Intelligent dish arrangement method, system and equipment based on large language model and storage medium

Abstract

The invention provides an intelligent dish arrangement method based on a large language model. The method comprises the steps that dining demand information input by a user is acquired based on terminal equipment; wherein the dining demand information at least comprises user basic information, historical ordering records, taste preferences, budget and banquet types; performing preliminary screening according to the dining demand information, and calling a dish database based on a corresponding food material label to perform dish screening; selecting a recommendation cue word template according to the type of the banquet, and calling a large language model according to the dish label to carry out multiple rounds of dialogue and semantic analysis so as to generate a recommendation menu meeting the requirements of the user; and performing optimization adjustment on the feedback information of the recommended menu based on the user to generate a final menu and output a dish arrangement result. The menu can be intelligently generated according to the multi-dimensional requirements (taste, health, budget and the like) of the user, and the user experience is greatly improved.

Description

Technical Field

[0001] This application belongs to the field of computer technology, and in particular relates to an intelligent menu arrangement method, system, device and storage medium based on a large language model. Background Technology

[0002] Existing ordering and banquet menu design largely relies on human experience. When ordering food or planning a banquet, users often need to spend a lot of time looking up menu information, matching dishes, calculating nutrition and budget. Traditional menu design methods cannot make intelligent recommendations based on users' taste preferences, health needs, budget constraints, and occasion characteristics, resulting in low ordering efficiency, poor user experience, and banquet menu design that easily overlooks elements such as nutritional balance or variety of dishes.

[0003] The traditional method of arranging dishes has the following defects and shortcomings: (1) Lack of personalization: It is unable to provide personalized intelligent recommendations based on users' historical ordering data, health status, taste preferences, etc.

[0004] (2) Inefficient: Manually preparing dishes is time-consuming and prone to errors, especially in complex settings such as banquets.

[0005] (3) Difficulty in balancing nutrition and budget: Existing systems have difficulty balancing nutritional balance and budget constraints, and lack automatic optimization capabilities.

[0006] (4) Lack of intelligent interaction: The recommendation results cannot be dynamically adjusted based on real-time user feedback, and the intelligent interactive experience is lacking. Summary of the Invention

[0007] In view of this, this application aims to propose an intelligent menu arrangement method, system, device and storage medium based on a large language model to solve at least one of the above problems.

[0008] To achieve the above objectives, the technical solution of this application is implemented as follows: Firstly, this application provides an intelligent menu-arranging method based on a large language model, including: The dining needs information input by the user is obtained based on the terminal device; wherein, the dining needs information includes at least the user's basic information, historical ordering records, taste preferences, budget and banquet type; Based on the dining needs information, a preliminary screening is performed, and then the menu is further filtered by calling the menu database based on the corresponding ingredient tags; Based on the type of banquet, a recommended prompt template is selected, and based on the dish tags, a large language model is invoked to conduct multi-turn dialogue and semantic analysis in order to generate a recommended menu that meets the user's needs. The recommended menu is optimized and adjusted based on user feedback to generate the final menu and output the dish arrangement result.

[0009] Secondly, based on the same inventive concept, this application also provides an intelligent menu-setting system based on a large language model, comprising: The information acquisition module is configured to acquire dining demand information input by the user based on the terminal device; wherein, the dining demand information includes at least the user's basic information, historical ordering records, taste preferences, budget and banquet type; The dish selection module is configured to perform preliminary selection based on the dining needs information, and to call the dish database to select dishes based on the corresponding ingredient tags; The menu generation module is configured to select a recommended prompt template based on the banquet type, and call a large language model to perform multi-turn dialogue and semantic analysis based on the dish tags in order to generate a recommended menu that meets the user's needs. The menu arrangement result output module is configured to optimize and adjust the recommended menu based on user feedback to generate the final menu and output the menu arrangement result.

[0010] Thirdly, based on the same inventive concept, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in the first aspect.

[0011] Fourthly, based on the same inventive concept, this application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions for causing the computer to perform the method as described in the first aspect.

