Baijiu selling method and system based on big data

Abstract

The invention discloses a Baijiu selling method and system based on big data, and relates to the technical field of Baijiu selling systems. Comprising a data acquisition module which performs user behavior data acquisition, market data acquisition and user feedback data acquisition; the knowledge graph construction module is used for carrying out structured representation and storage on the collected knowledge by applying a knowledge graph technology, and constructing a wine knowledge graph; and the algorithm analysis module analyzes the consumption habit of the user, predicts the future purchase behavior and demand of the user, and performs price grading on the wine in combination with the consumption habit analysis result and the wine knowledge graph. By comprehensively collecting user behaviors, markets and feedback data and combining algorithm analysis, the system can accurately grasp consumer demands and market trends, a scientific basis is provided for wine recommendation, price grading and wine use scheme planning, and decision-making efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of liquor sales systems, and in particular to a liquor sales method and system based on big data. Background Art

[0002] With the rapid development of information technology and the advent of the big data era, traditional liquor sales models are facing unprecedented challenges and opportunities. On the one hand, consumers are increasingly prioritizing personalization, quality, and experience when choosing liquor, placing higher demands on convenient sales channels, information transparency, and professional service. On the other hand, the widespread application of technologies such as the internet, the Internet of Things, and artificial intelligence is providing a powerful impetus for transformation and upgrading in the liquor industry, enabling precision marketing and intelligent recommendations based on big data analysis.

[0003] Although the current liquor sales system has marketing functions, it lacks personalized services and cannot meet user needs in different situations.

[0004] After searching, the application scheme of Chinese patent application number CN202310573229.3 discloses a liquor sales management method and system based on big data. By obtaining liquor sales big data in a preset area and dividing the liquor sales big data according to regions, regional sales data is obtained. Further online and offline sales feature analysis is performed based on the regional sales data to obtain regional online sales feature data and offline sales feature data. Based on the regional data features, a prediction model is imported to perform sales forecasting, and predicted sales data of each sales area in the preset area is obtained. The predicted sales data is imported into the logistics platform for logistics warehousing analysis to obtain logistics warehousing allocation plans and online promotion plans. The liquor sales management method and system in the above patent have the following deficiencies: insufficient personalized services, and cannot meet user needs in different situations. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and to propose a liquor sales method and system based on big data.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A liquor sales system based on big data, comprising:

[0008] A data collection module, which collects user behavior data, market data, and user feedback data;

[0009] The knowledge graph construction module uses knowledge graph technology to structure and store the collected knowledge and construct a wine knowledge graph;

[0010] Algorithm analysis module: This module analyzes users' consumption habits, predicts their future purchasing behavior and needs, and combines the results of consumption habit analysis with the wine knowledge graph to price-classify wines and use recommendation algorithms to recommend suitable wines to users.

[0011] Wine classification and guidance module: The wine classification and guidance module classifies and manages wines according to the classification results; it designs personalized guidance plans for wines of different price grades and levels;

[0012] User information processing module, which builds user portraits based on the information filled in by users;

[0013] Wine usage plan planning module: Based on user portraits and demand information, combined with wine knowledge graphs and inventory information, the wine usage plan planning module uses rule engines and algorithm models to plan wine usage plans for users;

[0014] The price list information generation module calculates the total price list based on the wine types and quantities selected in the wine usage plan and the price information of the wine types, and displays the price list information on the user interface.

[0015] Preferably, the data acquisition module includes:

[0016] User behavior data collection unit: records users' browsing behavior on the platform, including the wine product pages they browse, their dwell time, and clicks, and analyzes the level of user interest in different wine products; collects users' purchase history data to understand their consumption preferences and frequency; and monitors users' search keywords on the platform to understand their needs and interests.

[0017] Market data collection unit: regularly collects price fluctuation data, sales data, and new product release information from the liquor market to provide reference for algorithm analysis and liquor pricing;

[0018] User feedback data collection unit: Set up user evaluation and feedback functions to collect user evaluations and suggestions.

[0019] Preferably, the knowledge graph construction module performs knowledge graph construction, including the following steps:

[0020] Collect and organize relevant knowledge in the field of wine, use knowledge graph technology to structure the collected knowledge and store it, and build a knowledge graph for liquor; the nodes in the knowledge graph include wine types, brands, brewing processes, raw materials, and drinking occasions, and the nodes are connected by relationship edges to form a knowledge network.

