and product protection performance;

[0189] S1084. Use multiple regression analysis to determine the second-order influence coefficient of brand packaging material and quality on market sales volume;

[0190] S1085. Obtain relevant data on different promotional channels and content, including exposure across different channels and content, and consumer responses to the promotional content;

[0191] S1086. Use multiple regression analysis to determine the third-order influence coefficient of promotional channels and promotional content on market sales volume;

[0192] S1087. Obtain data on the target audience's response to brand packaging and promotional strategies, including consumer purchase intention and brand awareness;

[0193] S1088. Use multiple regression analysis to determine the fourth coefficient of influence of target audience response on market sales volume;

[0194] S1089. Determine the third influencing factor of tobacco product brand packaging and promotional strategy on market sales based on the first influencing coefficient, the second influencing coefficient, the third influencing coefficient, and the fourth influencing coefficient.

[0195] Specifically, data related to cigarette brand packaging design and color is obtained, including sales data, market research, and consumer feedback and commentary for different design and color schemes. Based on this data, multiple regression analysis is used to determine market sales. The specific impact of brand packaging design and color on market sales is determined by testing the significance and explanatory power of the regression coefficients. Data related to the materials and quality of different brand packaging is obtained, including consumer evaluations of the materials and quality of the brand packaging and product protection. Multiple regression analysis is used to determine the coefficient of influence of the brand packaging materials and quality on market sales. The coefficient of influence of the brand packaging materials and quality on market sales is determined by testing the significance and explanatory power of the regression coefficients. Data related to different promotional channels and content is obtained, including exposure to each channel and content, as well as consumer responses to the promotional content. Multiple regression analysis is used to determine the coefficient of influence of the promotional channels and content on market sales. The actual specific impact of the promotional channels and content on market sales is determined by testing the significance and explanatory power of the regression coefficients. Data on the target audience's response to the brand packaging and promotional strategies is obtained, including consumer purchase intention and brand awareness. Through multiple regression analysis, we analyze data to determine the specific impact coefficient of the target audience's response on market sales. Through the significance test and explanatory power of the regression coefficient, we can determine the actual specific impact of the target audience's response on market sales.

[0196] For example, suppose there are two different product brand packaging designs and color schemes, Plan A and Plan B. A regression analysis is conducted on the sales data for the two schemes, with the product brand packaging design and color as the independent variables and market sales as the dependent variable. By testing the significance and explanatory power of the regression coefficients, the specific impact of the product brand packaging design and color on market sales can be determined. Suppose the regression analysis results show that Plan A's product brand packaging design and color have a more significant impact on market sales, with a regression coefficient of 8, while Plan B's regression coefficient is only 3. This means that Plan A's product brand packaging design and color are more popular with consumers and have a greater impact on market sales. Similarly, by collecting data on the packaging materials and quality of different product brands, a multiple regression analysis can be conducted to determine the specific impact of these materials and quality on market sales. If the analysis results show that high-quality product brand packaging materials have a significant positive impact on market sales, with a regression coefficient of 11, while low-quality product brand packaging materials have a regression coefficient of only 2, it can be concluded that high-quality product brand packaging materials have a greater impact on market sales. Similarly, by collecting data on different promotional channels and content, a multiple regression analysis can be conducted to determine the specific impact of these channels and content on market sales. Suppose the analysis results show that, in television advertising, promotional content A has the greatest impact on sales, with a regression coefficient of 9, while in social media, promotional content B has a regression coefficient of 6. Therefore, we can conclude that the impact of promotional channel and content on sales differs. Furthermore, we can conduct surveys and analyses targeting different target audiences to collect data on their responses to product branding, packaging, and promotional strategies, and determine the impact of these responses on sales. In a business context, if the analysis results show that young people are more interested in Option A's product branding, packaging, and promotional content, with a regression coefficient of 7, while middle-aged people are more interested in Option B, with a regression coefficient of 5, we can conclude that different target audiences' responses have different impacts on sales. If product branding, packaging, and color have a greater impact on sales, we recommend optimizing the product branding, packaging, and color scheme. If promotional channel and content have different impacts on sales, we recommend adjusting the promotional strategy to better appeal to different target audiences.

[0197] In some optional embodiments, a support vector machine algorithm is used to automatically identify and mine hot topics related to the tobacco industry, as well as keywords of product brand packaging and promotional strategies.

