Intelligent face file personalized recommendation method and system

Abstract

The invention provides an intelligent face document personalized recommendation method and system. The method comprises the following steps: firstly, collecting multi-dimensional face document demand data (including historical consumption, real-time demand and preference change data) and face document supply data (including real-time supply, feature change and consumption feedback data) of a user; constructing a demand evolution trajectory based on the user demand data, and presenting a demand change trend and an association relationship; constructing a supply feature time sequence association set based on the supply data, and presenting a supply feature association mode and a change rule; calling an artificial intelligence model for interactive analysis to generate a dynamic demand supply matching rule; according to the method, the rule is iteratively optimized by using the user feedback data, and the personalized recommendation result of the face file is finally generated, so that the accurate dynamic matching of the demand and the supply is realized, and the user satisfaction and the face file operation benefit are improved.

Description

Technical Field

[0001] This invention relates to the field of digital online service technology, and more specifically, to a smart noodle personalization recommendation method and system. Background Technology

[0002] In the food service sector, especially in noodle stalls, providing accurate and personalized recommendations that meet users' needs is crucial for improving user satisfaction and the profitability of noodle stall operations. Traditional noodle stall recommendation methods have significant shortcomings.

[0003] On the one hand, traditional recommendations rely primarily on simple historical user consumption records, only able to extract user preferences from limited past data, failing to fully consider the multidimensional and dynamic nature of user needs. A user's dietary needs not only change due to factors such as season, health condition, and personal emotions, but real-time dietary needs are also difficult to accurately capture and record. For example, a user might need high-carbohydrate pasta on a particular day due to high activity levels, but traditional methods cannot promptly detect such temporary needs.

[0004] On the other hand, traditional methods lack in-depth analysis and dynamic tracking of noodle supply characteristics. The supply of noodles at noodle stalls changes with time, ingredient availability, and the chef's condition, causing fluctuations in characteristics such as taste, texture, and nutritional balance. Traditional recommendations do not consider the changing patterns of these supply characteristics and their correlation with user demand, resulting in recommended noodles that may not meet user expectations or the actual supply capacity of the noodle stall, failing to achieve an effective match between supply and demand. Summary of the Invention

[0005] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a smart noodle personalization recommendation method, the method comprising:

[0006] Collect multi-dimensional noodle stall demand data and noodle stall supply data from users. The multi-dimensional noodle stall demand data from users includes historical noodle stall consumption data, real-time dietary demand data, and dietary preference change data. The noodle stall supply data includes real-time noodle product supply data, noodle product characteristic change data, and noodle product consumption feedback data.

[0007] Based on multi-dimensional user profile demand data, a user profile demand evolution trajectory is constructed. This trajectory is used to present the changing trends and correlations of user profile demand in different time periods.

[0008] Based on the noodle stall supply data, a time-series correlation set of noodle stall supply characteristics is constructed. This time-series correlation set of noodle stall supply characteristics is used to present the correlation patterns and changing rules of noodle stall supply characteristics in different time periods.

[0009] The artificial intelligence model is invoked to perform interactive analysis on the evolution trajectory of user faceplate demand and the time-series correlation set of faceplate supply characteristics, and to generate dynamic demand and supply matching rules.

[0010] The dynamic demand and supply matching rules are iteratively optimized by utilizing user feedback data on the output results of the dynamic demand and supply matching rules, and personalized recommendation results are generated based on the optimized dynamic demand and supply matching rules.

[0011] In another aspect, embodiments of the present invention also provide a smart personalized recommendation system for face plates, including a processor and a machine-readable storage medium connected to the processor. The machine-readable storage medium is used to store programs, instructions, or code, and the processor is used to execute the programs, instructions, or code in the machine-readable storage medium to implement the above-described method.

[0012] Based on the above, this invention collects multi-dimensional user demand data and supply data for noodle stalls. It constructs a trajectory of user demand evolution based on the multi-dimensional demand data, revealing the changing trends and internal relationships of user demand across different time periods, and providing in-depth insights into the dynamic characteristics of user needs. Simultaneously, it constructs a time-series correlation set of noodle supply characteristics based on supply data, accurately presenting the correlation patterns and changing rules of noodle supply characteristics across different time periods. An artificial intelligence model is invoked to interactively analyze the demand evolution trajectory and the time-series correlation set of supply characteristics, generating dynamic demand-supply matching rules, thus achieving intelligent and dynamic matching of demand and supply. By iteratively optimizing the matching rules using user feedback data, the recommended results continuously align with actual user needs and noodle supply conditions. The resulting personalized noodle recommendations are highly accurate and adaptable, effectively improving user satisfaction with noodle stall services. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the execution flow of the intelligent facial profile personalized recommendation method provided in an embodiment of the present invention.

[0014] Figure 2 This is a schematic diagram of exemplary hardware and software components of the smart face plate personalized recommendation system provided in an embodiment of the present invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a personalized smart noodle recommendation method provided in one embodiment of the present invention. The personalized smart noodle recommendation method will be described in detail below.

[0016] Step S110: Collect multi-dimensional noodle stall demand data and noodle stall supply data from users. The multi-dimensional noodle stall demand data from users includes historical noodle stall consumption data, real-time dietary demand data, and dietary preference change data. The noodle stall supply data includes real-time noodle supply data, noodle characteristic change data, and noodle consumption feedback data.

[0017] This embodiment uses the "AI Noodle Shop" chain of smart noodle stalls as a unified application scenario, covering service models such as online mini-program ordering, offline in-store dining, and delivery. Noodle products include three main categories: soup noodles (clear broth, thick broth), mixed noodles (sesame sauce, scallion oil), and stir-fried noodles (soy sauce flavor, spicy flavor), with accompanying snacks including braised dishes and side dishes. Data is collected through multi-module collaboration, as detailed below.

[0018] User multi-dimensional noodle stall demand data is acquired through a three-layer collection mechanism: Historical noodle stall consumption data is synchronized from the membership system, POS system, and food delivery platform interface by the consumption record module. Each record includes the consumption time period (accurate to the hour), consumption channel identifier, noodle name, noodle feature combination (flavor, texture, ingredients, portion size), consumption amount, payment method, evaluation content (text + star rating), repurchase identifier, etc.; Real-time dietary demand data is collected by the real-time interaction module through the mini-program input box, voice recognition component, and scene perception sensor. For example, if a user inputs "light soup noodles, no cilantro", the voice recognition will say "a non-spicy mixed noodle", and the scene perception will infer "quick breakfast demand" through location (office building area), time (7:30-9:00), and device status (low battery of mobile device); Dietary preference change data is generated by the preference analysis module through historical consumption comparison, user active settings (mini-program "preference management" page), and evaluation keyword extraction. For example, the comparison shows that the user has switched from frequently ordering spicy noodles to clear soup noodles, the user checked the "low salt" option, and the evaluation includes statements such as "I like light flavors recently".

[0019] The noodle supply data is collected through the supply chain and operations system: real-time noodle supply data is provided by the kitchen management module, including the list of noodles on sale in each store, remaining supply, estimated serving time, and ingredient preparation status (such as "cilantro out of stock" and "beef in stock"); noodle characteristic change data is recorded by the product management module, covering seasonal adjustments (cold noodles in summer, warm soup noodles in winter), ingredient replacements (local vegetables replacing vegetables from other regions), and process upgrades (hand-pulled noodles replacing machine-pulled noodles), etc. Each record includes the time of change, a comparison of characteristics before and after, and the reason for the change; noodle consumption feedback data is integrated by the feedback processing module, with sources including mini-program reviews, store feedback cards, and takeaway platform comments. Each feedback includes a timestamp, noodle ID, feedback category (taste / portion / service), specific content, and associated consumption record ID.

[0020] Regarding privacy protection, user mobile phone numbers and ID card numbers (used for member registration) are anonymized. The middle part of the mobile phone number is replaced with asterisks, and only the birth date and gender code are retained in the ID card number. Payment information only stores the transaction serial number and does not retain the complete card number. Data transmission is encrypted with SSL, and sensitive fields are encrypted using the AES algorithm. The key is managed by the hardware security module. Access permissions are divided according to roles. Only the data processing and recommendation modules can access the anonymized data, and operation logs are recorded throughout the process.

[0021] When collecting multi-dimensional user profile data, the consumption record module obtains explicit authorization from users during membership registration and order placement before synchronizing data from the membership system, POS system, and food delivery platform interfaces. A clear and easy-to-understand user agreement explains the purpose, scope, and usage of data collection, and data collection only proceeds after the user checks their agreement. Similarly, when the real-time interaction module collects data through the mini-program, voice recognition component, and scene perception sensors, users are informed of data collection matters in advance in the mini-program's privacy policy; continued use of the mini-program constitutes consent. For data acquisition in the preference analysis module, a pop-up reminder appears when users actively set preferences or submit reviews, indicating that the data will be used for preference analysis and service optimization; relevant data is only recorded after user confirmation.

[0022] The collection of noodle supply data is also compliant. Before collecting data, the kitchen management, product management, and feedback processing modules all adhere to internal regulations and employee training to ensure that all data sources are legal, and that user privacy-related data, such as changes in noodle characteristics and consumer feedback, has been anonymized. Regarding data authorization, AI Noodle Restaurant has signed compliant data sharing agreements with third-party data cooperation platforms, clearly defining data usage boundaries and security responsibilities to prevent data misuse. Simultaneously, a professional legal team is regularly invited to review the data collection and authorization processes, updating them in a timely manner to match dynamic adjustments to laws and regulations, ensuring that the entire data lifecycle operates within a legal and compliant framework.

[0023] Step S120: Construct a user face profile demand evolution trajectory based on multi-dimensional user face profile demand data. The user face profile demand evolution trajectory is used to present the changing trends and correlations of user face profile demands in different time periods.

[0024] Using the demand data of "AI Noodle Shop" user A as a sample, the evolution trajectory is constructed through processes such as data extraction, timeline construction, and trend analysis. The specific steps are as follows.

[0025] Step S121: Extract consumption time data, noodle type data, noodle characteristic data, and consumption evaluation data from the user's historical noodle consumption data from the user's multi-dimensional noodle demand data.

[0026] Four core data categories were extracted from User A's historical consumption database: Consumption time data, which included specific consumption times within the past six months, recorded by year, month, day, hour, and minute; Noodle type data, categorized into three main types: soup noodles, mixed noodles, and stir-fried noodles, with each consumption type labeled; Noodle characteristic data, which detailed the flavor (clear broth / spicy / scallion oil, etc.), texture (chewy / soft), toppings (beef / shredded chicken / vegetables, etc.), and portion size (regular / extra large) of each noodle ordered; and Consumption evaluation data, which included star ratings and text reviews after each consumption, with keywords extracted from the text reviews (such as "fresh," "spicy," and "generous portion").

[0027] The extracted data is sorted chronologically by consumption time to generate a structured historical consumption list. Each record is associated with a unique consumption ID for easy tracing later.

[0028] Step S122: Extract the current dietary scenario data, current dietary requirement description data, and current dietary preference data from the user's multi-dimensional profile demand data in the real-time dietary demand data.

[0029] Extracting User A's real-time demand data: Current dietary scenario data is generated by the scenario awareness module, including scenario type (office building breakfast / community lunch / family dinner) and scenario attributes (time urgency: high / medium / low; number of companions: 1 / multiple), for example, "Scenario type: office building breakfast, time urgency: high, number of companions: 1 person"; Current dietary demand description data is the original text entered by the user through the mini-program, such as "I want a bowl of light soup noodles, no cilantro, and I want it made as soon as possible"; Current dietary preference data is generated by extracting keywords from the real-time demand description and recent consumption records, such as "light", "soup noodles", "no cilantro", and "quick".