[0021] Preferably: the algorithm analysis module includes:

[0022] Consumption Habit Analysis Algorithm Unit: Based on user behavior data and purchase history data, it uses data mining and machine learning algorithms to analyze user consumption habits; establish user consumption habit models and predict users' future purchasing behavior and needs;

[0023] Wine recommendation algorithm unit: Combines consumption habit analysis results with wine knowledge graph and uses recommendation algorithms to recommend suitable wines to users;

[0024] Price grading algorithm unit: Use data analysis to price grade wines based on their market positioning, brand awareness, brewing process, and raw material cost factors; establish a price grading model to dynamically adjust the price grading of wines based on market changes and user demand.

[0025] Preferably, the wine classification and guidance module includes:

[0026] Wine classification management unit: classifies and manages wines according to the classification results determined by the price classification algorithm; establishes a clear wine classification catalog in the system, and sets detailed descriptions and labels for each wine classification;

[0027] Guidance program design unit: Design personalized guidance programs for wines of different price categories and levels.

[0028] Preferably: the user information processing module includes:

[0029] User demand collection unit: Set up an information filling area on the user interface to guide users to fill in relevant information according to their needs;

[0030] User portrait construction unit: Builds user portraits based on the information filled in by the user and the data collected by the data collection module.

[0031] Preferably, the wine program planning module includes:

[0032] Solution Generation Unit: Based on user profiles and demand information, combined with wine knowledge graphs and inventory information, it uses rule engines and algorithm models to plan wine plans for users. Wine plans include wine selection, pairing suggestions, and dosage calculations.

[0033] Plan adjustment unit: The wine plan planning module provides an editing function, through which users can manually adjust the generated plan; the system updates the plan in real time according to the user's adjustment operation.

[0034] Preferably, the price list information generating module includes:

[0035] Price calculation and integration unit: calculates the total price based on the wine types and quantities selected in the wine plan and the price information of the wine types;

[0036] Price list display and explanation unit: clearly display price list information on the user interface.

[0037] Preferably, the specific method of the solution generating unit for planning a wine solution for the user is:

[0038] Based on the data collection module, collect user information to build user portraits;

[0039] Demand information collection: before users place an order, collect users’ wine demand information;

[0040] Use algorithm models to find matching wines in the wine knowledge graph based on user profiles and demand information;

[0041] The rule engine and algorithm model generate wine usage plans based on user portraits, demand information, wine knowledge graphs and inventory information.

[0042] Preferably, the liquor sales method of the liquor sales system comprises the following steps:

[0043] S1: The data collection module collects user behavior data, market data, and user feedback data;

[0044] S2: The knowledge graph construction module uses knowledge graph technology to structure and store the collected knowledge to build a wine knowledge graph;

[0045] S3: The user information processing module builds a user profile based on the information filled in by the user;

[0046] S4: The user is ready to purchase alcohol. If the user does not indicate a demand, the process proceeds to S5. If the user indicates a demand, the process proceeds to S6.

[0047] S5: The algorithm analysis module analyzes the user's consumption habits, predicts the user's future purchasing behavior and needs, and recommends suitable wines to the user, which then proceeds to S8;

[0048] S6: The wine usage plan module uses a rule engine and algorithm model to plan wine usage plans for users based on user profiles and demand information, combined with wine knowledge graphs and inventory information;

[0049] S7: The price list information generation module calculates the total price list based on the wine type and quantity selected in the wine usage plan and the price information of the wine type, and displays the price list information;

[0050] S8: User decision making.

[0051] The beneficial effects of the present invention are:

[0052] 1. By comprehensively collecting user behavior, market and feedback data and combining it with algorithm analysis, the system can accurately grasp consumer demand and market dynamics, provide a scientific basis for wine recommendations, price grading and wine usage plan planning, and improve decision-making efficiency and accuracy.

[0053] 2. The present invention utilizes user portraits and refined classification management. The system can provide customized wine recommendations, guidance programs and wine usage plans based on the preferences, consumption capabilities and drinking scenarios of different users, thereby enhancing user satisfaction and loyalty.

[0054] 3. The system of the present invention can dynamically adjust wine price classification, recommendation strategies and guidance plans according to market changes, user feedback and the evolution of consumption habits, so as to ensure the timeliness and effectiveness of services and maintain market competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of a sales method of a liquor sales system based on big data proposed by the present invention;

[0056] Figure 2 This is a flowchart of a big data-based liquor sales system proposed by the present invention for analyzing user consumption habits. DETAILED DESCRIPTION

[0057] The technical solution of the present invention will be further described in detail below in conjunction with specific implementation methods.