[0012] Compared with existing technologies, the intelligent menu scheduling method, system, device, and storage medium based on a large language model described in this application have the following advantages: The intelligent menu-setting method based on a large language model described in this application can automatically analyze user needs based on multi-dimensional user requirements, intelligently generate personalized ordering suggestions and banquet menus, and greatly improve ordering efficiency and banquet planning quality. Attached Figure Description

[0013] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of an intelligent menu-setting method based on a large language model, as described in an embodiment of this application. Figure 2 This is a schematic diagram of a scenario-based solution for intelligent food sorting as described in an embodiment of this application; Figure 3This is a schematic diagram of the recommended ratio rules described in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of an intelligent menu-setting system based on a large language model, as described in an embodiment of this application. Figure 5 This is a schematic diagram of the hardware structure of the electronic device described in the embodiments of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0015] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0016] The embodiments of this application are described in detail below with reference to the accompanying drawings.

[0017] Please see Figure 1 As shown, this embodiment provides an intelligent menu-setting method based on a large language model, which specifically includes the following steps: Step S101: Obtain the dining needs information input by the user based on the terminal device; wherein, the dining needs information includes at least the user's basic information, historical ordering records, taste preferences, budget and banquet type.

[0018] Specifically, in this embodiment, the following dining needs information of the user is collected based on the ordering terminal provided by the restaurant, including: Obtain basic information about users, such as age, gender, and health status (e.g., whether they have diabetes, allergies, etc.); The system records users' past order history and extracts information such as user preferences, frequently ordered dishes, and allergies from it. The system can obtain users' taste preferences, such as spicy, mild, sweet and sour, etc. This information can be obtained from the user's historical order records or by allowing the user to input their taste preferences on the ordering terminal.

[0019] Users can select the occasion of the banquet (business dinner, family gathering, festival activities, etc.) and their budget range to help the system consider the fit between budget and occasion when making recommendations.

[0020] Step S102: Perform preliminary screening based on dining needs information, and use the corresponding ingredient tags to call the dish database for dish selection.

[0021] Specifically, in this embodiment, the menu database is the core data source of the entire system, containing rich menu information and corresponding tags. These tags and data help the recommendation engine filter and recommend dishes based on user needs. This includes: The menu information includes: flavor labels, such as spicy, mild, sweet and sour, to help the system recommend suitable dishes based on the user's taste preferences; nutritional information, including calories, fat, protein, and carbohydrates, primarily for users with health needs (such as low-fat, low-salt, etc.); occasion-appropriate labels, such as family gatherings, business banquets, and festive parties, to help recommend different types of dishes for different occasions; and allergen labels, such as seafood, nuts, and dairy products, to avoid recommending related dishes to users with a history of allergies.

[0022] The menu is categorized into traditional and innovative dishes. The system will recommend traditional home-style dishes or innovative specialty dishes based on the needs of different users. Seasonal and signature dishes are also included. To ensure the timeliness and distinctiveness of the menu, the system will prioritize seasonal and restaurant signature dishes.

[0023] It should be noted that, as the market changes and seasons change, the recipes, prices, nutritional information, and other information in the database will be updated regularly; at the same time, the system will avoid recommending dishes that have been discontinued by obtaining real-time inventory status.

[0024] Step S103: Select a recommended prompt template based on the banquet type, and call a large language model to perform multi-turn dialogue and semantic analysis based on the dish tags to generate a recommended menu that meets the user's needs.

[0025] Specifically, in this step, this embodiment uses an AI recommendation engine. The AI ​​recommendation engine combines a large language model with rule algorithms to conduct in-depth analysis of user needs and generate personalized menu recommendations based on the user's input.

[0026] The system analyzes user input using Natural Language Processing (NLP) to extract key information (such as number of people, budget, taste preferences, health restrictions, etc.). Based on the user's historical data, taste preferences, health status, and banquet type, the system combines a large language model (which optimizes the accuracy of recommendations based on a rich corpus and contextual information) to generate a personalized menu.