[0030] The three types of data are integrated into a real-time demand structure, which includes data generation timestamps, scenario identifiers, demand text, and a list of preferred keywords.

[0031] Step S123: Extract the time node data of preference change, the comparison data before and after preference change, and the data of factors influencing preference change from the user's multi-dimensional needs data.

[0032] The preference analysis module extracts user A's preference change data: the preference change time point data is the consumption time when the preference change first occurs, such as "the time when the user first ordered clear soup noodles instead of spicy noodles"; the comparison data before and after the preference change is classified by feature dimensions, such as flavor dimension (before the change: mainly spicy; after the change: mainly clear soup) and topping dimension (before the change: no aversion to toppings; after the change: repeatedly noted "no cilantro"); the data on factors influencing preference changes are extracted through evaluation text, user active feedback, and scene changes, such as the evaluation mentioning "wanting to eat something light because of the hot weather", the user adding "cilantro allergy" in the preference settings, and the season changing from spring to summer.

[0033] Establish a related profile for each change node, including consumption samples before and after the change, sources of influencing factors, and credibility labels.

[0034] Step S124: Arrange the consumption time data in chronological order, establish a timeline framework, and map the consumption noodle type data, consumption noodle characteristic data, and consumption evaluation data at the corresponding time points to the corresponding positions in the timeline framework.

[0035] A timeline framework for user A's needs is constructed using "week" as the time unit: the horizontal axis represents continuous time nodes (from the first week to the last week of the past six months), and the vertical axis is divided into three sub-dimensions: type distribution, feature distribution, and evaluation performance. Consumption time data is categorized by week, with each time node corresponding to all consumption records for that week; consumption noodle type data is mapped to the consumption percentage of each type of noodle (e.g., soup noodles accounted for 80% and dry noodles for 20% in a certain week); consumption noodle feature data is statistically distributed by flavor, texture, ingredients, and portion size (e.g., clear broth accounted for 90% and spicy for 10% in a certain week); consumption evaluation data is mapped to the average star rating and high-frequency evaluation keywords for that week (e.g., average star rating 4.8, keywords "light," "fresh," "no cilantro").

[0036] The timeline framework uses an extensible data structure for storage, supporting data querying and filtering by time node and feature dimension.

[0037] Step S125: Mark the current dietary scenario data, current dietary requirement description data, and current dietary preference data in the real-time dietary requirement data at the latest time node position in the timeline frame.

[0038] Real-time demand data is marked at the latest time node on the timeline (the last day of the last week of the past six months): Current dietary scenario data is displayed in the form of tags, such as "office building breakfast" and "high urgency"; Current dietary demand description data is linked in the form of floating text boxes, which can be clicked to view the full content; Current dietary preference data is presented in the form of keyword cloud, with the size of the keywords reflecting their weight, such as "light", "soup noodles" and "no cilantro" having the highest weight and being displayed in the largest font size.

[0039] The labeled data is distinguished from historical data by using bright colors, making it easy to intuitively identify the relationship between real-time needs and historical trends.

[0040] Step S126: Analyze the time node data of preference change in the dietary preference change data, mark the preference change time node in the corresponding position of the time axis frame, and associate the preference change before and after the preference change time node with the comparison data before and after the preference change.

[0041] Locate the time nodes of preference changes on the timeline and mark them with vertical red lines; mark the preference data before the change to the left of the red line, including the proportion of types, feature distribution, and evaluation keywords (such as the high frequency of keywords "spicy" and "mixed noodles" before the change); mark the preference data after the change to the right of the red line (such as the high frequency of keywords "clear soup" and "soup noodles" after the change); connect the changes in the corresponding feature dimensions with arrows, such as "flavor: spicy - clear soup" and "ingredients: with cilantro - without cilantro".

[0042] Clicking on each change node will expand a detailed comparison table, which includes information such as the frequency of consumption before and after the change, and changes in rating stars.

[0043] Step S127: Based on the data of each time node in the time axis framework, calculate the rate of change of the type of consumer goods, the similarity of the characteristics of consumer goods, and the fluctuation value of consumer evaluation between adjacent time nodes.

[0044] For two adjacent time points on the timeline (such as week N and week N+1), the quantitative indicators are calculated according to the following logic:

[0045] Step S1271: Extract the consumption noodle product type data of two adjacent time nodes in the time axis frame, count the number and specific types of consumption noodle product types of the previous time node, and count the number and specific types of consumption noodle product types of the next time node.

[0046] Extract the types of noodles consumed in the previous node (week N), and record the number of types and the specific types (e.g., 2 types: soup noodles, dry noodles); extract the types of noodles consumed in the next node (week N+1), and record the number of types and the specific types (e.g., 2 types: soup noodles, fried noodles).

[0047] Step S1272: Calculate the number of new noodle product types and the number of noodle product types that disappeared at the next time point compared to the previous time point. Divide the sum of the number of new noodle product types and the number of noodle product types that disappeared by the number of noodle product types at the previous time point to obtain the rate of change of noodle product types.

[0048] After counting the number of new types added to the node (e.g., 1 type: fried noodles) and the number of types that disappeared (e.g., 1 type: mixed noodles), calculate the sum of the two; divide the sum by the number of types of the previous node to obtain the type change rate. The higher the change rate, the more significant the type change.

[0049] Step S1273: Extract the consumption noodle product feature data of two adjacent time nodes. The consumption noodle product feature data includes noodle product flavor features, noodle product texture features, noodle product ingredient features, and noodle product portion features.

[0050] Extract feature data of the preceding and following nodes in four dimensions: flavor, texture, ingredients, and portion size. For example, the preceding node is mainly spicy, while the following node is mainly clear soup; the preceding node contains cilantro, while the following node does not.

[0051] Step S1274: For each noodle product feature, convert the noodle product feature data of adjacent time nodes into standardized feature vectors, calculate the similarity value between two standardized feature vectors, and take the average of the similarity values ​​corresponding to all noodle product features as the similarity of the consumer noodle product features.

[0052] Convert the data for each feature dimension into a standardized vector (vector elements represent the proportion of each feature value), such as a flavor vector of [spicy proportion, clear soup proportion, scallion oil proportion, etc.]. Calculate the cosine similarity between the feature vectors corresponding to the preceding and following nodes to obtain the similarity for each dimension. Take the average of the similarities of the four dimensions as the overall feature similarity. The lower the similarity, the more significant the feature change.

[0053] Step S1275: Extract consumer evaluation data from two adjacent time points. This consumer evaluation data includes evaluation scores, the number of positive words in the evaluation text, and the number of negative words in the evaluation text.

[0054] Extract the average evaluation score of the preceding and following nodes, the number of positive words (such as "fresh", "good", "satisfied" etc.), and the number of negative words (such as "salty", "slow", "few" etc.).

[0055] Step S1276: Calculate the difference between the evaluation score of the next time node and the evaluation score of the previous time node, and calculate the difference between the ratio of the number of positive words to the number of negative words at the next time node and the ratio of the number of positive words to the number of negative words at the previous time node.

[0056] Calculate the difference in evaluation scores (score of the later node minus score of the earlier node); calculate the ratio of "number of positive words / number of negative words" between the earlier and later nodes, and then calculate the difference between the two ratios (if the number of negative words is 0, the ratio is set according to the preset rules).

[0057] Step S1277: The difference between the evaluation scores and the difference between the vocabulary ratios are weighted and summed. The weighting is set according to the importance of the evaluation scores and vocabulary ratios in the consumer evaluation, and the fluctuation value of the consumer evaluation is obtained.

[0058] Based on business experience, set the weights of the evaluation score and the vocabulary ratio (e.g., score weight 0.6, ratio weight 0.4). Multiply the two differences by their respective weights and sum them to obtain the evaluation fluctuation value. A positive fluctuation value indicates an improvement in the evaluation, while a negative value indicates a decline in the evaluation.

[0059] Step S1278: Record the calculated rate of change of consumer noodle types, similarity of consumer noodle features, and fluctuation value of consumer evaluation, and store them in association with the corresponding adjacent time nodes.

[0060] The three calculated indicators are associated with the corresponding time point pairs (such as week N to week N+1) and stored in the quantitative indicator database. Each record contains a time point pair, type change rate, feature similarity, and evaluation fluctuation value.

[0061] Step S1279: Summarize the calculation results of all adjacent time node pairs to form a set of quantitative indicators of changes in user profile requirements for each time period in the time axis framework.

[0062] Traverse all adjacent node pairs in the timeline, summarize their quantitative indicators, organize them by time period (such as monthly or quarterly) to form an indicator set for each time period, which facilitates the overall analysis of demand change patterns.

[0063] Step S128: Based on the calculated rate of change, similarity and fluctuation value, construct a trend curve of user profile demand change. The horizontal axis of the trend curve of user profile demand change is the time node, and the vertical axis is the quantitative indicator of demand change.

[0064] Plot three trend curves with time points on the horizontal axis: the type change rate curve (vertical axis represents the change rate), the feature similarity curve (vertical axis represents the similarity), and the evaluation fluctuation value curve (vertical axis represents the fluctuation value). Smooth the curves to eliminate fluctuations caused by occasional consumption and highlight the overall trend. For example, a peak in the type change rate curve at a certain time point indicates a significant change in the choice of noodle types during that period; a trough in the feature similarity curve indicates a significant change in preference characteristics during that period.

[0065] Step S129: Associate the data on factors influencing changes in dietary preferences with the fluctuation nodes in the trend curve of user demand for face food, and label the influencing factors corresponding to each fluctuation node.

[0066] Identify fluctuation nodes in the trend curve (such as peak values ​​of type change rate, trough values ​​of feature similarity, and time nodes corresponding to extreme values ​​of evaluation fluctuation). Associate the extracted factors affecting preference changes with the fluctuation nodes and annotate them in the corresponding positions on the curve, such as "Fluctuation node: Week 10, influencing factors: warmer weather, user setting no preference for cilantro".

[0067] Each influencing factor label must include its source (evaluation extraction / active setting / scenario change) and a credibility score (based on the reliability of the data source).

[0068] Step S1210: Integrate the timeline framework, the user profile demand change trend curve, and the influencing factor annotations to form the user profile demand evolution trajectory. This user profile demand evolution trajectory includes the distribution of demand data in the time dimension, the user profile demand change trend, and the correlation information of the influencing factors.

[0069] The timeline framework, three trend curves, and the influencing factors of fluctuation nodes are integrated into a structured evolution trajectory document. The document is divided into three parts: time dimension data distribution (presenting detailed data on type, characteristics, and evaluation by week), trend analysis (interpreting the curve trend, such as "after week 10, the flavor changed from spicy to clear soup, and the evaluation improved accordingly"), and influencing factor association table (listing each fluctuation node and its corresponding influencing factors, source, and credibility).

[0070] This evolutionary trajectory supports dynamic updates, automatically updating the timeline, trend curves, and related information of influencing factors when users generate new consumption or demand data.

[0071] Step S130: Construct a time-series correlation set of noodle supply characteristics based on noodle supply data. The time-series correlation set of noodle supply characteristics is used to present the correlation patterns and change rules of noodle supply characteristics in different time periods.