[0058] Example 1:

[0059] A liquor sales system based on big data, comprising:

[0060] A data collection module, which collects user behavior data, market data, and user feedback data;

[0061] The knowledge graph construction module uses knowledge graph technology to structure and store the collected knowledge and construct a wine knowledge graph;

[0062] User information processing module, which builds user portraits based on the information filled in by users;

[0063] Algorithm analysis module: This module analyzes users' consumption habits, predicts their future purchasing behavior and needs, and combines the results of consumption habit analysis with the wine knowledge graph to price-classify wines and use recommendation algorithms to recommend suitable wines to users.

[0064] Wine classification and guidance module: The wine classification and guidance module classifies and manages wines according to the classification results; it designs personalized guidance plans for wines of different price grades and levels;

[0065] Wine usage plan planning module: Based on user portraits and demand information, combined with wine knowledge graphs and inventory information, the wine usage plan planning module uses rule engines and algorithm models to plan wine usage plans for users;

[0066] The price list information generation module calculates the total price list based on the wine types and quantities selected in the wine usage plan and the price information of the wine types, and clearly displays the price list information on the user interface.

[0067] Wherein, the data acquisition module includes:

[0068] User behavior data collection unit: records users' browsing behavior on the platform, including the wine pages they browse, dwell time, clicks, etc., to analyze users' interest in different wines; collects users' purchase history data, such as the name, price, quantity, and time of purchase, to understand users' consumption preferences and frequency; monitors users' search keywords on the platform to understand users' needs and interests;

[0069] Market data collection unit: regularly collects liquor market price fluctuation data, sales data, new product release information, etc. from sales platforms, industry reports, news media and other channels to provide reference for algorithm analysis and liquor pricing;

[0070] User feedback data collection unit: Set up user evaluation and feedback functions to collect users' evaluations and suggestions on wine taste, quality, packaging, service, etc.

[0071] The knowledge graph construction module performs knowledge graph construction, including the following steps:

[0072] Collect and organize relevant knowledge in the field of wine, including knowledge about wine products (such as brewing technology, origin of raw materials, taste characteristics, aroma components, etc.), drinking culture (such as drinking etiquette, wine table culture, regional drinking customs, etc.), health knowledge (such as the benefits of drinking in moderation, precautions for drinking and health, etc.); use knowledge graph technology to structure the collected knowledge and store it to construct a knowledge graph for liquor; the nodes in the knowledge graph can include wine products, brands, brewing technology, raw materials, drinking occasions, etc., and the nodes are connected by relationship edges to form a knowledge network.

[0073] Wherein, the algorithm analysis module includes:

[0074] Consumption Habit Analysis Algorithm Unit: Analyzes user consumption habits using data mining and machine learning algorithms based on user behavior data and purchase history data. For example, cluster analysis can be used to categorize users into different consumer groups, each with similar consumption characteristics, such as price sensitivity, brand loyalty, and quality pursuit. A user consumption habit model is established to predict future purchasing behavior and demand. For example, based on a user's past purchase history and browsing behavior, the time of next purchase, type of wine, and price range can be predicted.

[0075] Wine Recommendation Algorithm Unit: This unit combines consumption habit analysis results with the wine knowledge graph to apply a recommendation algorithm to recommend suitable wines to users. This algorithm can employ collaborative filtering, content-based recommendations, or hybrid recommendations. It also considers users' individual needs and preferences, such as drinking occasions, taste preferences, and budget constraints, to optimize and rank recommendation results, improving recommendation accuracy and satisfaction.

[0076] Price grading algorithm unit: Based on factors such as the wine's market positioning, brand awareness, brewing process, and raw material costs, the wine is price graded using a combination of data analysis and expert evaluation. The wine is divided into different price levels, such as high-end, mid-end, and low-end, and each price level is further subdivided into different levels. A price grading model is established to dynamically adjust the price grading of the wine according to market changes and user needs to ensure the rationality and timeliness of the grading results.

[0077] The wine classification and guidance module includes:

[0078] Wine classification management unit: classify and manage wines according to the classification results determined by the price classification algorithm; establish a clear wine classification catalog in the system to facilitate user browsing and searching; set detailed descriptions and labels for each wine classification, including price range, wine characteristics, applicable occasions, etc., to help users better understand different categories of wines.