[0198] Specifically, a text dataset related to the tobacco industry, including news reports, social media comments, and industry reports, is obtained. Each sample in the dataset is labeled to indicate its relevance to the industry. The text dataset is converted into feature vectors using TF-IDF weights that can be used for training a support vector machine algorithm. The extracted feature vectors are standardized or normalized to ensure that each feature has a similar scale. The dataset is divided into a training set and a test set, and a support vector machine algorithm is used to train the model on the labeled training data. Model performance is optimized by adjusting hyperparameters and using cross-validation methods. The trained model is evaluated using the test dataset, and precision and recall metrics are calculated. Based on the model training results, information gain or chi-square tests are used to identify the most influential features for topics related to the tobacco industry, as well as keywords related to product brand packaging and promotional strategies. The trained model is used to make predictions on a new text dataset, determine its relevance to the tobacco industry, and identify trending topics and keywords related to product brand packaging and promotional strategies.

[0199] For example, consider a dataset containing 1,000 text samples, each labeled to indicate its relevance to the tobacco industry. These samples can come from news articles, social media comments, and industry reports. Next, use TF-IDF weights to convert the text dataset into feature vectors. By calculating the TF-IDF weights for each word in each text sample, the text data can be converted into numerical feature vectors. The extracted feature vectors can then be standardized or normalized to ensure that the features have similar scales. Next, the dataset is divided into a training set and a test set. The training set is used to train the support vector machine (SVM) model, while the test set is used to evaluate the performance of the trained model. The SVM algorithm can then be used to train the model on the labeled training data. During model training, model performance can be optimized by adjusting hyperparameters and using cross-validation. Different kernel functions, regularization parameters, and penalty parameters can be tried, and cross-validation can be used to select the optimal parameter combination. The trained model can then be evaluated using the test dataset. Metrics such as precision and recall can be calculated to measure model performance. The model achieved an accuracy of 80% and a recall of 85% on the test dataset. Based on the model training results, information gain or chi-square tests can be used to identify keywords related to tobacco industry topics, as well as product branding, packaging, and promotional strategies. These keywords can help understand the industry's hot topics and key strategies. Using the chi-square test, the keywords "health risks" and "marketing" were identified. Finally, the trained model can be used to predict new text datasets, determine their relevance to the tobacco industry, and identify keywords related to hot topics and key strategies. A new text sample can be input into the trained model, which will output its relevance to the tobacco industry. Keyword extraction methods can also be used to identify keywords related to hot topics and key strategies.

[0200] S109. Determine the target brand packaging and target promotion strategy for tobacco products based on acceptance information, preference information, diversification direction information, market brand packaging trends and competitive situation, relationship curves, first influencing factors, second influencing factors, and third influencing factors.

[0201] Specifically, based on market research and consumer preference data, obtain data on their preferences for tobacco product appearance, materials, functions, and convenience. Based on consumer preference data, determine the tobacco product brand packaging design that is most similar to their preference data. Use environmentally friendly materials to create product brand packaging, emphasizing the sustainability of the product to meet consumer demand for environmentally friendly products. Compare the content and format of the labels on the product brand packaging materials in the proposal with industry regulations. Use a unique product brand packaging design and promotional language to highlight the differentiation of the product proposal from competitors. Based on the differentiation results, optimize the tobacco product promotion strategy to highlight the value and benefits of the product and convey the product's characteristics and functions to consumers. Determine the final optimized tobacco product brand packaging design and promotional method plan.

[0202] For example, market research data reveals that consumers' preferences for tobacco product appearance are 50% for minimalist designs, 30% for ornate designs, and 20% for innovative designs. Based on this consumer preference data, it can be determined that the tobacco product brand packaging design with the highest similarity to these preferences is minimalist. Regarding product brand packaging materials, environmentally friendly practices require the use of sustainable materials, such as biodegradable paper boxes. Industry regulations require that labels on product brand packaging materials include the product name, supplier information, ingredients, and health warnings. Labels should comply with specified font, size, and color requirements. Regarding differentiated design and promotional methods, suppose competitors' product brand packaging designs favor ornate and innovative styles, while your product brand packaging design opts for minimalism. Bold fonts and colors can be used on the product brand packaging to highlight its simplicity and high quality. Furthermore, unique promotional language can be used to emphasize the product's uniqueness and differentiation from competitors. For example, the promotional language could be "Simple yet not simplistic, unique taste," highlighting the product's minimalist style and distinctive characteristics. By optimizing tobacco product promotional strategies, consumers can be conveyed the product's value and benefits, such as reduced environmental impact and a comfortable smoking experience. Through a collaborative approach between industry and commerce, the ultimate optimized tobacco product brand packaging design and promotional approach could include a minimalist design, using environmentally friendly materials, and ensuring that label content and format comply with industry regulations. Furthermore, through unique promotional language and differentiated design, the uniqueness and value of the product can be highlighted, meeting consumer demand for environmentally friendly products.