[0027] Furthermore, Figure 2 The demonstration showcased a scenario-based solution for intelligent menu planning, involving the assembly of prompts, the invocation of a large AI model, and the generation of menus tailored to customer needs through multi-turn AI dialogue. Specifically, this included: Choose different recommended prompt templates based on the type of banquet. For example, if it is a business banquet, more emphasis is placed on the sense of ceremony, so choose the prompt template for business banquets. If it is a family gathering, more emphasis is placed on the quantity, so choose the prompt template for family gatherings. If it is a specific holiday, choose the prompt template for that specific holiday.

[0028] Based on the dish labels, a large-scale model is used to exclude dishes with dietary restrictions, such as those that prohibit seafood consumption. Dishes with seafood labels or names containing seafood references are excluded. Additionally, dishes detrimental to customer health are excluded; for example, if a customer has high uric acid levels, high-purine foods are excluded using the large-scale model. (Filtering is done using a large-scale model based on labels and dish names.) By analyzing customer semantics through natural language processing, we can directly help customers select their must-order dishes. If a customer doesn't eat a staple food, we can remove that dish from their menu. We can also create precise recommendations based on customer demographics, such as seniors, children, or women.

[0029] It should be noted that when using the large model to recommend seasonal dishes, signature dishes, and special dishes set by the restaurant, these should all be prioritized while meeting customer needs.

[0030] Finally, the final menu is obtained through two rounds of large-scale model dialogue, along with two rounds of large-scale model dish type recommendations and price percentage settings. In this embodiment, two rounds of model dialogue are used because dishes are generally divided into appetizers and main courses. These two core structures are not closely related, and the price range is relatively fixed. If a single-round recommendation mode were used, the accuracy, final effect, and speed of menu arrangement would all be inferior to the results obtained through two separate dialogues.

[0031] Step S104: Optimize and adjust the recommended menu based on user feedback to generate the final menu and output the dish arrangement result.

[0032] Specifically, in this step, users can rate the taste of each dish. The system adjusts future dish recommendations based on user feedback to ensure they better suit the user's taste. If a user reports that certain dishes have adverse effects on their health, the system will avoid recommending similar dishes in future updates. Furthermore, users can adjust the menu according to their needs, such as adding or removing dishes or modifying taste preferences. The system optimizes the menu and updates the recommendations in real time.

[0033] After multiple rounds of rule filtering and optimization, the final list of recommended dishes is generated, and detailed menu information is output.

[0034] The intelligent menu-setting method based on a large language model described in this embodiment can automatically analyze user needs based on multi-dimensional user requirements, intelligently generate personalized ordering suggestions and banquet menus, and greatly improve ordering efficiency and banquet planning quality.

[0035] This embodiment provides the following menu design logic (generate the menu by following these steps): Step A1: Define the core framework (seasonal + unique features + wellness + customer needs), including: Seasonal: Select dishes tagged "seasonal" from the database that are in season (accounting for ≥30%), such as "yellow croaker", "bamboo shoots", and "moringa seedlings" in May; light staple food (small portion fried rice) for lunch, and heavier staple food (claypot rice) for dinner.

[0036] Features: Select dishes tagged "signature dishes / traditional dishes" (≥20%), recommend them to new customers (such as "sand garlic braised bean noodles" and "caviar crystal yellow croaker jelly"), and pair them with new / customized dishes for old customers (such as newly developed seasonal dishes).

[0037] Health preservation: Match health preservation needs according to the season (select mushrooms / goji berries to nourish the liver in spring; select chicken head rice / sour soup to strengthen the spleen in summer; select lily bulbs / white fungus to nourish the lungs in autumn; select sea cucumber / mutton to nourish the kidneys in winter), and include 1-2 "high-end products" (fish maw / bird's nest / sea cucumber).

[0038] Customer needs: For new customers, choose dishes that are reasonably priced, have a clear theme (such as "Fresh Seasonal Foods in May"), and offer good value for money; for returning customers, highlight the specialties (new products / limited-edition ingredients) and retain their favorite dishes; prioritize dishes with dietary restrictions / preferences (such as Muslims avoiding pork, and recommending "Spicy Braised Scallops and Sea Cucumbers" for those who like spicy food).