[0079] Guidance program design unit: Design personalized guidance programs for wines of different price ranges and levels; the guidance programs include wine recommendations, pairing suggestions, tasting knowledge introduction, etc., to help users find the most suitable wine for different needs; for example, for high-end wines, professional tasting techniques and suggestions for pairing with high-end food can be provided; for mid- and low-end wines, the emphasis can be on cost-effectiveness and suitability for daily drinking occasions.

[0080] The user information processing module includes:

[0081] User demand collection unit: Set up an information filling area on the user interface to guide users to fill in relevant information according to their needs; for example, drinking occasions (such as business banquets, family gatherings, individual drinking, etc.), taste preferences (such as sauce-flavored, strong-flavored, light-flavored, etc.), budget range, number of drinkers, etc.

[0082] User portrait construction unit: Build a user portrait based on the information filled in by the user and the data collected by the data collection module; the user portrait includes multiple dimensions such as the user's basic information, consumption habits, demand preferences, purchasing power, etc., and comprehensively reflects the user's characteristics and needs.

[0083] The wine program planning module includes:

[0084] Solution Generation Unit: Based on user profiles and demand information, combined with a wine knowledge graph and inventory information, the system uses a rule engine and algorithm model to plan wine plans for users. Wine plans include wine selection, pairing suggestions, and dosage calculations. For example, based on the drinking occasion and the number of people, the system can automatically recommend the appropriate amount and type of wine and provide corresponding drink pairing suggestions.

[0085] Plan adjustment unit: The wine plan planning module provides an editing function, through which users can manually adjust the generated plan; users can change the wine type and adjust the dosage according to their preferences; the system updates the plan in real time based on the user's adjustment operations.

[0086] The price list information generation module includes:

[0087] Price calculation and integration unit: Calculates the total price based on the wine type and quantity selected in the wine plan, combined with the wine price information. The price information includes the wine price, shipping costs, taxes and other detailed expenses, allowing users to clearly understand the consumption amount.

[0088] Price list display and description unit: clearly display price list information on the user interface, including total price, detailed charges and discount descriptions.

[0089] The consumption habit analysis algorithm unit analyzes the consumption habits of users in the following specific ways:

[0090] The data collection module collects user behavior data and records various user behaviors on the platform by embedding data collection code in the front end of the sales system;

[0091] Clean the collected data to remove duplicate, erroneous, or incomplete data. For example, this can remove noise caused by users clicking multiple times on a page within a short period of time. Standardize the data, such as formatting time fields and converting amount fields in a standardized manner.

[0092] Perform feature extraction on the data. Extract features from user behavior data, such as the total time users spend on a wine page within a certain price range, the frequency of browsing a certain type of wine, etc.; extract features such as purchase frequency, average purchase amount per purchase, and brand loyalty (by calculating the proportion of purchases of the same brand) from purchase history data.

[0093] Consumption habit analysis models are constructed and trained, using clustering algorithms (such as K-Means clustering) to divide users into different consumer groups; based on factors such as the price range of the alcohol purchased by users, brand preferences, and purchase frequency, users are divided into high-end consumer groups, cost-effective groups, brand loyal groups, etc.; at the same time, association rule mining algorithms (such as the Apriori algorithm) are used to discover the association patterns between user purchasing behaviors. For example, it is found that users who purchase a certain high-end liquor are likely to also purchase high-end wine glasses and other wine utensils.

[0094] Model training: The preprocessed data is divided into a training set and a test set. The selected model is trained using the training set. For example, in clustering algorithms, the cluster centers and the number of clusters are continuously adjusted to ensure that the consumption habits of users within the same cluster are as similar as possible, while the consumption habits of users in different clusters are significantly different. In association rule mining, valuable association rules are mined by adjusting the support and confidence thresholds.

[0095] Model evaluation and optimization: Use the test set to evaluate the trained model. For clustering algorithms, the clustering effect can be evaluated by calculating indicators such as the silhouette coefficient. For association rule mining, the quality of the rules can be evaluated by calculating indicators such as accuracy and recall. Based on the evaluation results, the model can be optimized, such as adjusting model parameters, adding or removing features, etc.