[0203] The above describes the method steps of the embodiment of the present invention. It can be recognized that the embodiment of the present invention crawls text data such as industry association announcements and news reports, uses text mining technology to extract attributes related to cigarette product brand packaging requirements, and analyzes the specific impact of these attributes on product brand packaging requirements; through the analysis of tobacco taxes and tobacco control regulations, uses the random forest algorithm to predict the impact of taxes and industry regulations on cigarette product brand packaging and publicity; uses public opinion analysis tools to analyze consumers' acceptance of cigarette product brand packaging and health warnings through discussions on social media and professional health websites; conducts sentiment analysis on consumers' online feedback and comments on product brand packaging, and explores consumers' preferences for different product brand packaging design elements; uses market sales data and collaborative filtering algorithms to analyze consumers' demand for product brand packaging of different flavors, qualities and types, and determine product brand packaging. Diversification direction: By monitoring competitors' product brand packaging design and promotional strategies, convolutional neural networks are used to identify product brand packaging design images, and the K-means clustering algorithm is used to analyze popular product brand packaging trends and competitive situations in the market. Association rule mining is used to analyze the relationship between advertising, sales performance, and different product brand packaging types. Multiple regression analysis and prediction models are used to evaluate the specific impact of product brand packaging and promotional strategies on market sales. The Apriori algorithm is used to design different price combinations of tobacco products on tobacco display counters to maximize tobacco sales. Based on consumer preferences, market trends, and industry regulations, product brand packaging and promotional strategies are optimized, improving the accuracy and reliability of tobacco product brand packaging and promotional strategy formulation, thereby enhancing the market competitiveness of tobacco companies and tobacco products. From the unique perspective of analyzing product brand packaging and promotional strategies in the cigarette industry, the embodiments of the present invention, through a collaborative model between industry and commerce, increase cigarette retail customer revenue and consumer satisfaction, further promote tobacco tax revenue, and enhance the image of tobacco companies in society.

[0204] Reference Figure 2 The embodiment of the present invention provides a tobacco product brand packaging and promotion strategy determination system based on data mining, including:

[0205] The first impact factor determination module is used to crawl negative evaluation information of the cigarette industry based on industry association announcements and news, extract key information related to brand packaging requirements based on the negative evaluation information, and analyze the first impact factor of tobacco product brand packaging based on the key information;

[0206] The second impact factor determination module is used to analyze tobacco taxation and tobacco control regulations and predict the second impact factor of changes in taxation and industry requirements on tobacco product brand packaging and promotional strategies;

[0207] An acceptance information determination module, which uses public opinion analysis tools to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites;

[0208] The preference information determination module is used to mine consumer preferences for different brand packaging design elements based on online consumer feedback and comments on tobacco product brand packaging using sentiment analysis technology;

[0209] A diversification direction information determination module is used to analyze consumer demand for brand packaging of different qualities and types based on market sales data, and determine diversification direction information for tobacco product brand packaging;

[0210] The module for determining market brand packaging trends and competitive situations is used to crawl competitors' brand packaging design data, use convolutional neural networks to identify brand packaging images, and analyze market brand packaging trends and competitive situations.

[0211] A relationship curve determination module is used to determine the relationship curve between sales performance and different brand packaging and different promotional strategies based on advertising activities, tobacco product brand packaging design and tobacco product sales data;

[0212] The third influencing factor determination module is used to evaluate the third influencing factor of tobacco product brand packaging and promotional strategies on market sales based on the promotional effects of tobacco product brand packaging and market feedback, using multiple regression analysis and prediction models;

[0213] The brand packaging and promotion strategy determination module is used to determine the target brand packaging and target promotion strategy of tobacco products based on acceptance information, preference information, diversification direction information, market brand packaging trends and competitive situation, relationship curves, first influencing factors, second influencing factors and third influencing factors.