[0072] Using the supply data of a certain store of "AI Noodle House" as a sample, a set was constructed through processes such as time interval division, feature extraction, and correlation analysis. The specific steps are as follows.

[0073] Step S131: Extract the supply time data, noodle name data, noodle ingredient data, noodle preparation method data, and noodle quantity data from the noodle stall supply data in the real-time noodle supply data.

[0074] Real-time supply data is extracted from the store's kitchen management system: Supply time data includes the start and end times of each operating period on the day (e.g., morning market 6:00-10:00, lunch market 11:00-14:30, dinner market 17:00-21:00); Supply noodle name data includes the complete name of the noodles available for sale during each period (e.g., "Bone Broth Beef Noodles", "Scallion Oil Chicken Noodles", "Soy Sauce Fried Noodles"); Supply noodle ingredient data includes a detailed ingredient list for each type of noodle (e.g., "Bone Broth Beef Noodles: Beef bone broth base, hand-pulled noodles, beef slices, greens, ginger slices, scallions"); Supply noodle preparation method data records the preparation process for each type of noodle (e.g., "Hand-pulled noodles, slow-cooked bone broth, made to order"); Supply noodle quantity data includes the remaining supply of each type of noodle and the total stock available for the day.

[0075] Step S132: Extract the characteristic change time data, pre-change characteristic data, post-change characteristic data, and characteristic change reason data from the noodle supply data.

[0076] Extract noodle product feature change data from the product management module: Feature change time data is the specific time of noodle product feature adjustment (accurate to the day); Before the change, the noodle product feature data is the noodle product feature before the adjustment (e.g., "Old Altar Pickled Cabbage Noodles: Traditional Pickled Cabbage, Ordinary Soup Base"); After the change, the noodle product feature data is the noodle product feature after the adjustment (e.g., "Old Altar Pickled Cabbage Noodles: Upgraded Pickled Cabbage, Bone Broth Soup Base"); Feature change reason data explains the adjustment motivation (e.g., "Ingredient Upgrade", "User Feedback Optimization", "Seasonal Adjustment").

[0077] Step S133: Extract feedback time data, feedback noodle identification data, feedback content data, and corresponding consumption quantity data from the noodle supply data.

[0078] Extract noodle consumption feedback data from the feedback processing module: feedback time data is the timestamp of the user submitting feedback; feedback noodle identifier data is the unique ID of the noodle corresponding to the feedback; feedback content data is the specific text of the user's feedback (such as "the bone broth beef noodle soup is very fresh" or "the scallion oil noodles are too salty"); and the consumption quantity data corresponding to the feedback is the purchase quantity in the consumption record associated with the feedback.

[0079] Step S134: Divide the supply time data into multiple consecutive time intervals in chronological order, based on operating hours (morning market, noon market, evening market) and date (each day is an independent time unit), forming a two-level time interval structure of "day-time period", such as "August 1st morning market", "August 1st noon market", "August 2nd morning market", etc. For each time interval, statistically analyze the distribution of the following data: the name of the supplied noodle products (e.g., "August 1st morning market supplies beef broth noodles, scallion oil noodles, and vegetable and egg noodles"), the ingredients of the supplied noodles (organize the ingredient list by noodle product), the preparation method of the supplied noodles (record the preparation process by noodle product), and the quantity of supplied noodles (statistically analyze the total stock, remaining supply, and total sales of each type of noodle product), forming a snapshot of the supply data for each time interval.

[0080] Step S135: For each time interval, analyze the correspondence between the data on the ingredients of the supplied noodles and the data on the methods of producing the supplied noodles, and determine the core characteristic combination of the noodles supply within that time interval.

[0081] Feature association analysis was performed on the supply data snapshots for each time interval: Taking the "August 1st morning market" as an example, the correspondence between the ingredients (beef bone broth, hand-pulled noodles, etc.) and preparation methods (slow-cooked bone broth, hand-pulled noodles) of "beef bone broth noodles" was analyzed, and "bone broth + hand-pulled noodles" was determined as the characteristic combination of this noodle dish; the correspondence between the ingredients (scallion oil, hand-pulled noodles, cucumber shreds) and preparation methods (hand-pulled noodles, freshly prepared scallion oil) of "scallion oil seasoning + hand-pulled noodles" was analyzed, and "scallion oil seasoning + hand-pulled noodles" was determined as its characteristic combination. The characteristic combinations of all noodle dishes within this time interval were summarized, and the characteristic combination with the highest frequency and the largest number of noodle dishes covered was selected as the core characteristic combination. For example, the core characteristic combination of the "August 1st morning market" was "hand-pulled noodles + traditional soup base (bone broth, clear broth)," and this core characteristic combination covered 80% of the noodle dishes supplied during this period.

[0082] Step S136: Map the feature change time data in the noodle feature change data to the divided time interval, and associate the noodle feature data before the change, the noodle feature data after the change, and the feature change reason data within the time interval.

[0083] Based on the time data (accurate to the day) of feature changes in noodle product data, it is matched to the corresponding "day-time period" interval. For example, if a feature change of a noodle product occurs at 9:00 AM on August 1st (morning market time), then the change is associated with the "August 1st morning market" time interval. In the supply data of this time interval, the product feature data before the change (e.g., "before the change: ordinary soup base, machine-pulled noodles"), the product feature data after the change (e.g., "after the change: bone broth soup base, hand-pulled noodles"), and the reason data for the feature change (e.g., "reason: user feedback to improve taste") are marked to ensure that feature changes within each time interval are traceable.

[0084] Step S137: Map the feedback time data to the divided time intervals, and count the data types and proportions of feedback content corresponding to different feedback product identifier data in each time interval.

[0085] Based on the timestamp of the feedback time data, each feedback is associated with the corresponding "day-time period" interval. For example, if a feedback is submitted at 12:00 on August 1st (lunchtime), it is classified into the "August 1st Lunchtime" time interval. For each time interval, feedback is categorized by noodle product identifier data, and the data type of feedback content for each noodle product is statistically analyzed (e.g., flavor, portion size, service). For example, in the "August 1st Lunchtime" feedback for "Bone Broth Beef Noodles" is categorized as flavor (70%) and portion size (30%). The percentage of each feedback type in the total number of feedbacks for that noodle product, and the percentage of feedback for each noodle product in the total number of feedbacks for that time interval, are calculated to form feedback distribution statistics.

[0086] Step S138: Calculate the similarity of the core feature combinations of noodle supply within adjacent time intervals to determine the inheritance relationship and variation range of noodle supply features.

[0087] Analyze the core feature combinations and relationships between adjacent time intervals using the following process:

[0088] Step S1381: Extract the core feature combination of noodle supply for two adjacent time intervals. Each core feature combination of noodle supply includes a subset of noodle ingredient features, a subset of noodle preparation method features, and a subset of noodle quantity distribution features.

[0089] Extract the core feature combination of "August 1st morning market" and "August 1st afternoon market": The former includes the ingredient feature subset (bone broth, clear broth, handmade noodle toppings), the preparation method feature subset (handmade noodles, slow-cooked bone broth), and the quantity distribution feature subset (bone broth beef noodles account for 40%, scallion oil noodles account for 30%); the latter includes the ingredient feature subset (bone broth, soy sauce flavor, handmade noodle toppings), the preparation method feature subset (handmade noodles, stir-fried with soy sauce flavor), and the quantity distribution feature subset (bone broth beef noodles account for 20%, stir-fried noodles with soy sauce flavor account for 40%).

[0090] Step S1382: Convert each feature subset in the core feature combination of the noodle supply in the previous time interval into a feature matrix. The rows of the feature matrix represent feature types, and the columns represent specific attribute values ​​of the features.

[0091] Convert the ingredient feature subset of "August 1st Morning Market" into a matrix: rows represent "Soup Base Type", "Noodle Type", and "Side Dish Type", and columns represent specific attribute values ​​(e.g., the Soup Base Type column includes the proportion of bone broth and clear broth); the preparation method feature subset matrix: rows represent "Ramen Method" and "Soup Base Preparation", and columns represent specific attribute values ​​(e.g., the Ramen Method column includes the proportion of handmade ramen); the quantity distribution feature subset matrix: rows represent "Noodle Type", and columns represent specific percentage values.

[0092] Step S1383: Convert each feature subset in the core feature combination of the noodle supply in the next time interval into a feature matrix with the same dimension as the feature matrix of the previous time interval.

[0093] Based on the matrix dimensions of the previous interval, transform the feature subset of "August 1st lunch market": in the ingredient feature matrix, the soup base type column includes the proportion of bone broth and soy sauce flavor, ensuring that the rows are consistent with the previous matrix; in the preparation method matrix, add a "stir-fry method" row, but retain the "ramen method" row, ensuring dimension matching; the quantity distribution matrix is ​​arranged by the same noodle type, and the column value of no corresponding noodle is set to 0.

[0094] Step S1384: For the feature matrix corresponding to each feature subset, the matrix similarity algorithm is used to calculate the similarity value of the feature matrices in adjacent time intervals. The matrix similarity algorithm calculates the overall similarity based on the correspondence of matrix elements.

[0095] For the ingredient feature matrix, calculate the element similarity of the corresponding rows (such as the difference in the proportion of bone broth in soup base types), and then take the average value as the matrix similarity; for the preparation method matrix, calculate the element similarity of the corresponding rows in the same way; for the quantity distribution matrix, calculate the cosine similarity of the proportion of noodle products.

[0096] Step S1385: Take the average of the similarity values ​​of all feature subsets as the overall similarity of the core feature combination of noodle supply in adjacent time intervals.

[0097] The average similarity values ​​of the three feature subsets of ingredients, preparation method, and quantity distribution are used to obtain the overall similarity of the core feature combination of adjacent intervals. For example, the overall similarity between "August 1st morning market" and "August 1st afternoon market" is 0.6.

[0098] Step S1386: If the overall similarity is higher than the set inheritance threshold, it is determined that the core feature combination of the noodle supply in the later time interval has inherited the core feature combination of the noodle supply in the previous time interval, and the inherited feature subset and specific features are recorded.

[0099] The inheritance threshold was set at 0.5. Since the overall similarity of 0.6 was higher than the threshold, it was determined that the lunch market inherited the core feature combination of the morning market. The inherited features were recorded: "handmade noodle ingredients" in the ingredients and "handmade noodles" in the preparation method.

[0100] Step S1387: If the overall similarity is lower than the set inheritance threshold, analyze the difference feature subsets of the core feature combination of the noodle supply in the two time intervals, and count the number of feature changes and the type of feature attributes in the difference feature subsets.

[0101] If the overall similarity of a certain adjacent interval is 0.4 (below the threshold), analyze the difference feature subsets: for example, if the soup base type of the ingredient feature subset changes from "bone broth, clear broth" to "spicy soup base, tomato soup base", and the preparation method changes from "handmade noodles" to "machine-made noodles", the number of changes is counted as 2 feature subsets, and the change attribute type is "soup base type" and "preparation method type".

[0102] Step S1388: Calculate the change magnitude of attribute values ​​of the changing features in the subset of differential features. Divide the difference between the attribute value of the next time interval and the attribute value of the previous time interval for each changing feature by the attribute value of the previous time interval to obtain the change magnitude of a single changing feature.