[0096] Based on the clustering results, targeted marketing strategies are formulated for each consumer group. For example, for high-end consumer groups, limited edition and collector's edition high-end liquors can be recommended, and personalized packaging and delivery services can be provided. For those who pursue cost-effectiveness, cost-effective liquors can be recommended, and combination package discounts can be launched. The results of association rule mining can be used to make personalized recommendations and cross-selling. For example, when a user purchases a certain type of liquor, related drinking utensils or food to go with it can be recommended based on the association rules.

[0097] The specific method of the wine recommendation algorithm unit to recommend suitable wines to users is as follows:

[0098] The construction and integration of a wine knowledge graph involves collecting wine knowledge from multiple sources, including detailed information provided by wine manufacturers (such as brewing process, raw material origin, alcohol content, flavor, etc.), reports from professional wine evaluation agencies, wine encyclopedias, and user-generated comments and reviews. This collected knowledge is represented in a structured manner to construct a knowledge graph. For example, nodes represent entities such as wine, brand, brewing process, and raw materials, and edges represent relationships between entities, such as "wine-belongs-to-brand" and "wine-adopts-brewing process." The knowledge graph can be stored in a graph database (such as Neo4j) for efficient query and retrieval.

[0099] Knowledge fusion and updating: integrating knowledge from different sources to resolve conflicts and redundancies. For example, when different sources have inconsistent descriptions of the aroma of a certain wine, the accurate description can be determined by weighting.

[0100] Based on the characteristics of alcoholic beverages purchased or browsed by the user in the past, as well as information in the alcoholic beverage knowledge graph, the system finds alcoholic beverages similar to the user's preferences and recommends them. For example, if the user frequently purchases Maotai-flavor liquor, the system searches the knowledge graph for other alcoholic beverages with Maotai-flavor characteristics and ranks and recommends them based on other attributes of the alcoholic beverages (such as price and brand awareness). In specific implementation, feature vectors are constructed for each alcoholic beverage and user, and the degree of match between the vectors (such as cosine similarity) is calculated to measure the degree of match between the alcoholic beverage and the user's preferences.

[0101] Recommendations are made based on behavioral similarities between users. First, based on the clustering information from the consumption habit analysis results, other users who belong to the same consumer group as the target user (i.e., neighboring users) are found. Then, based on the purchase and evaluation behavior of neighboring users, the target user's preference for unpurchased alcoholic beverages is predicted. For example, if most neighboring users have purchased a certain alcoholic beverage and rated it highly, then that beverage is recommended to the target user. Collaborative filtering can be divided into user-based collaborative filtering and item-based collaborative filtering. User-based collaborative filtering focuses on the similarity between users, while item-based collaborative filtering focuses on the similarity between alcoholic beverages, determining the similarity of alcoholic beverages by calculating indicators such as the co-occurrence frequency between them.

[0102] Combining the advantages of content-based and collaborative filtering recommendations improves the accuracy and diversity of recommendations. For example, a content-based recommendation algorithm can be used to screen a subset of candidate wines, and then a collaborative filtering algorithm can be used to further sort and filter these candidate wines. Alternatively, the weights of the two recommendation algorithms can be dynamically adjusted based on different scenarios and user needs.

[0103] According to the recommendation scores calculated by the recommendation algorithm, the wines are sorted and a recommendation list is generated. The recommendation list is arranged from high to low according to the recommendation scores of the wines. At the same time, factors such as the price and inventory of the wines are taken into consideration to ensure that the recommended wines are of interest to users and can be purchased.

[0104] The specific method of the guidance scheme design unit to design a personalized guidance scheme is as follows:

[0105] The formulation of price grading standards takes into account factors such as the cost of alcoholic beverages (including raw material costs, production costs, and transportation costs), market positioning, and brand value. For example, alcoholic beverages can be divided into several major price categories, such as high-end (priced above XXX yuan), mid-range (priced between XX and XXX yuan), and low-end (priced below XX yuan). Within each price category, further tiers can be broken down based on factors such as brand awareness and the complexity of the brewing process. For example, high-end alcoholic beverages can be divided into premium luxury and high-end boutique categories.

[0106] Establish a dynamic adjustment mechanism for price grading, and regularly reclassify price tiers and grades based on factors such as market price changes, cost fluctuations, and new product launches. For example, if a rise in the price of a certain raw material leads to a significant increase in the production cost of a wine product, the price tier of that wine product will be adjusted accordingly.

[0107] When users browse wines, the system automatically triggers corresponding guidance plans based on the price grade and level of the wines. For example, when users view the page of high-end wines, the system displays the brand story, tasting knowledge and other content of the high-end wines on the page; when users view the page of low-end wines, the system displays information such as affordable purchase suggestions and simple drinking methods.