[0214] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0215] Reference Figure 3 The embodiment of the present invention provides a device for determining tobacco product brand packaging and promotional strategies based on data mining, comprising:

[0216] at least one processor;

[0217] at least one memory for storing at least one program;

[0218] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned method for determining tobacco product brand packaging and promotion strategies based on data mining.

[0219] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0220] An embodiment of the present invention also provides a computer-readable storage medium storing a program executable by a processor. When executed by the processor, the program executable by the processor is used to execute the above-mentioned method for determining tobacco product brand packaging and promotional strategies based on data mining.

[0221] A computer-readable storage medium in an embodiment of the present invention can execute a method for determining tobacco product brand packaging and promotional strategies based on data mining provided by an embodiment of the method of the present invention, can execute any combination of implementation steps of the method embodiment, and has the corresponding functions and beneficial effects of the method.

[0222] The embodiment of the present invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs Figure 1 The method shown.

[0223] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the above-mentioned boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operation and logic flow presented herein. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0224] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the above-mentioned functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present invention. More specifically, given the properties, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the routine skills of an engineer. Therefore, a person skilled in the art can implement the present invention set forth in the claims using ordinary skills without undue experimentation. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0225] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium 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 above methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0226] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0227] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable media on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0228] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0229] In the above description of this specification, reference to the terms "one embodiment / example," "another embodiment / example," or "certain embodiments / examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0230] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

[0231] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for determining tobacco product brand packaging and promotion strategy based on data mining, characterized in that: The following steps are involved: According to the announcements and news of industry associations, crawl the negative evaluation information of the cigarette industry, extract the key information related to the brand packaging requirements based on the negative evaluation information, and analyze the first influencing factor on the brand packaging of tobacco products based on the key information; Analyze tobacco taxation and tobacco control regulations to predict the secondary impact of changes in taxation and industry requirements on tobacco product brand packaging and promotional strategies; Use public opinion analysis tools to analyze consumer acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites; Based on consumers’ online feedback and comments on tobacco product brand packaging, sentiment analysis technology is used to mine information about consumers’ preferences for different brand packaging design elements; Analyze consumers’ demand for brand packaging of different qualities and types based on market sales data, and determine the diversification direction of tobacco product brand packaging; Crawl competitor brand packaging design data, use convolutional neural networks to identify brand packaging images, and analyze market brand packaging trends and competitive situations; Based on advertising campaigns, tobacco product brand packaging designs and tobacco product sales data, determine the relationship curve between sales performance and different brand packaging and different promotional strategies; Based on the publicity effect of tobacco product brand packaging and market feedback, multiple regression analysis and prediction models are used to evaluate the third influencing factor of tobacco product brand packaging and publicity strategy on market sales; The target brand packaging and target promotion strategy of tobacco products are determined based on the acceptance level information, the preference level information, the diversification direction information, the market brand packaging trend and competition situation, the relationship curve, the first influencing factor, the second influencing factor and the third influencing factor.

2. A method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: According to the announcements and news of the industry association, the negative evaluation information of the cigarette industry is crawled, the key information related to the brand packaging requirements is extracted according to the negative evaluation information, and the first influencing factor on the brand packaging of tobacco products is analyzed according to the key information, which specifically includes: Crawling first text data of industry association announcements and news reports, denoising, segmenting, and removing stop words from the first text data to obtain a word sequence; Using a named entity recognition algorithm to identify the first entity related to tobacco products and tobacco companies in the word sequence; According to the first entity, extracting the key information related to the industry regulation changes and brand packaging requirements from the first text data by using keyword extraction or topic modeling; Performing word frequency analysis on the key information to determine the importance and frequency of occurrence of each word; Determine the relevance and context of each word based on its order and position in the text; Performing sentiment analysis on the key information to obtain the ratio of positive sentiment, negative sentiment and neutral sentiment, and determining the sentiment attitude and sentiment change; Determine the topic distribution of the key information, determine the topic words and topic weights, construct a topic association diagram, and analyze the association relationship between topics; Determine the specific content of the changes announced by the industry association and the first influencing factor on tobacco product brand packaging.