[0039] Step A2: Combine dishes according to the matching principles, including: Flavor and color: Must choose three colors of dishes: white (such as "Caviar Crystal Yellow Croaker Jelly"), green (such as "Spicy Moringa Sprouts"), and red (such as "Salt and Pepper Mini Sun Tower"), taking into account sour / sweet / salty / umami.

[0040] Main and secondary dishes should be arranged in a balanced way, with high and low prices (e.g., side dishes such as "fresh croaker maw, morel mushroom, and sour soup with razor clams" + main dishes such as "yellow croaker" + regular dishes such as "scallion and bean noodles") to avoid price concentration.

[0041] A balanced diet of meat and vegetables: Seafood: Meat: Vegetables ≈ 4:3:3 (e.g., choose yellow croaker / clams for seafood, sea cucumber / braised pork for meat, and bamboo shoots / moringa seedlings for vegetables).

[0042] Nutritional balance: The combination of nourishing (fish maw / sea ​​cucumber) and cooling (seasonal vegetables) is in line with seasonal health practices.

[0043] Utensil pairings: High-end dishes use exquisite utensils (such as caviar served with a silver spoon), while ceremonial dishes (such as "crab fried rice" served in the dining room) use special tools.

[0044] Reasonable gross profit: Based on the gross profit data of dishes in the database, ensure that the overall gross profit meets the store's standards.

[0045] Step A3, quantity matching, including: For 2-4 people: 4 appetizers, with a reasonable combination of meat and vegetables, flavors, and colors; soup served individually; 4 main courses (small portions): 1 fish, 1 small seafood dish, 1 meat dish, and 1 vegetable dish; 1 staple food; 1 dessert. For 6-10 people: 6 appetizers, with a reasonable combination of meat and vegetables, flavors and colors; soup in a simmering pot; 6 main courses, including 1 fish, 2 small seafood dishes, 1 meat dish, and 2 vegetable dishes; 1 staple food; and 1 dessert.

[0046] For groups of 10 or more: 6-8 appetizers, with a reasonable combination of meat and vegetables, flavors and colors; 1-1.5 soups; 6-8 main courses, 1.5 servings, including 1-2 fish, 2-3 small seafood dishes, 1-2 meat dishes, and 2-3 vegetable dishes; 1 staple food; and 1 dessert.

[0047] For small groups (2-4 people): We recommend individual dishes (such as soup / vegetables) with a variety of options (6-8 dishes in total).

[0048] For larger groups (12 or more people): Increase the portion sizes of dishes (e.g., choose "portion" instead of "per serving" for main courses) to avoid an excessive number of dishes (total dishes ≤ 15).

[0049] For all dishes where the unit is "person", the quantity Item Num must be less than or equal to the number of diners. The default quantity is the number of diners.

[0050] The execution logic is as follows: Step B1: Analyze the needs and extract the number of people, scenarios, budget, new and old customers, preferences, tastes, dietary restrictions, and special needs (such as a sense of ritual). Step B2: Select dishes from the database that match the tags "seasonal", "specialty", and "health-preserving", and exclude those with dietary restrictions (such as spicy dishes). Step B3: Combine and match ingredients according to the order of "appetizer - soup - main course - vegetarian dish - staple food - dessert," matching them according to the rules of flavor / color / meat / vegetarian balance, such as... Figure 3 As shown; Step B4: Adjust the budget to ensure the total price is within the budget. Step B5: Ensure the uniqueness of dishes. No duplicate dish names should appear in the generated menu. Each dish can only appear once to avoid recommending the same dish repeatedly (e.g., "Caviar Chicken with Mushrooms and Crispy White Mushroom Fish" can only appear once). Step B 6: Optimize highlights and add dishes with a sense of ritual (such as dine-in fried rice), specialty dishes (such as braised bean noodles with garlic), and health-preserving dishes (such as chicken head rice dessert); Step B 7: Output the final selected dishes.

[0051] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0052] Based on the same inventive concept, and corresponding to any of the above embodiments, the embodiments of this application also provide an intelligent menu-setting system based on a large language model.