[0103] Regarding the change in "soup base type", bone broth accounted for 40% in the first section and spicy soup base accounted for 50% in the second section, with the change range being (50%-40%) / 40%; regarding the change in "preparation method", hand-pulled noodles accounted for 100% in the first section and machine-pulled noodles accounted for 100% in the second section, with the change range being (100%-100%) / 100%=0.

[0104] Step S1389: Take the average value of the change range of all change characteristics as the overall change range of the core characteristic combination of noodle supply in adjacent time intervals.

[0105] The average of the changes in each individual change feature is used to obtain the overall change range, which is used to quantify the degree of difference in supply features between adjacent intervals.

[0106] Step S139: Calculate the correlation between the data type of feedback content and the core feature combination of noodle supply within each time interval, and determine the user feedback tendency corresponding to different supply features.

[0107] For each time interval, analyze the correlation between the data type of feedback content and the core feature combination: for example, the core feature combination for "August 1st morning market" is "handmade noodles + bone broth base". Calculate the proportion of "fresh broth" and "chewy noodles" mentioned in taste feedback within this interval (80% of taste feedback) and the proportion of "sufficient portion" mentioned in portion feedback (70% of portion feedback). Calculate the correlation between these positive feedbacks and the core feature combination (e.g., correlation degree = number of positive feedbacks / total number of noodle products covered by the core feature combination). Simultaneously, calculate the correlation between negative feedback (e.g., "salty broth") and the core feature combination, determining that the user feedback tendency corresponding to the "handmade noodles + bone broth base" feature combination is "mainly positive feedback, primarily acknowledging the taste and broth flavor".

[0108] Step S1310: Integrate the core feature combination, feature change information, feedback correlation information and feature correlation relationship of noodle supply in each time interval to form a noodle supply feature time sequence correlation set. This noodle supply feature time sequence correlation set includes time interval division, supply features of each interval, feature time sequence correlation and feedback correlation information.

[0109] The system integrates the "day-time period" division, the core feature combination of each period, feature change records (including features before and after changes and their causes), feedback correlation analysis results (distribution of feedback types, degree of correlation with core features, feedback tendency), and similarity and inheritance / change information of adjacent periods to form a structured time-series correlation set of surface file supply features. This time-series correlation set of surface file supply features is stored using a time-series data structure, supports querying supply features by time period and querying feedback tendency by feature combination, and intuitively presents the evolution of supply features over time and the correlation pattern with user feedback.

[0110] Step S140: Call the artificial intelligence model to perform interactive analysis on the evolution trajectory of user faceplate demand and the time-series correlation set of faceplate supply characteristics, and generate dynamic demand and supply matching rules.

[0111] The system calls upon an AI model built on the Transformer architecture (trained with massive amounts of demand and supply data) to generate matching rules through processes such as feature interaction, filtering, and rule generation. The specific steps are as follows.

[0112] Step S141: Convert the timeline framework, trend curve of user profile demand change, and influencing factor annotations in the user profile demand evolution trajectory into demand feature vectors that can be recognized by the artificial intelligence model.

[0113] The evolution trajectory of user profile requirements is characterized by feature encoding: the type proportion and feature distribution data in the timeline framework are converted into numerical feature components; the trend characteristics of the demand change curve (such as rising, falling, or stable) are converted into trend feature components; and the influencing factor labels are converted into categorical feature components (such as "weather factors" or "preference settings"). The above components are arranged in a preset order to form a fixed-dimensional demand feature vector, where each element corresponds to a demand feature attribute.

[0114] Step S142: Convert the time interval division, supply characteristics of each interval, feature time sequence correlation and feedback correlation information in the face stock supply feature time sequence correlation set into supply feature vectors that can be recognized by artificial intelligence models.

[0115] Feature encoding is performed on the time-series correlation set of supply characteristics: the core supply characteristic combinations within a time interval are converted into feature combination components; the similarity and magnitude of change in the time-series correlations are converted into time-series correlation components; and the degree of correlation and feedback tendency in the feedback correlation information are converted into feedback feature components. These components are then arranged in dimensional order corresponding to the demand feature vector to form a supply feature vector, ensuring dimensional matching with the demand feature vector for easy interactive analysis.

[0116] Step S143: Input the demand feature vector and the supply feature vector that the artificial intelligence model can identify into the feature interaction layer of the artificial intelligence model. The feature interaction layer uses a cross-attention mechanism to capture the correlation points between demand features and supply features in the time dimension.

[0117] The demand feature vector and the supply feature vector are input into the feature interaction layer. The cross-attention mechanism of this feature interaction layer locates the correlation between the two in the time dimension by calculating the attention weights of the demand feature and the supply feature. For example, the time feature of "preferring light soup noodles in summer" in the demand feature is highly correlated with the time feature of "supplying cold noodles and clear soup noodles in summer" in the supply feature; the preference feature of "rejecting cilantro" in the demand feature is correlated with the ingredient feature of "cilantro is out of stock" in the supply feature.

[0118] Step S144: Calculate the correlation strength between each element in the demand feature vector that the artificial intelligence model can identify and each element in the supply feature vector that the artificial intelligence model can identify through the feature interaction layer, and generate a correlation strength matrix. The element values ​​in the correlation strength matrix represent the degree of correlation between the corresponding demand feature and the supply feature.

[0119] Generate the correlation strength matrix using the following procedure:

[0120] Step S1441: Determine the number of elements in the demand feature vector that the artificial intelligence model can recognize and the demand feature type corresponding to each element in the demand feature vector that the artificial intelligence model can recognize. The demand feature type includes time demand features, preference demand features, scenario demand features and evaluation demand features.

[0121] The number of elements in the demand feature vector is counted (e.g., 100 elements), and each element is labeled with the demand feature type: for example, elements 1-20 correspond to time demand features (distribution of consumption time periods), elements 21-50 correspond to preference demand features (taste, ingredient preferences), elements 51-70 correspond to scenario demand features (scenario type, time urgency), and elements 71-100 correspond to evaluation demand features (rating stars, keyword distribution).

[0122] Step S1442: Determine the number of elements in the supply feature vector that the artificial intelligence model can recognize and the supply feature type corresponding to each element in the supply feature vector that the artificial intelligence model can recognize. The supply feature type includes time supply feature, product feature, quantity supply feature and feedback supply feature.

[0123] The number of elements in the supply feature vector is counted (consistent with the demand feature vector, 100 elements), and the supply feature types are labeled: for example, elements 1-20 correspond to time supply features (supply period, feature change time), elements 21-50 correspond to noodle features (ingredients, production method), elements 51-70 correspond to quantity supply features (supply quantity, sales distribution), and elements 71-100 correspond to feedback supply features (feedback type, degree of correlation).

[0124] Step S1443: Initialize the correlation strength matrix. The number of rows in the correlation strength matrix is ​​equal to the number of elements in the demand feature vector that the artificial intelligence model can identify. The number of columns in the correlation strength matrix is ​​equal to the number of elements in the supply feature vector that the artificial intelligence model can identify. The initial element values ​​of the correlation strength matrix are all set to the initial correlation benchmark values.

[0125] Construct a 100×100 correlation strength matrix, with all elements initially set to 0 (initial correlation baseline value), representing that demand and supply characteristics are not correlated in the initial state.

[0126] Step S1444: For each element in the demand feature vector that the artificial intelligence model can identify, extract the feature attributes, feature value range, and importance weight of the element in the user profile demand evolution trajectory.

[0127] For element 25 (preference demand feature, corresponding to "light taste preference") in the demand feature vector, the feature attribute is extracted as "taste preference type", the feature value range is "0-1" (0 represents no preference, 1 represents strong preference), and the importance weight is 0.8 (set according to the degree of influence of this feature on demand).

[0128] Step S1445: For each element in the supply feature vector that the artificial intelligence model can identify, extract the feature attributes, feature value range, and importance weight of the element in the temporal correlation set of surface supply features.

[0129] For element 25 (noodle product feature, corresponding to "proportion of clear soup base") in the supply feature vector, the feature attribute is extracted as "proportion of soup base type", the feature value range is "0-100%", and the importance weight is 0.7 (based on the representativeness of the supply according to this feature).

[0130] Step S1446: Calculate the attribute matching degree and numerical range overlap degree of the elements in the demand feature vector that the AI ​​model can identify and the elements in the supply feature vector that the AI ​​model can identify. Then, perform a weighted calculation on the attribute matching degree, numerical range overlap degree, and the importance weights of the elements in the demand feature vector and the supply feature vector that the AI ​​model can identify to obtain the correlation strength value between a single element in the demand feature vector and a single element in the supply feature vector that the AI ​​model can identify.

[0131] Calculate the attribute matching degree (both are related to taste / soup base, matching degree is 1) and numerical range overlap degree (the value of the demand element is 0.9, the value of the supply element is 0.8, and the overlap degree is 0.8) between element 25 (demand) and element 25 (supply). The association strength value is calculated as (attribute matching degree × 0.5 + numerical range overlap degree × 0.5) × (demand weight × 0.5 + supply weight × 0.5). The specific association strength value is obtained through this weighted calculation.

[0132] Step S1447: Fill the calculated association strength value into the corresponding position in the association strength matrix, replacing the initial association benchmark value in the association strength matrix.

[0133] Fill the above correlation strength value into the 25th row and 25th column of the correlation strength matrix, replacing the initial value 0.

[0134] Step S1448: Traverse all elements of the demand feature vector that the artificial intelligence model can identify and all elements of the supply feature vector that the artificial intelligence model can identify, and repeat the above process of calculating the correlation strength value until all elements of the correlation strength matrix are updated with the calculated correlation strength value to form a complete correlation strength matrix.

[0135] The correlation strength values ​​of all element pairs in the demand feature vector and the supply feature vector are calculated sequentially, and the corresponding positions in the matrix are updated to form a complete correlation strength matrix. Each element value in the matrix represents the degree of correlation between a pair of demand and supply features.

[0136] Step S145: Input the association strength matrix into the feature filtering layer of the artificial intelligence model, and use the feature filtering layer to retain feature pairs whose association strength meets the set conditions, and remove feature pairs whose association strength does not meet the set conditions.

[0137] A threshold for association strength is set (e.g., 0.5). The feature filtering layer iterates through all elements in the association strength matrix: retaining feature pairs with element values ​​≥ 0.5 (e.g., "light flavor preference - clear broth percentage" and "no cilantro preference - cilantro out of stock"), and removing feature pairs with element values ​​< 0.5 (e.g., "spicy flavor preference - clear broth percentage" and "fast demand - slow cooking method"). Simultaneously, the association strength value of the retained feature pairs and the corresponding demand-supply feature type are recorded, forming a filtered list of feature pairs. This filtered list of feature pairs only contains feature combinations with strong associations.

[0138] Step S146: Classify the retained feature pairs and divide them into different matching categories according to the demand feature type and the supply feature type. Each matching category corresponds to a type of demand and supply matching relationship.

[0139] Based on the combination of demand feature types (time demand, preference demand, scenario demand, evaluation demand) and supply feature types (time supply, noodle product characteristics, quantity supply, feedback supply) in the retained feature pairs, they are classified as follows: feature pairs combining "preference demand - noodle product characteristics," such as "light flavor preference - clear broth percentage" and "no cilantro preference - cilantro availability," are classified into the "preference-feature matching category"; feature pairs combining "time / scenario demand - time / quantity supply," such as "early market demand - early market noodle product list" and "fast demand - short serving time noodles," are classified into the "scenario-supply matching category"; and feature pairs combining "high evaluation demand - high feedback rating noodles," such as "evaluation demand - feedback supply," are classified into the "evaluation-feedback matching category." Each matching category contains multiple feature pairs, and feature pairs within the same category have similar demand-supply matching logic.