3. The method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: The analysis of tobacco taxation and tobacco control regulations predicts the second impact factor of changes in taxation and industry requirements on tobacco product brand packaging and promotion strategies, which specifically includes: Crawl the attribute data of product brand packaging specifications, brand packaging design features, printing information, brand packaging materials, promotion channels, and promotion content; Based on the second text data of tobacco tax and tobacco control regulations, the impact on tobacco product brand packaging and promotion strategies is used as the outcome label; A random forest algorithm was used to construct a prediction model of the impact of tax regulation changes on brand packaging promotion, to determine the changes in packaging specifications, design features, materials, promotion channels, and promotional content of tobacco product brand packaging under different tobacco tax and tobacco control regulations; Based on the prediction results, the second influencing factor of tobacco tax and tobacco control regulations on tobacco product brand packaging and promotion strategies is determined.

4. The method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: The public opinion analysis tool is used to analyze consumers' acceptance of tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites, including: Crawl comment texts on social media platforms and professional health websites, and collect consumers’ evaluation texts on tobacco product brand packaging design, information delivery effectiveness, and health warning acceptance; The TF-IDF algorithm is used to extract keywords related to the attractiveness of tobacco product brand packaging, the fit of brand image, and the degree of conformity of design style with personal aesthetics; Using the TextRank algorithm, we determined consumers’ acceptance of different brands’ packaging and health warnings. Use sentiment analysis algorithms to analyze consumers’ perceptions of the impact of health warnings on smoking behavior and their level of concern for their own health; Statistics are collected on the frequency of key words in consumers’ comments on the environmental friendliness of tobacco product brand packaging, and information on consumers’ recognition of the environmental friendliness of brand packaging materials and their acceptance of environmentally friendly brand packaging is analyzed.

5. The method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: Based on consumers' online feedback and comments on tobacco product brand packaging, sentiment analysis technology is used to mine information on consumers' preference for different brand packaging design elements, which specifically includes: Obtain data containing evaluations and sentiments about brand packaging colors based on consumers’ online feedback and reviews; Use sentiment analysis algorithms to analyze consumers’ preferences and emotional responses to different colors, and determine consumers’ preferences for packaging designs of different colors; Obtain data containing evaluations and sentiments about brand packaging images and graphics based on consumers’ online feedback and reviews; Use sentiment analysis algorithms to analyze consumers’ preferences and emotional responses to different images and patterns, and determine consumers’ preferences for packaging designs with different images and patterns; Extract data containing evaluations and sentiments about brand packaging fonts and typography based on consumers’ online feedback and reviews; Use sentiment analysis algorithms to analyze consumers’ preferences and emotional responses to different fonts and layouts, and determine consumers’ preferences for packaging designs with different fonts and layouts; Extract data containing evaluations and emotional expressions of brand packaging materials and textures based on consumers’ online feedback and comments; Sentiment analysis algorithms are used to analyze consumers’ preferences and reactions to different materials and textures, and to determine consumers’ preference information for packaging designs with different materials and textures.

6. A method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: The above-mentioned analysis of consumers' demands for brand packaging of different qualities and types based on market sales data determines the diversified direction information of tobacco product brand packaging, which specifically includes: Obtaining purchase history data, preference data, and evaluation data of consumers purchasing tobacco products; Based on consumers’ purchase history and preference data, obtain consumers’ demand for brand packaging of different qualities and types; Based on the tobacco products purchased by consumers and their corresponding brand packaging characteristics, obtain consumers' brand packaging needs for tobacco product characteristics and brand image; Based on different consumers’ purchase history and evaluation data, collaborative filtering algorithms are used to analyze consumers’ preferences and relevance for different tobacco product brand packaging attributes, and to determine consumers’ preferences and requirements for brand packaging attributes such as safety, sustainability, visual appeal, and convenience; According to consumers' preferences and correlations for the packaging attributes of various tobacco product brands, a collaborative filtering algorithm is used to analyze the tobacco product brand packaging attributes that consumers value most, and determine the diversified direction information of tobacco product brand packaging.

7. The method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: The crawling of competitors' brand packaging design data, using convolutional neural networks to identify brand packaging images, and analyzing market brand packaging trends and competitive situations specifically include: Obtain and label competitor brand packaging design image datasets to obtain label data containing brand packaging colors, shapes, patterns, and texts; Based on the acquired brand packaging design image dataset and label data, a convolutional neural network algorithm is used to build a brand packaging image recognition model to determine the brand packaging color, shape, pattern and text attributes of the new brand packaging image; The competitor's brand packaging design images are divided into several clusters, each cluster represents a brand packaging trend or competitive situation, and the K-means clustering algorithm is used to build a brand packaging feature clustering analysis model; Based on the characteristic attributes of the cluster, the brand packaging design trends in the market and the brand packaging strategies of competitors are determined.