[0108] The specific method of the solution generation unit for planning a wine solution for the user is as follows:

[0109] Based on the data collection module, user information is collected to construct a user profile, including basic user information (such as age, gender, region, etc.), consumption habit data (such as purchase frequency, average consumption amount, preferred alcohol type, etc.), behavioral data (such as browsing history, search keywords, etc.), and social data (such as drinking-related content shared through social platforms). These data are integrated and analyzed to form a comprehensive user profile. For example, through analysis, it is found that a user is between 30 and 40 years old, male, lives in an area with high liquor consumption, frequently purchases mid- to high-end liquor, and the search keyword is mostly "business banquet liquor". He has shared photos of drinking parties on social platforms. Then, this user can be labeled as "middle-aged male", "mid- to high-end liquor consumer", "primarily for business purposes", "socially active", etc., forming a complete user profile.

[0110] Demand information collection: before users place an order, collect users’ wine demand information;

[0111] Establish basic rules to meet the basic needs of wine usage plan planning. For example, according to the drinking occasion rules, wedding wines are generally selected with festive packaging and auspicious meanings; business banquets are generally selected with well-known brands and mid-to-high-end price points. According to the number of people rules, the approximate amount of wine corresponding to different number of people is calculated. For example, the amount of white wine per person is calculated in milliliters (taking into account differences in drinking levels and habits), while also considering the amount of spare wine. According to the budget rules, wines are screened within the user's set budget, giving priority to wines with high cost performance.

[0112] Special rules can be added based on business needs and special circumstances. For example, for users who are allergic to alcohol (as determined through user profiles or demand information), the rule engine automatically excludes alcoholic beverages and recommends non-alcoholic drinks as alternatives. For drinking occasions with regional cultural characteristics (such as traditional festival gatherings in certain regions), the rule engine recommends alcoholic beverages based on local customs.

[0113] An algorithmic model is used to search for matching wines in the wine knowledge graph based on user profiles and demand information. For example, a similarity calculation algorithm is used to calculate the similarity between user needs and wine characteristics. Wine characteristics include attributes such as flavor, alcohol content, price, and brand, while user needs include requirements for wine attributes derived from factors such as occasion, number of people, and budget. By calculating similarity, a candidate set of wines that best matches the user needs is selected.

[0114] The consumption calculation algorithm uses an algorithm to calculate the required alcohol consumption based on factors such as the number of drinkers, the expected drinking duration, and the drinking intensity (estimated by the user's drinking habits in the user profile or the user's specified drinking level in the demand information). For example, for a standard bottle of liquor, the algorithm calculates the reasonable number of bottles opened and the expected consumption based on the average person's drinking speed and drinking range, combined with the number of drinkers and the duration of the drink. At the same time, a certain margin is taken into account to cope with the possibility of increased drinking.

[0115] The rule engine and algorithm model generate wine usage plans based on user portraits, demand information, wine knowledge graphs and inventory information; the wine usage plan includes a list of selected wines (including wine names, brands, specifications, quantities, etc.), pairing suggestions (such as food pairings, other beverage pairings, etc.), estimated total price, drinking precautions, etc.

[0116] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A liquor sales system based on big data, characterized in that: include: A data collection module, which collects user behavior data, market data, and user feedback data; The knowledge graph construction module uses knowledge graph technology to structure and store the collected knowledge and construct a wine knowledge graph; Algorithm analysis module: This module analyzes users' consumption habits, predicts their future purchasing behavior and needs, and combines the results of consumption habit analysis with the wine knowledge graph to price-classify wines and use recommendation algorithms to recommend suitable wines to users. Wine classification and guidance module: The wine classification and guidance module classifies and manages wines according to the classification results; it designs personalized guidance plans for wines of different price grades and levels; User information processing module, which builds user portraits based on the information filled in by users; Wine usage plan planning module: Based on user portraits and demand information, combined with wine knowledge graphs and inventory information, the wine usage plan planning module uses rule engines and algorithm models to plan wine usage plans for users; The price list information generation module calculates the total price list based on the wine types and quantities selected in the wine usage plan and the price information of the wine types, and displays the price list information on the user interface.