[0053] like Figure 4 As shown, the intelligent menu scheduling system based on a large language model includes: The information acquisition module 11 is configured to acquire dining demand information input by the user based on the terminal device; wherein, the dining demand information includes at least the user's basic information, historical ordering records, taste preferences, budget and banquet type; The dish selection module 12 is configured to perform preliminary selection based on dining needs information and to call the dish database based on the corresponding ingredient tags to select dishes. The menu generation module 13 is configured to select a recommended prompt template based on the banquet type, and call a large language model to perform multi-turn dialogue and semantic analysis based on the dish tags in order to generate a recommended menu that meets the user's needs. The menu arrangement result output module 14 is configured to optimize and adjust the recommended menu based on user feedback to generate the final menu and output the menu arrangement result.

[0054] For ease of description, the above system is described by dividing it into various modules based on their functions. Of course, in implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware.

[0055] The system described in the above embodiments is used to implement the corresponding method in any of the foregoing embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0056] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the methods described in any of the above embodiments.

[0057] Figure 5 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0058] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0059] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0060] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0061] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0062] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0063] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0064] The electronic devices described above are used to implement the corresponding methods in any of the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0065] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to perform the methods described in any of the above embodiments.

[0066] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0067] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to perform the methods described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0068] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0069] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0070] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A smart menu-setting method based on a large language model, characterized in that, include: The dining needs information input by the user is obtained based on the terminal device; wherein, the dining needs information includes at least the user's basic information, historical ordering records, taste preferences, budget and banquet type; Based on the dining needs information, a preliminary screening is performed, and then the menu is further filtered by calling the menu database based on the corresponding ingredient tags; Based on the type of banquet, a recommended prompt template is selected, and based on the dish tags, a large language model is invoked to conduct multi-turn dialogue and semantic analysis in order to generate a recommended menu that meets the user's needs. The recommended menu is optimized and adjusted based on user feedback to generate the final menu and output the dish arrangement result.

2. The method according to claim 1, characterized in that, The selection of dishes includes: Based on the ingredient labels of the dishes and combined with the user's health information, dishes that do not meet the requirements are automatically removed. Based on the user's historical ordering records and taste preferences, the system prioritizes recommending dishes from the user's most recent orders. It also prioritizes recommending dishes that fit the user's budget based on the number of people in the group and the set budget range, and adjusts the recommendations based on the price and portion size of the dishes.

3. The method according to claim 2, characterized in that, Also includes: Based on seasonal changes and the restaurant's signature dishes, we prioritize recommending seasonal ingredients and specialty dishes.

4. The method according to claim 1, characterized in that: When generating the recommended menu, the menu structure is optimized according to the banquet type and nutritional needs, dishes are combined according to the matching principle, and the gross profit data of dishes in the database is combined to match the rules of taste / color / meat and vegetable / nutritional balance.

5. The method according to claim 1, characterized in that: The system obtains user ratings for each dish and provides feedback, adjusting recommended dishes based on user feedback.

6. The method according to claim 1, characterized in that: After multiple rounds of rule filtering and optimization, a final list of recommended dishes is generated, and the dish ranking results are output. The dish ranking results include the dish name and quantity, the total price of the menu, the nutritional analysis results, and the target audience for each dish.

7. An intelligent menu scheduling system based on a large language model, characterized in that, include: The information acquisition module is configured to acquire dining demand information input by the user based on the terminal device; wherein, the dining demand information includes at least the user's basic information, historical ordering records, taste preferences, budget and banquet type; The dish selection module is configured to perform preliminary selection based on the dining needs information, and to call the dish database to select dishes based on the corresponding ingredient tags; The menu generation module is configured to select a recommended prompt template based on the banquet type, and call a large language model to perform multi-turn dialogue and semantic analysis based on the dish tags in order to generate a recommended menu that meets the user's needs. The menu arrangement result output module is configured to optimize and adjust the recommended menu based on user feedback to generate the final menu and output the menu arrangement result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that, in, The non-transitory computer-readable storage medium stores computer instructions for causing a computer to perform the method described in any one of claims 1-6.

Images