[0140] Step S147: Input the classified feature pairs into the rule generation layer of the artificial intelligence model, and use the rule generation layer to generate initial matching rules based on the feature pair attributes of each matching category, including feature association conditions, matching priority and matching result output format.

[0141] The rule generation layer generates initial rules for each matching category: For the "preference-feature matching category," rules are generated based on features to attributes, such as "If a user has a preference for light flavors (demand feature) and the proportion of clear broth in the noodle supply is ≥ a set proportion (supply feature), then this type of noodle will be matched first." The feature association condition is "light flavor preference identifier = yes and clear broth proportion ≥ a set proportion." The matching priority is set to the highest (because preferences directly affect user choice intentions), and the matching result output format is "noodle name + clear broth feature label." For the "scenario-supply matching category," rules are generated such as "If a user is in a fast-demand scenario at a morning market (demand feature)..." For the "evaluation-feedback matching" category, the generation rule is as follows: "If the user prefers highly rated noodles (demand feature) and the noodle's historical feedback rating is ≥ the set level (supply feature), then match this type of noodle." The feature association condition is "Scenario type = morning market and time urgency = high and serving time ≤ set time," the matching priority is set to medium, and the output format is "noodle name + serving time label."

[0142] Step S148: Apply the initial matching rule to the historical demand and supply data sample, obtain the matching result of the initial matching rule on the historical demand and supply data sample, analyze the degree of fit between the matching result of the initial matching rule on the historical demand and supply data sample and the historical actual consumption result, and adjust the feature association conditions and matching priority in the initial matching rule.

[0143] A sample of historical supply and demand data from the past six months (including demand data from multiple users, corresponding noodle stall supply data, and actual consumption records) was selected. Initial matching rules were applied to the sample: For user B's data of "light taste + fast demand at the morning market," the rules matched "clear broth vegetable noodles (fast service, clear broth)," which, combined with user B's actual consumption of "clear broth vegetable noodles," showed a high degree of fit. For user C's data of "light taste + high rating demand," the rules matched "clear broth beef noodles (high rating)," but user C's actual consumption was "light chicken noodles," resulting in a low degree of fit. The reason for the low fit was analyzed as follows: the initial rules did not consider user C's implicit demand for "chicken toppings," indicating incomplete feature association conditions. Accordingly, the rules were adjusted as follows: the "preference-feature matching" rule was supplemented with the "ingredient preference" association condition, and the "light flavor preference-clear soup base ratio" rule was optimized to "light flavor preference = yes and clear soup base ratio ≥ set ratio and ingredients include user-preferred ingredients"; at the same time, based on sample fit data, the priority of "preference-feature matching" was adjusted from "highest" to "higher than scenario" to ensure that it is more in line with actual consumption habits.

[0144] Step S149: Repeat the process of adjusting the initial matching rule until the initial matching rule reaches a stable state on the historical demand and supply data sample, and determine the initial matching rule that has reached a stable state as the dynamic demand and supply matching rule.

[0145] Repeat the "apply rules - analyze fit - adjust rules" process: apply the optimized rules to historical samples again. If the fit improves and the fit fluctuation range is ≤ the set threshold after multiple consecutive applications (indicating that the rules are stabilizing), then stop adjusting. The final dynamic demand-supply matching rules include three types of optimized sub-rules: "preference-feature matching sub-rules" (including multi-dimensional preference association conditions such as taste, ingredients, and texture, with the highest priority), "scenario-supply matching sub-rules" (including scenario association conditions such as time period and urgency, with medium priority), and "evaluation-feedback matching sub-rules" (including association conditions such as ratings and feedback keywords, with the second highest priority). Each sub-rule can be dynamically adapted according to changes in demand-supply data.

[0146] Step S150: Iteratively optimize the dynamic demand and supply matching rule using user feedback data on the output results of the dynamic demand and supply matching rule, and generate personalized recommendation results based on the optimized dynamic demand and supply matching rule.

[0147] The rules are continuously optimized based on user feedback to generate the final recommendation results. The specific steps are as follows.

[0148] Step S151: Obtain the initial recommendation result output by the dynamic demand and supply matching rule. The initial recommendation result includes the recommended noodle name, the recommended noodle feature description and the recommendation basis.

[0149] Input the current user demand data and noodle stall supply data into the dynamic demand and supply matching rules, and output the initial recommendation results: For example, for user A's "light taste, no coriander, quick morning market demand", the recommendation results include "Recommended noodle dish 1: Clear soup vegetable noodles (features: handmade noodles, clear soup base, no coriander, short serving time; recommendation basis: meets the preference for light taste, the requirement of no coriander and the condition of quick morning market supply)" and "Recommended noodle dish 2: Shredded chicken clear soup noodles (features: handmade noodles, chicken soup base, no coriander, high rating; recommendation basis: meets the preference and high rating demand)".

[0150] Step S152: Present the initial recommendation results to the user and collect the user's feedback operation data on the initial recommendation results. The feedback operation data includes the user's selection data of the recommended noodle products, the user's evaluation data of the basis for the recommendation, and the user's supplementary description of needs.

[0151] The initial recommendation results are presented in a list format via the mini-program, and user feedback is collected: If a user selects "clear soup vegetable noodles" (selection data), the evaluation of the recommendation basis is "the description of the meal preparation time is accurate, but the portion size is not mentioned" (evaluation data), and the supplementary requirement is described as "I need a regular portion size" (supplementary data); if the user does not select a recommended noodle dish and enters "I want noodles with a bit of chewiness", the selection data is "no recommended noodle dish selected", and the supplementary requirement is described as "I prefer chewy noodles".

[0152] Step S153: Classify and organize the feedback operation data, mark the recommended noodles selected by the user as positive feedback noodles, and mark the recommended noodles not selected by the user as negative feedback noodles, and extract the evaluation keywords of the user's basis for the recommendation and the keywords of the user's supplementary demand description.

[0153] Feedback data was categorized and organized: "Clear Broth Vegetable Noodles" was marked as a positive feedback noodle dish, and "Chicken Broth Noodles" was marked as a negative feedback noodle dish; the keyword "missing portion size" was extracted from the evaluation data, and the keywords "regular portion size" and "chewy noodles" were extracted from the supplementary descriptions. The time of the feedback, the user's ID, and the corresponding recommendation result ID were also recorded for easy correlation analysis.

[0154] Step S154: Analyze the feature association conditions in the dynamic demand and supply matching rules corresponding to the positive feedback noodle products, and statistically analyze the frequency and association strength of the feature association conditions in the positive feedback cases.

[0155] The analysis focused on the rule-related conditions for positive feedback regarding the noodle dish "clear broth vegetable noodles": "light flavor = yes", "no cilantro = yes", and "serving time ≤ set value". The frequency of these three conditions was analyzed across all positive feedback cases over the past week: "light flavor = yes" appeared most frequently, followed by "serving time ≤ set value". The correlation strength between each condition and the positive feedback was calculated (correlation strength = number of positive cases containing that condition / total number of positive cases), identifying "light flavor" and "no cilantro" as strong correlation conditions.

[0156] Step S155: Analyze the feature association conditions in the dynamic demand supply matching rules corresponding to non-positive feedback products, find the contradictions between the feature association conditions and user feedback, and determine the feature association conditions in the dynamic demand supply matching rules that need to be adjusted.

[0157] Analyzing the non-positive feedback for the noodle dish "Chicken Broth Noodles," the rule conditions were: "Light flavor = Yes," "No cilantro = Yes," and "Rating ≥ Set Level." Combining this with the user's unselected feedback and supplementary request for "chewy noodles," a contradiction was identified: the noodle type for this dish is "regular ramen," failing to meet the user's implicit "chewy" texture requirement, while the initial rule did not include a "texture preference" related condition. Therefore, the rule needs adjustment: a "texture preference" feature related condition needs to be added to the "Preference-Feature Matching Sub-Rule."

[0158] Step S156: Convert the user-supplemented demand description keywords into new demand features, determine the correlation between the new demand features and existing supply features, and add the correlation to the dynamic demand and supply matching rules.

[0159] The supplementary demand keywords "regular portion" and "chewy noodles" are converted into new demand features: "portion preference = regular" and "texture preference = chewy". The relationship between the new demand features and supply features is determined: "portion preference = regular" is associated with the supply feature "noodle portion = regular", and "texture preference = chewy" is associated with the supply feature "noodle type = handmade noodles (chewy)". The above relationships are added to the "preference-feature matching sub-rule", and the optimized rule conditions are "light taste = yes and no cilantro = yes and portion = regular and noodle type = handmade noodles (chewy)".

[0160] Step S157: Adjust the correlation strength of feature correlation conditions in the dynamic demand and supply matching rules, increase the correlation strength of feature correlation conditions that occur more frequently than the set frequency in positive feedback cases, and decrease the correlation strength of feature correlation conditions that contradict non-positive feedback.

[0161] Based on the statistical results of positive feedback cases, the correlation strength of the conditions "light taste" and "no cilantro" was increased from the original set value (because the frequency of occurrence was higher than the set frequency); for cases where the condition "rating ≥ set level" in non-positive feedback has no direct correlation with user choices, the correlation strength of this condition was reduced to avoid over-reliance on ratings and ignoring key preferences such as taste.

[0162] Step S158: Adjust the matching priority in the dynamic demand and supply matching rules, increase the priority of the matching category corresponding to positive feedback noodle products, and decrease the priority of the matching category corresponding to non-positive feedback noodle products.

[0163] Based on the feedback that the user chose "clear broth vegetable noodles" (corresponding to "preference-feature matching category") instead of "chicken shredded clear broth noodles" (corresponding to "evaluation-feedback matching category"), the priority of "preference-feature matching category" is further increased from "highest" to "evaluation-feedback matching category" to ensure that the user's core preferences are met first.

[0164] Step S159: Apply the adjusted dynamic demand and supply matching rules to the new demand and supply data sample, obtain the matching results of the adjusted dynamic demand and supply matching rules on the new demand and supply data sample, and collect user feedback data on the matching results.

[0165] Select a new sample of user demand supply data (such as user D's "light, chewy, and regular portion" requirements), apply the adjusted rules, match "handmade chewy clear soup noodles (regular portion)", present the result to the user and collect feedback: if the user selects the noodle product and rates it "meets their needs", it is considered positive feedback.

[0166] Step S1510: Compare the user feedback satisfaction of the dynamic demand and supply matching rules before and after the adjustment on the same data sample. If the user feedback satisfaction of the adjusted dynamic demand and supply matching rules is higher than that of the original dynamic demand and supply matching rules, then retain the adjusted dynamic demand and supply matching rules. If the user feedback satisfaction of the adjusted dynamic demand and supply matching rules is lower than that of the original dynamic demand and supply matching rules, then adjust the parameters of the dynamic demand and supply matching rules in reverse until the user feedback satisfaction of the dynamic demand and supply matching rules reaches the set level.

[0167] Compare and optimize satisfaction levels using the following process:

[0168] Step S1510-1: Select the demand and supply data samples that were applied to the dynamic demand and supply matching rules before the adjustment as the comparison sample set. The comparison sample set contains the demand data of multiple users and the corresponding surface supply data.

[0169] One hundred user demand supply samples that were used before the rule was adjusted were selected as the comparison sample set, covering different demand types (preferences, scenarios, evaluation needs) and supply scenarios.