2. A liquor sales system based on big data according to claim 1, characterized in that: The data acquisition module includes: User behavior data collection unit: records users' browsing behavior on the platform, including the wine product pages they browse, their dwell time, and clicks, and analyzes the level of user interest in different wine products; collects users' purchase history data to understand their consumption preferences and frequency; and monitors users' search keywords on the platform to understand their needs and interests. Market data collection unit: regularly collects price fluctuation data, sales data, and new product release information from the liquor market to provide reference for algorithm analysis and liquor pricing; User feedback data collection unit: Set up user evaluation and feedback functions to collect user evaluations and suggestions.

3. The liquor sales system based on big data according to claim 2, characterized in that: The knowledge graph construction module performs knowledge graph construction, including the following steps: Collect and organize relevant knowledge in the field of wine, use knowledge graph technology to structure the collected knowledge and store it, and build a knowledge graph for liquor; the nodes in the knowledge graph include wine types, brands, brewing processes, raw materials, and drinking occasions, and the nodes are connected by relationship edges to form a knowledge network.

4. The liquor sales system based on big data according to claim 3 is characterized in that: The algorithm analysis module includes: Consumption Habit Analysis Algorithm Unit: Based on user behavior data and purchase history data, it uses data mining and machine learning algorithms to analyze user consumption habits; establish user consumption habit models and predict users' future purchasing behavior and needs; Wine recommendation algorithm unit: Combines consumption habit analysis results with wine knowledge graph and uses recommendation algorithms to recommend suitable wines to users; Price grading algorithm unit: Use data analysis to price grade wines based on their market positioning, brand awareness, brewing process, and raw material cost factors; establish a price grading model to dynamically adjust the price grading of wines based on market changes and user demand.

5. The liquor sales system based on big data according to claim 4 is characterized in that: The wine classification and guidance module includes: Wine classification management unit: classifies and manages wines according to the classification results determined by the price classification algorithm; establishes a clear wine classification catalog in the system, and sets detailed descriptions and labels for each wine classification; Guidance program design unit: Design personalized guidance programs for wines of different price categories and levels.

6. The liquor sales system based on big data according to claim 5 is characterized in that: The user information processing module includes: User demand collection unit: Set up an information filling area on the user interface to guide users to fill in relevant information according to their needs; User portrait construction unit: Builds user portraits based on the information filled in by the user and the data collected by the data collection module.

7. The liquor sales system based on big data according to claim 6 is characterized in that: The wine program planning module includes: Solution Generation Unit: Based on user profiles and demand information, combined with wine knowledge graphs and inventory information, it uses rule engines and algorithm models to plan wine plans for users. Wine plans include wine selection, pairing suggestions, and dosage calculations. Plan adjustment unit: The wine plan planning module provides an editing function, through which users can manually adjust the generated plan; the system updates the plan in real time according to the user's adjustment operation.

8. The liquor sales system based on big data according to claim 7 is characterized in that: The price list information generating module includes: Price calculation and integration unit: calculates the total price based on the wine types and quantities selected in the wine plan and the price information of the wine types; Price list display and explanation unit: clearly display price list information on the user interface.

9. The liquor sales system based on big data according to claim 8, characterized in that: The specific method of the solution generation unit for planning a wine solution for the user is as follows: Based on the data collection module, collect user information to build user portraits; Demand information collection: before users place an order, collect users’ wine demand information; Use algorithm models to find matching wines in the wine knowledge graph based on user profiles and demand information; The rule engine and algorithm model generate wine usage plans based on user portraits, demand information, wine knowledge graphs and inventory information.

10. The liquor sales system based on big data according to claim 9, characterized in that: The liquor sales method of the liquor sales system comprises the following steps: S1: The data collection module collects user behavior data, market data, and user feedback data; S2: The knowledge graph construction module uses knowledge graph technology to structure and store the collected knowledge to build a wine knowledge graph; S3: The user information processing module builds a user profile based on the information filled in by the user; S4: The user is ready to purchase alcohol. If the user does not indicate a demand, the process proceeds to S5. If the user indicates a demand, the process proceeds to S6. S5: The algorithm analysis module analyzes the user's consumption habits, predicts the user's future purchasing behavior and needs, and recommends suitable wines to the user, which then proceeds to S8; S6: The wine usage plan module uses a rule engine and algorithm model to plan wine usage plans for users based on user profiles and demand information, combined with wine knowledge graphs and inventory information; S7: The price list information generation module calculates the total price list based on the wine type and quantity selected in the wine usage plan and the price information of the wine type, and displays the price list information; S8: User decision making.

Images