8. The method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: The relationship curve between sales performance and different brand packaging and different promotional strategies is determined based on advertising activities, tobacco product brand packaging design and tobacco product sales data, which specifically includes: Obtain sales data and information on brand packaging types, including brand packaging types, sales performance, and brands; Through data preprocessing, the sales data and brand packaging types are encoded and association rules are mined; Constructing frequent item sets of different brand packaging types and sales performance, wherein the frequent item sets represent the association relationship between different brand packaging and sales performance; Construct association rules and extract association rules from frequent item sets according to the confidence threshold; Evaluate the importance of association rules based on support and confidence, and select association rules with practical significance; According to the obtained association rules, the relationship curve between sales performance and different brand packaging and different promotional strategies is determined.

9. The method for determining tobacco product brand packaging and promotion strategy based on data mining according to claim 1, characterized in that: According to the publicity effect and market feedback of tobacco product brand packaging, multiple regression analysis and prediction model are used to evaluate the third influencing factor of tobacco product brand packaging and publicity strategy on market sales, which specifically includes: Obtain data on tobacco product brand packaging design and color, including sales data, market research, and consumer feedback comment text for different designs and color schemes; Use multiple regression analysis to determine the first influencing coefficient of brand packaging design and color on market sales; Obtain relevant data on packaging materials and quality of different brands, including consumers’ evaluation of brand packaging materials and quality and product protection performance; Use multiple regression analysis to determine the second coefficient of influence of brand packaging material and quality on market sales volume; Obtain relevant data on different promotional channels and promotional content, including exposure of different channels and content and consumer response to the promotional content; Use multiple regression analysis to determine the third influence coefficient of publicity channels and publicity content on market sales volume; Obtain data on the target audience's response to brand packaging and promotional strategies, including consumer purchase intention and brand awareness; Use multiple regression analysis to determine the fourth coefficient of influence of the target audience's response on market sales; The third influencing factor of tobacco product brand packaging and publicity strategy on market sales is determined according to the first influencing coefficient, the second influencing coefficient, the third influencing coefficient and the fourth influencing coefficient.

10. A tobacco product brand packaging and promotion strategy determination system based on data mining, characterized in that: include: A first influencing factor determination module is used to crawl negative evaluation information of the cigarette industry based on industry association announcements and news, extract key information related to brand packaging requirements based on the negative evaluation information, and analyze the first influencing factor on tobacco product brand packaging based on the key information; The second impact factor determination module is used to analyze tobacco taxation and tobacco control regulations and predict the second impact factor of changes in taxation and industry requirements on tobacco product brand packaging and promotion strategies; The module for determining acceptance information is used to analyze consumer acceptance information on tobacco product brand packaging and health warnings based on discussions on social media platforms and professional health websites using public opinion analysis tools; The preference information determination module is used to mine consumers' preference information for different brand packaging design elements based on consumers' online feedback and comments on tobacco product brand packaging using sentiment analysis technology; A diversified direction information determination module is used to analyze consumers' demand for brand packaging of different qualities and types based on market sales data, and determine diversified direction information for brand packaging of tobacco products; The module for determining the market brand packaging trend and competition situation is used to crawl the brand packaging design data of competitors, use convolutional neural networks to identify brand packaging images, and analyze the market brand packaging trend and competition situation; A relationship curve determination module, used to determine the relationship curve between sales performance and different brand packaging and different promotional strategies based on advertising activities, tobacco product brand packaging design and tobacco product sales data; The third influencing factor determination module is used to evaluate the third influencing factor of tobacco product brand packaging and publicity strategy on market sales based on the publicity effect of tobacco product brand packaging and market feedback, using multiple regression analysis and prediction models; The brand packaging and promotion strategy determination module is used to determine the target brand packaging and target promotion strategy of tobacco products based on the acceptance level information, the preference level information, the diversification direction information, the market brand packaging trend and competition situation, the relationship curve, the first influencing factor, the second influencing factor and the third influencing factor.

Images