[0170] Step S1510-2: Apply the adjusted dynamic demand and supply matching rules to the comparison sample set to generate the adjusted recommendation results of the adjusted dynamic demand and supply matching rules on each sample in the comparison sample set.

[0171] The adjusted rules are applied to each of the 100 samples to generate an adjusted recommendation for each sample. For example, for users in the sample who want "light flavor, no cilantro, and chewy texture", the adjusted recommendation is "handmade chewy clear soup noodles", while the original recommendation is "regular clear soup noodles".

[0172] Step S1510-3: Retrieve the pre-adjustment recommendation results generated by the dynamic demand and supply matching rules on the comparison sample set, as well as the user feedback data collected at that time.

[0173] Retrieve recommendation results and feedback data from the database when the rules were applied to the sample set before the adjustment. For example, the user selection rate for "plain clear soup noodles" was 60%, and the satisfaction rating was "average".

[0174] Step S1510-4: Collect user feedback data on the adjusted recommendation results. This feedback data includes the recommended noodle selection rate, the degree of acceptance of the recommendation basis, and the demand satisfaction score.

[0175] Feedback on the adjusted recommendation results: The selection rate for "Handmade Chewy Clear Broth Noodles" was 85%, the acceptance rate of the recommendation basis (the proportion of users who believe that the recommendation reasons meet their needs) was 90%, and the average score for meeting needs (1-5 points) was 4.5 points.

[0176] Step S1510-5: Calculate the user feedback satisfaction corresponding to the dynamic demand and supply matching rule before adjustment. The user feedback satisfaction calculation is based on the recommended noodle selection rate before adjustment, the recognition of the recommendation basis before adjustment, and the demand satisfaction score before adjustment. A weighted summation method is used, and the weighting weight is set according to the importance of each indicator.

[0177] Set the selection rate weight to 0.4, the approval weight to 0.3, and the satisfaction rating weight to 0.3. Calculate the satisfaction level before adjustment as follows: 60% × 0.4 + 70% × 0.3 + 3.5 × 0.2 (calculated after converting the rating from 1 to 5 to the 0-1 range) = satisfaction value before adjustment.

[0178] Step S1510-6: Calculate the user feedback satisfaction corresponding to the adjusted dynamic demand and supply matching rule. The user feedback satisfaction is calculated using the same weighted summation method as the user feedback satisfaction corresponding to the dynamic demand and supply matching rule before the adjustment, based on the adjusted recommended noodle selection rate, the adjusted recognition of the recommendation basis, and the adjusted demand satisfaction score.

[0179] Adjusted satisfaction rate calculated with equal weights = 85% × 0.4 + 90% × 0.3 + 4.5 × 0.2 = Adjusted satisfaction rate value.

[0180] Step S1510-7: Compare the user feedback satisfaction with the dynamic demand-supply matching rule before and after the adjustment for each sample in the comparison sample set, and count the number of samples with improved user feedback satisfaction, the number of samples with decreased user feedback satisfaction, and the number of samples with unchanged user feedback satisfaction.

[0181] Statistical results: 80 samples showed an increase in satisfaction, 15 samples remained unchanged, and 5 samples showed a decrease.

[0182] Step S1510-8: Calculate the overall improvement in user feedback satisfaction. This overall improvement in user feedback satisfaction is the sum of the differences between the user feedback satisfaction of the adjusted dynamic demand supply matching rule and the user feedback satisfaction of the unadjusted dynamic demand supply matching rule for all samples in the comparison sample set, divided by the total number of samples in the comparison sample set.

[0183] The improvement rate is calculated as follows: (sum of satisfaction differences among all samples) / 100 = improvement rate value.

[0184] Step S1510-9: If the overall improvement in user feedback satisfaction is greater than the set improvement threshold, and the proportion of the number of samples with improved user feedback satisfaction to the total number of samples in the comparison sample set is greater than the set proportion threshold, then the adjusted dynamic demand supply matching rule is determined to be better than the original dynamic demand supply matching rule.

[0185] If the increase is greater than the set threshold (e.g., 5%) and the increase in the sample proportion (80%) is greater than the set proportion threshold (70%), then the adjusted rule is considered better and is retained.

[0186] Step S1510-10: If the above conditions are not met, record the demand and supply characteristics corresponding to the samples where user feedback satisfaction has decreased, and analyze the reasons for the inapplicability of the adjusted dynamic demand and supply matching rules in the samples.

[0187] If the target is not met, analyze the characteristics of the five declining samples: if they are all users with the "high score priority" requirement, the adjusted rule reduced the weight of the scoring condition, causing matching bias, and the weight of this condition needs to be adjusted in reverse.

[0188] Step S1511: Generate personalized recommendation results for face files based on the optimized dynamic demand and supply matching rules. The specific process is as follows.

[0189] For example, step S1511-1: Convert the current user's multi-dimensional profile demand data into a current demand feature vector that conforms to the optimized dynamic demand supply matching rule input format.

[0190] Convert User A's real-time requirements ("light flavor, no cilantro, chewy texture, quick at the morning market, regular portion") into a requirement feature vector, which includes feature components such as "light flavor = 1, no cilantro = 1, chewy texture = 1, morning market scene = 1, time urgency = 1, regular portion = 1", with the format consistent with the rule input requirements.

[0191] Step S1511-2: Convert the supply data of the current profile into a current supply feature vector that conforms to the input format of the optimized dynamic demand and supply matching rules.

[0192] The current supply data of the stores ("Handmade chewy clear soup noodles: clear soup base, no coriander, regular portion, short serving time, high rating; Chicken shredded clear soup noodles: clear soup base, no coriander, regular ramen, medium serving time, high rating") are converted into supply feature vectors, which include feature components such as the flavor, ingredients, texture, serving time, and rating of each noodle dish.

[0193] Step S1511-3: Input the current demand feature vector and the current supply feature vector into the optimized dynamic demand and supply matching rule. The optimized dynamic demand and supply matching rule selects the supply features that match the current demand features based on the feature association conditions.

[0194] The rules, based on the conditions "light flavor = 1, no cilantro = 1, chewy texture = 1, regular portion size = 1, and short serving time = 1", selected the supply characteristics of "handmade chewy clear soup noodles".

[0195] Step S1511-4: Sort the noodle products corresponding to the selected supply characteristics according to the matching priority in the optimized dynamic demand supply matching rules, and extract the detailed information of the top K noodle products. The detailed information includes the noodle product name, noodle product ingredients, noodle product preparation method, noodle product flavor description and the suitable dietary scenarios for the noodle product.

[0196] Based on the optimized matching priority (preference-feature matching > scenario-supply matching > evaluation-feedback matching), the selected noodle dishes, such as "Handmade Chewy Clear Broth Noodles," "Chewy Vegetable Clear Broth Noodles," and "Quick Handmade Clear Broth Noodles," were ranked as follows: "Handmade Chewy Clear Broth Noodles" ranked first because it simultaneously met all preference and scenario conditions of "light flavor, no cilantro, chewy texture, regular portion size, and short serving time"; "Chewy Vegetable Clear Broth Noodles" ranked second because it lacked the scenario condition of "short serving time"; and "Quick Handmade Clear Broth Noodles" ranked third because its ingredients did not include vegetables, which are a potential preference of users. Detailed information for the top three ranked noodle dishes (K=3) was extracted: "Handmade Chewy Clear Broth Noodles"—Ingredients: Hand-pulled noodles, beef bone broth, greens, ginger slices; Preparation method: Hand-pulled noodles, slow-cooked bone broth followed by quick portioning; Flavor description: Light and fresh, without irritating seasonings; Suitable scenarios: Office building breakfast, quick meal for one person. "Chewy Vegetable Clear Broth Noodles" - Ingredients: Hand-pulled noodles, chicken bone broth, various seasonal vegetables, scallions; Preparation: Hand-pulled noodles, freshly cooked in clear broth; Flavor: Light and refreshing, with a rich vegetable aroma; Suitable for: Community lunches, healthy light meals. "Quick Handmade Clear Broth Noodles" - Ingredients: Hand-pulled noodles, concentrated clear broth, egg, greens; Preparation: Pre-made handmade noodles, ready to eat after heating; Flavor: Light and convenient, chewy texture; Suitable for: Takeout delivery, meals with limited time.

[0197] Step S1511-5: Based on the matching criteria in the optimized dynamic demand supply matching rules, generate a recommendation reason for each recommended noodle product. The recommendation reason includes the specific matching points between the noodle product and the user's demand characteristics.

[0198] The top-ranked "Handmade Chewy Clear Broth Noodles" recommendation reason is: "This noodle dish uses hand-pulled noodles, satisfying your preference for a 'chewy' texture; the broth is beef bone clear soup, meeting your 'light' taste requirements; the toppings do not include cilantro, matching your 'cilantro-free' requirement; the portion size is standard, suitable for your portion size needs; the short serving time fits the 'quick morning market' scenario, making it the highest overall match." The "Chewy Vegetable Clear Broth Noodles" recommendation reason is: "This noodle dish uses hand-pulled noodles, satisfying your preference for a 'chewy' texture; the chicken bone clear broth meets the 'light' taste requirements and does not include cilantro; the toppings include a variety of seasonal vegetables, rich in nutrition; however, the serving time is slightly longer than the top recommendation, making it the second highest overall match." The "Quick Handmade Clear Broth Noodles" recommendation reason is: "This noodle dish uses hand-pulled noodles, satisfying your preference for a 'chewy' texture; the concentrated clear broth meets the 'light' taste requirements and does not include cilantro; the fast serving speed fits the 'quick' scenario; however, the toppings have fewer vegetable varieties, making it the third highest overall match."

[0199] Step S1511-6: Organize the detailed information and reasons for recommending the recommended noodles into a structured recommendation list in sorted order. Each record in the structured recommendation list contains complete information about a recommended noodle and the corresponding reason for recommending it.

[0200] The structured recommendation list is organized in the format of "Rank - Noodle Information - Recommendation Reason": Rank 1 - Noodle Name: Handmade Chewy Clear Broth Noodles; Ingredients: Hand-pulled noodles, beef bone broth, green vegetables, ginger slices; Preparation Method: Hand-pulled noodles, slow-cooked bone broth followed by quick portioning; Flavor Description: Light and fresh, no pungent seasonings; Suitable Scenarios: Office breakfast, quick meal for one person; Recommendation Reason: [See corresponding content in step S1511-5 for details]. Rank 2 - Noodle Name: Chewy Vegetable Clear Broth Noodles; Ingredients: Hand-pulled noodles, chicken bone broth, various seasonal vegetables, scallions; Preparation Method: Hand-pulled noodles, freshly cooked in clear broth; Flavor Description: Light and refreshing, with a rich and fragrant vegetable flavor; Suitable Scenarios: Community lunch, healthy light meal needs; Recommendation Reason: [See corresponding content in step S1511-5 for details]. Rank 3 – Noodle Name: Quick Handmade Clear Broth Noodles; Ingredients: Hand-pulled noodles, concentrated clear broth, egg, greens; Preparation Method: Pre-made hand-pulled noodles, ready to eat after heating; Flavor Description: Light and convenient, chewy texture; Suitable Scenarios: Takeout delivery, time-sensitive meals; Reason for Recommendation: [See steps S1511-5 for details].

[0201] Step S1511-7: Integrate the noodle information in the structured recommendation list, and determine the integrated structured recommendation list as the noodle personalized recommendation result. The noodle personalized recommendation result includes recommended noodles sorted by matching priority, detailed information of each recommended noodle, and corresponding recommendation reasons.

[0202] The noodle information in the structured recommendation list is integrated according to three dimensions: "basic information - feature matching - scenario adaptation". The basic information dimension summarizes the noodle name, ingredients, and preparation method; the feature matching dimension extracts the matching items between each noodle and user preferences (light, no cilantro, chewy, etc.); the scenario adaptation dimension clarifies the suitable dining scenarios and serving characteristics. After integration, a complete personalized recommendation result for noodle stalls is formed, ensuring that the information hierarchy is clear, including detailed information of individual noodle items, and intuitively reflecting the matching differences between different noodle items.

[0203] Step S1511-8: Output the personalized recommendation results of the face profile in a displayable format.

[0204] The output format is selected based on the user's usage scenario: If the user orders through a mini-program, it is displayed in a card-style list format, with each recommended noodle dish corresponding to a card. The top of the card displays the sorting position and a "Recommended" label, the middle displays the noodle dish name, flavor, and suitable occasions, and the bottom presents the recommendation reason and an "Add to Cart" button. Clicking the card expands to view complete ingredient and preparation details. If the user operates through a self-service ordering machine in a store, it is displayed in a split-screen format, with the left side showing the recommendation list and the right side showing detailed information and images of the selected noodle dish, supporting touch-screen swiping to switch between noodle dishes. If it is pushed by a food delivery platform, it is sent to the user's message center in a concise text list format. Each recommendation includes the noodle dish name, key matching points, and a jump link, which can be clicked to directly enter the order page. During the output process, the noodle dish images and ingredient information in the recommendation results are checked for compliance to ensure there is no false advertising content, while protecting the user's ordering preferences and other privacy data, and not disclosing them to third parties.

[0205] Figure 2 The diagram illustrates exemplary hardware and software components of a smart face profile personalized recommendation system 100 that can implement the ideas of this application, according to some embodiments of this application. For example, a processor 120 can be used in the smart face profile personalized recommendation system 100 and to perform the functions described in this application.

[0206] The intelligent face photo personalized recommendation system 100 can be a general-purpose server or a special-purpose server; both can be used to implement the intelligent face photo personalized recommendation method of this application. Although only one server is shown in this application, for convenience, the functions described in this application can be implemented in a distributed manner on multiple similar platforms to balance the load.

[0207] For example, the smart face card personalized recommendation system 100 may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and various forms of storage media 140, such as a disk, ROM, or RAM, or any combination thereof. Exemplarily, the smart face card personalized recommendation system 100 may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The methods of this application can be implemented according to the program instructions. The smart face card personalized recommendation system 100 also includes an I / O interface 150 between the computer and other input / output devices.

[0208] For ease of explanation, only one processor is described in the intelligent facial recognition personalized recommendation system 100. However, it should be noted that the intelligent facial recognition personalized recommendation system 100 of this application may also include multiple processors, and therefore the steps performed by one processor described in this application may also be performed jointly or individually by multiple processors. For example, if the processor of the intelligent facial recognition personalized recommendation system 100 performs steps A and B, it should be understood that steps A and B may also be performed jointly by two different processors or individually by one processor. For example, the first processor performs step A, the second processor performs step B, or the first processor and the second processor jointly perform steps A and B.

[0209] Furthermore, this embodiment of the invention also provides a readable storage medium, wherein computer-executable instructions are preset in the readable storage medium, and when the processor executes the computer-executable instructions, the above-mentioned smart face personalized recommendation method is implemented.

[0210] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A personalized recommendation method for smart noodle stalls, characterized in that, The method includes: Collect multi-dimensional noodle stall demand data and noodle stall supply data from users. The multi-dimensional noodle stall demand data from users includes historical noodle stall consumption data, real-time dietary demand data, and dietary preference change data. The noodle stall supply data includes real-time noodle product supply data, noodle product characteristic change data, and noodle product consumption feedback data. Based on multi-dimensional user profile demand data, a user profile demand evolution trajectory is constructed. This trajectory is used to present the changing trends and correlations of user profile demand in different time periods. Based on the noodle stall supply data, a time-series correlation set of noodle stall supply characteristics is constructed. This time-series correlation set of noodle stall supply characteristics is used to present the correlation patterns and changing rules of noodle stall supply characteristics in different time periods. The artificial intelligence model is invoked to perform interactive analysis on the evolution trajectory of user faceplate demand and the time-series correlation set of faceplate supply characteristics, and to generate dynamic demand and supply matching rules. The dynamic demand and supply matching rules are iteratively optimized by utilizing user feedback data on the output results of the dynamic demand and supply matching rules, and personalized recommendation results are generated based on the optimized dynamic demand and supply matching rules.

2. The personalized recommendation method for smart noodle stalls according to claim 1, characterized in that, The construction of the user profile requirement evolution trajectory based on multi-dimensional user profile requirement data includes: Extract consumption time data, noodle type data, noodle characteristic data, and consumption evaluation data from users' multi-dimensional noodle demand data; Extract current dietary scenario data, current dietary demand description data, and current dietary preference data from real-time dietary demand data from multi-dimensional user demand data; Extract data on the time points of preference changes, comparison data before and after preference changes, and data on factors influencing preference changes from multi-dimensional user demand data; Arrange the consumption time data in chronological order to establish a timeline framework, and map the consumption type data, consumption characteristic data, and consumption evaluation data of the corresponding time points to the corresponding positions in the timeline framework. Mark the current dietary scenario data, current dietary demand description data, and current dietary preference data in the real-time dietary demand data at the latest time node position in the timeline frame; Analyze the time nodes of preference changes in the dietary preference change data, mark the time nodes of preference changes in the corresponding positions of the time axis frame, and associate them with the comparison data of preference changes before and after the time nodes of preference changes; Based on the data at each time point in the time axis framework, the rate of change of consumer noodle types, the similarity of consumer noodle characteristics, and the fluctuation value of consumer evaluation are calculated between adjacent time points. Based on the calculated rate of change, similarity, and fluctuation value, a trend curve of user profile demand change is constructed. The horizontal axis of the trend curve of user profile demand change is the time node, and the vertical axis is the quantitative indicator of demand change. The data on factors influencing changes in dietary preferences are correlated with the fluctuation nodes in the trend curve of changes in user noodle demand, and the influencing factors corresponding to each fluctuation node are labeled. By integrating the timeline framework, the trend curve of user profile demand changes, and the annotation of influencing factors, a user profile demand evolution trajectory is formed. This trajectory includes the distribution of demand data over time, the trend of user profile demand changes, and the correlation information of influencing factors.

3. The personalized recommendation method for smart noodle stalls according to claim 2, characterized in that, The calculation of the rate of change of consumer noodle types, the similarity of consumer noodle characteristics, and the fluctuation value of consumer evaluation between adjacent time nodes, based on data from each time node in the time axis framework, includes: Extract the consumption noodle product type data between two adjacent time nodes in the timeline frame, count the number and specific types of noodle product types consumed at the previous time node, and count the number and specific types of noodle product types consumed at the next time node. Calculate the number of new noodle product types and the number of noodle product types that disappeared at the next time point compared to the previous time point. Divide the sum of the number of new noodle product types and the number of noodle product types that disappeared by the number of noodle product types at the previous time point to obtain the rate of change of noodle product types. Extract the consumption feature data of noodles from two adjacent time points. This consumption feature data includes noodle flavor features, noodle texture features, noodle ingredient features, and noodle portion features. For each type of noodle product feature, the noodle product feature data of adjacent time nodes are converted into standardized feature vectors, the similarity value between two standardized feature vectors is calculated, and the average of the similarity values ​​corresponding to all noodle product features is taken as the similarity of the consumption noodle product features. Extract consumer review data from two adjacent time points. This consumer review data includes review scores, the number of positive words in the review text, and the number of negative words in the review text. Calculate the difference between the evaluation score at the next time point and the evaluation score at the previous time point, and calculate the difference between the ratio of the number of positive words to the number of negative words at the next time point and the ratio of the number of positive words to the number of negative words at the previous time point. The difference between the evaluation scores and the difference between the vocabulary ratios are weighted and summed. The weighting is set according to the importance of the evaluation scores and vocabulary ratios in the consumer evaluation, and the fluctuation value of the consumer evaluation is obtained. Record the calculated rate of change of consumer noodle product type, similarity of consumer noodle product characteristics, and fluctuation value of consumer evaluation, and store them in association with the corresponding adjacent time nodes; The calculation results of all adjacent time node pairs are summarized to form a set of quantitative indicators of changes in user profile requirements for each time period in the time axis framework.

4. The personalized recommendation method for smart noodle stalls according to claim 1, characterized in that, The construction of a time-series correlation set of surface-area supply characteristics based on surface-area supply data includes: Extract the supply time data, noodle name data, noodle ingredient data, noodle preparation method data, and noodle quantity data from the noodle stall supply data in real-time noodle supply data; Extract the time data of feature changes, the feature data of the product before changes, the feature data of the product after changes, and the data on the reasons for feature changes from the noodle supply data. Extract feedback time data, noodle product identification data, feedback content data, and corresponding consumption quantity data from the noodle supply data; The supply time data is divided into multiple consecutive time intervals in chronological order, and the distribution of the supply noodle product name data, supply noodle product ingredient data, supply noodle product preparation method data, and supply noodle product quantity data in each time interval is statistically analyzed. For each time interval, analyze the correspondence between the data on the ingredients of the supplied noodles and the data on the methods of producing the supplied noodles, and determine the core characteristic combination of the noodles supply within that time interval; Map the feature change time data in the noodle product feature change data to the divided time interval, and associate the noodle product feature data before the change, the noodle product feature data after the change, and the feature change reason data within the time interval. The feedback time data is mapped to the divided time intervals, and the data types and proportions of feedback content corresponding to different feedback product identification data are statistically analyzed within each time interval. Calculate the similarity of core feature combinations of noodle supply within adjacent time intervals to determine the inheritance relationship and variation range of noodle supply features; Calculate the correlation between the data type of feedback content and the core feature combination of noodle supply within each time interval, and determine the user feedback tendency corresponding to different supply features; By integrating the core feature combination, feature change information, feedback correlation information, and feature correlation relationship of noodle supply in each time interval, a noodle supply feature time sequence correlation set is formed. This noodle supply feature time sequence correlation set includes time interval division, supply features of each interval, feature time sequence correlation, and feedback correlation information.

5. The personalized recommendation method for smart noodle stalls according to claim 4, characterized in that, The calculation of the similarity of core feature combinations of noodle supply within adjacent time intervals, and the determination of the inheritance relationship and variation range of noodle supply features, includes: Extract the core feature combination of noodle supply between two adjacent time intervals. Each core feature combination of noodle supply includes a subset of noodle ingredient features, a subset of noodle preparation method features, and a subset of noodle quantity distribution features. Each subset of features in the core feature combination of the noodle supply in the previous time interval is converted into a feature matrix. The rows of the feature matrix represent the feature type, and the columns represent the specific attribute values ​​of the features. Transform each feature subset in the core feature combination of the noodle supply in the next time interval into a feature matrix with the same dimension as the feature matrix of the previous time interval. For each feature subset, a matrix similarity algorithm is used to calculate the similarity value of feature matrices in adjacent time intervals. This matrix similarity algorithm calculates the overall similarity based on the correspondence between matrix elements. The average of the similarity values ​​of all feature subsets is taken as the overall similarity of the core feature combination of noodle supply in adjacent time intervals. If the overall similarity is higher than the set inheritance threshold, it is determined that the core feature combination of the noodle supply in the later time interval has inherited the core feature combination of the noodle supply in the previous time interval, and the inherited feature subset and specific features are recorded. If the overall similarity is lower than the set inheritance threshold, then analyze the difference feature subsets of the core feature combination of the noodle supply in the two time intervals, and count the number of feature changes and the type of feature attributes in the difference feature subsets. Calculate the magnitude of attribute value change for each variable feature in the subset of variable features. Divide the difference between the attribute value in the next time interval and the attribute value in the previous time interval for each variable feature by the attribute value in the previous time interval to obtain the magnitude of change for a single variable feature. The average value of the changes in all the changing characteristics is taken as the overall change in the combination of core characteristics of noodle supply in adjacent time intervals. Record the similarity of core feature combinations of noodle supply in adjacent time intervals, the results of inheritance relationship determination, and the overall change range, and associate them with the corresponding time interval of the noodle supply feature time sequence association set.

6. The personalized recommendation method for smart noodle stalls according to claim 1, characterized in that, The process involves using an artificial intelligence model to interactively analyze the evolution trajectory of user faceplate demand and the time-series correlation set of faceplate supply characteristics, generating dynamic demand-supply matching rules, including: The timeline framework, trend curve of user profile demand change, and influencing factor annotation in the user profile demand evolution trajectory are converted into demand feature vectors that can be recognized by artificial intelligence models. The time interval division, supply characteristics of each interval, feature time sequence correlation and feedback correlation information in the face stock supply feature time sequence correlation set are converted into supply feature vectors that can be recognized by artificial intelligence models. The demand feature vector and the supply feature vector that can be identified by the artificial intelligence model are input into the feature interaction layer of the artificial intelligence model. The feature interaction layer uses a cross-attention mechanism to capture the correlation points between demand features and supply features in the time dimension. The correlation strength matrix is ​​generated by calculating the correlation strength between each element in the demand feature vector that the artificial intelligence model can identify and each element in the supply feature vector that the artificial intelligence model can identify through the feature interaction layer. The element values ​​in the correlation strength matrix represent the degree of correlation between the corresponding demand feature and the supply feature. The correlation strength matrix is ​​input into the feature filtering layer of the artificial intelligence model. The feature filtering layer retains feature pairs whose correlation strength meets the set conditions and removes feature pairs whose correlation strength does not meet the set conditions. The retained feature pairs are classified and divided into different matching categories according to the type of demand feature and the type of supply feature. Each matching category corresponds to a type of demand and supply matching relationship. The classified feature pairs are input into the rule generation layer of the artificial intelligence model. The rule generation layer uses the feature pair attributes of each matching category to generate initial matching rules that include feature association conditions, matching priorities, and matching result output formats. The initial matching rule is applied to the historical demand and supply data sample to obtain the matching result of the initial matching rule on the historical demand and supply data sample. The degree of fit between the matching result of the initial matching rule on the historical demand and supply data sample and the historical actual consumption result is analyzed, and the feature association conditions and matching priority in the initial matching rule are adjusted. The process of repeatedly adjusting the initial matching rules continues until the initial matching rules reach a stable state in terms of their fit with historical demand and supply data samples. The initial matching rules that have reached a stable state are then identified as dynamic demand and supply matching rules.

7. The personalized recommendation method for smart noodle stalls according to claim 6, characterized in that, The step of calculating the correlation strength between each element in the demand feature vector recognizable by the AI ​​model and each element in the supply feature vector recognizable by the AI ​​model through the feature interaction layer, and generating a correlation strength matrix, includes: Determine the number of elements in the demand feature vector that the artificial intelligence model can recognize and the demand feature type corresponding to each element in the demand feature vector that the artificial intelligence model can recognize. The demand feature type includes time demand features, preference demand features, scenario demand features and evaluation demand features. Determine the number of elements in the supply feature vector that the artificial intelligence model can recognize and the type of supply feature corresponding to each element in the supply feature vector that the artificial intelligence model can recognize. The type of supply feature includes time supply feature, product feature, quantity supply feature and feedback supply feature. Initialize the association strength matrix. The number of rows in the association strength matrix is ​​equal to the number of elements in the demand feature vector that the artificial intelligence model can identify. The number of columns in the association strength matrix is ​​equal to the number of elements in the supply feature vector that the artificial intelligence model can identify. The initial element values ​​of the association strength matrix are all set to the initial association baseline values. For each element in the demand feature vector that can be identified by the artificial intelligence model, extract the feature attributes, feature value range, and importance weight of the element in the user profile demand evolution trajectory. For each element in the supply feature vector that can be identified by the artificial intelligence model, extract the feature attributes, feature value range, and importance weight of the element in the temporal correlation set of surface supply features. The attribute matching degree and numerical range overlap of the elements in the demand feature vector that can be identified by the artificial intelligence model and the supply feature vector that can be identified by the artificial intelligence model are calculated. The attribute matching degree, numerical range overlap and the importance weight of the elements in the demand feature vector that can be identified by the artificial intelligence model and the importance weight of the elements in the supply feature vector that can be identified by the artificial intelligence model are weighted and calculated to obtain the correlation strength value of a single element in the demand feature vector that can be identified by the artificial intelligence model and a single element in the supply feature vector that can be identified by the artificial intelligence model. Fill the calculated association strength value into the corresponding position in the association strength matrix, replacing the initial association reference value in the association strength matrix; Iterate through all elements of the demand feature vector and the supply feature vector that the AI ​​model can identify, repeating the above process of calculating the correlation strength value until all elements of the correlation strength matrix are updated with the calculated correlation strength value, forming a complete correlation strength matrix.

8. The personalized recommendation method for smart noodle stalls according to claim 1, characterized in that, The iterative optimization of the dynamic demand-supply matching rule using user feedback data on the output results of the dynamic demand-supply matching rule includes: Obtain the initial recommendation results output by the dynamic demand and supply matching rules. These initial recommendation results include the recommended noodle names, descriptions of recommended noodle features, and the basis for the recommendation. The initial recommendation results are presented to the user, and the user's feedback data on the initial recommendation results is collected. This feedback data includes the user's selection data of the recommended noodle products, the user's evaluation data of the basis for the recommendation, and the user's supplementary description of needs. The feedback data is categorized and organized. Recommended noodles selected by users are marked as positive feedback noodles, and recommended noodles not selected by users are marked as negative feedback noodles. Keywords for users' evaluation of the basis for the recommendations and keywords for users' supplementary descriptions of their needs are extracted. Analyze the feature association conditions in the dynamic demand and supply matching rules corresponding to positive feedback noodle products, and statistically analyze the frequency and association strength of the feature association conditions in positive feedback cases; Analyze the feature association conditions in the dynamic demand and supply matching rules corresponding to non-positive feedback noodle products, find the contradictions between the feature association conditions and user feedback, and determine the feature association conditions in the dynamic demand and supply matching rules that need to be adjusted. The user-supplemented demand description keywords are converted into new demand features, the correlation between the new demand features and existing supply features is determined, and the correlation is added to the dynamic demand and supply matching rules. Adjust the correlation strength of feature correlation conditions in the dynamic demand and supply matching rules, increase the correlation strength of feature correlation conditions that occur more frequently than the set frequency in positive feedback cases, and decrease the correlation strength of feature correlation conditions that contradict non-positive feedback. Adjust the matching priority in the dynamic demand and supply matching rules, increase the priority of the matching category corresponding to positive feedback noodle products, and decrease the priority of the matching category corresponding to non-positive feedback noodle products; The adjusted dynamic demand and supply matching rules are applied to the new demand and supply data samples to obtain the matching results of the adjusted dynamic demand and supply matching rules on the new demand and supply data samples, and user feedback data on the matching results are collected. Compare the user feedback satisfaction with the dynamic demand and supply matching rules before and after the adjustment on the same data sample. If the user feedback satisfaction corresponding to the adjusted dynamic demand and supply matching rules is higher than that corresponding to the original dynamic demand and supply matching rules, then the adjusted dynamic demand and supply matching rules are retained. If the user feedback satisfaction corresponding to the adjusted dynamic demand and supply matching rules is lower than that corresponding to the original dynamic demand and supply matching rules, then the parameters of the dynamic demand and supply matching rules are adjusted in reverse until the user feedback satisfaction corresponding to the dynamic demand and supply matching rules reaches the set level.

9. The personalized recommendation method for smart noodle stalls according to claim 8, characterized in that, The comparison of user satisfaction with the dynamic demand-supply matching rules before and after the adjustment on the same data sample includes: The demand and supply data samples that were applied under the dynamic demand and supply matching rules before the adjustment were selected as the comparison sample set. This comparison sample set contains demand data and corresponding surface supply data of multiple users. The adjusted dynamic demand and supply matching rules are applied to the comparison sample set to generate the adjusted recommendation results for each sample in the comparison sample set. Retrieve the pre-adjustment recommendation results generated on the comparison sample set by the dynamic demand-supply matching rules before the adjustment, as well as the user feedback data collected at that time. Collect user feedback data on the adjusted recommendation results. This feedback data includes the selection rate of recommended noodle products, the degree of acceptance of the recommendation basis, and the score of demand satisfaction. Calculate the user feedback satisfaction corresponding to the dynamic demand-supply matching rule before the adjustment. The user feedback satisfaction calculation is based on the recommended noodle selection rate before the adjustment, the recognition of the recommendation basis before the adjustment, and the demand satisfaction score before the adjustment. The weighted summation method is used, and the weight is set according to the importance of each indicator. Calculate the user feedback satisfaction corresponding to the adjusted dynamic demand and supply matching rule. The user feedback satisfaction is calculated using the same weighted summation method as the user feedback satisfaction corresponding to the original dynamic demand and supply matching rule, based on the adjusted recommended noodle selection rate, the adjusted recognition of the recommendation basis, and the adjusted demand satisfaction score. Compare the user feedback satisfaction with the dynamic demand-supply matching rule before and after the adjustment for each sample in the sample set, and count the number of samples with improved user feedback satisfaction, decreased user feedback satisfaction, and unchanged user feedback satisfaction. The overall improvement in user feedback satisfaction is calculated as the sum of the differences between the user feedback satisfaction of the adjusted dynamic demand supply matching rule and the user feedback satisfaction of the unadjusted dynamic demand supply matching rule across all samples in the comparison sample set, divided by the total number of samples in the comparison sample set. If the overall improvement in user feedback satisfaction exceeds the set improvement threshold, and the proportion of the number of samples with improved user feedback satisfaction to the total number of samples in the comparison sample set exceeds the set proportion threshold, then the adjusted dynamic demand supply matching rule is determined to be superior to the original dynamic demand supply matching rule. If the above conditions are not met, record the demand and supply characteristics corresponding to the samples where user feedback satisfaction has decreased, and analyze the reasons for the inapplicability of the adjusted dynamic demand and supply matching rules in the samples.

10. A smart noodle stall personalized recommendation system, characterized in that, The device includes a processor and a memory, the memory and the processor being connected. The memory is used to store programs, instructions or code, and the processor is used to execute the programs, instructions or code in the memory to implement the smart face card personalized recommendation method according to any one of claims 1-9.

Images