Information processing system, information processing method and program
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- GURUNAVI
- Filing Date
- 2023-06-13
- Publication Date
- 2026-06-19
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to an information processing system, an information processing method, and a program capable of transmitting response information to a question about a menu from a user terminal in a restaurant. [Background technology]
[0002] Conventionally, a system (mobile ordering system) that allows a user to order a menu from a user terminal such as a smartphone without going through a waiter at a restaurant is known. In such a mobile ordering system, a technique for suggesting menus to a user is also known.
[0003] For example, Patent Document 1 listed below describes a method of managing a user's preference information including an order history at a first restaurant in correspondence with the user's identification information, acquiring the user identification information and a store ID indicating a second restaurant in a different chain from the first restaurant from the user terminal, and, based on the preference information corresponding to the user identification information and the menu information of the second restaurant indicated by the store ID, arranging the menu contents included in the menu information in an order corresponding to the preference information using the user's order history corresponding to menu contents of the first restaurant that are the same as or alternative to the menu contents included in the menu information, and transmitting the arranged menu contents to the user terminal for display. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] Patent No. 7065320 Summary of the Invention [Problem to be solved by the invention]
[0005] Incidentally, the desires of a user regarding the menu they wish to order may differ from time to time, for example, the user's mood or hunger level on that day, or the other users (family, friends, etc.) who dine with them at the restaurant, etc. However, the technology described in the above Patent Document 1 suggests menu items using the user's past order history, and therefore cannot suggest menu items that meet the user's desires at any given time.
[0006] In view of the above circumstances, an object of the present invention is to provide an information processing system, an information processing device, an information processing method, and a program that are capable of suggesting a menu that meets the user's needs at any given time when the user orders a menu at a restaurant without going through a waiter. [Means for solving the problem]
[0007] In order to achieve the above-mentioned objective, an information processing system according to one embodiment of the present invention includes a control unit that receives, via a chatbot, question information indicating a question regarding a menu item desired by the user from the menu of the restaurant that the user is using, receives response information indicating a menu item to be suggested in response to the question from a learning model that has previously undergone machine learning to learn information regarding the restaurant's menu, and transmits the response information to the user terminal or the ordering terminal via the chatbot.
[0008] With this configuration, the chatbot can accept questions about the menu requested by the user and return answers using the learning model, making it possible to suggest menu items that meet the user's needs at the time when the user orders a menu at a restaurant without the assistance of a waiter (for example, by mobile order or tabletop order). The learning model can be a generative AI based on a large-scale language model (LLM), such as ChatGPT.
[0009] The control unit may train the learning model so that, when there are multiple menu options that can be suggested in response to the question information, the control unit selects one of the multiple menu options that satisfies certain conditions related to sales history based on POS information of the restaurant, and generates the response information.
[0010] This makes it possible to suggest menu items that meet the user's needs, including menu items with high sales (popular menu items) and menu items with low sales (menus that the restaurant is more interested in selling).
[0011] The control unit may train the learning model so that, when there are multiple menu options that can be suggested in response to the question information, based on inventory information of the ingredients for the restaurant's menu, the control unit selects one of the multiple menu options that uses an ingredient with a large inventory and generates the response information.
[0012] This makes it possible to propose a menu that meets the user's needs and requires less stock of ingredients.
[0013] The control unit may train the learning model so that, when the question information includes information unrelated to the restaurant, the response information is not generated for the unrelated information.
[0014] This can prevent unnecessary interactions between the user and the restaurant that are not related to ordering a menu item.
[0015] The control unit may train the learning model to generate the response information for the related information when the question information includes information unrelated to the restaurant and information related to the restaurant.
[0016] This allows the system to respond to questions that contain information unrelated to restaurants or menus by providing only the relevant information.
[0017] The control unit may train the learning model to generate the response information by combining multiple menus of different categories when the question information includes a request to combine multiple menus within a specified budget.
[0018] This allows a well-balanced combination of menus from multiple different categories to be proposed to the user.
[0019] The control unit may display a chatbot screen including the question information and the response information on the user terminal or the ordering terminal, and may display an order button on the chatbot screen that accepts an order operation for ordering the combined menu items included in the response information all at once.
[0020] This makes it possible to avoid the hassle of accepting orders for a combination of multiple menus proposed to a user for each menu item, thereby improving user convenience.
[0021] The control unit may display a chatbot screen including the question information and the response information on the user terminal or the order terminal, and if the response information includes multiple menus, may display order buttons that accept ordering operations for each of the menus in a position adjacent to each of the multiple menus on the chatbot screen.
[0022] This allows the user, who has confirmed the response to the question, to intuitively place an order for at least one of the multiple suggested menu items.
[0023] An information processing method according to another aspect of the present invention includes: receiving question information indicating a question regarding a menu item requested by the user from among menu items of the restaurant via a chatbot from a user terminal of the user or an ordering terminal installed in a restaurant used by the user; Receive response information indicating a menu to be proposed in response to the question from a learning model that has previously machine-learned information about the restaurant's menu; The response information is transmitted to the user terminal or the order terminal via the chatbot.
[0024] According to yet another aspect of the present invention, there is provided a program for an information processing device, receiving question information indicating a question regarding a menu item desired by the user from among menu items of the restaurant via a chatbot from a user terminal of the user or an ordering terminal installed in the restaurant currently used by the user; receiving response information indicating a menu to be proposed in response to the question from a learning model that has previously machine-learned information about the restaurant's menu; and transmitting the response information to the user terminal or the order terminal via the chatbot. Effect of the Invention
[0025] As described above, according to the present invention, when a user orders a menu at a restaurant without going through a waiter, a menu that meets the user's needs at that time can be suggested. However, this effect does not limit the present invention. [Brief description of the drawings]
[0026] [Figure 1] 1 is a diagram showing a configuration of a restaurant menu suggestion system according to an embodiment of the present invention. [Diagram 2] 1 is a diagram showing a hardware configuration of a restaurant information providing server according to an embodiment of the present invention. [Diagram 3] 2 is a diagram showing a configuration of a database included in a restaurant information providing server according to an embodiment of the present invention. FIG. [Figure 4] 11 is a flowchart showing a flow of a menu suggestion process performed by a restaurant information providing server according to an embodiment of the present invention. [Diagram 5]FIG. 13 is a diagram showing an example of a chatbot screen on which a restaurant information providing server according to one embodiment of the present invention responds to a question from a user terminal about a menu. [Figure 6] FIG. 13 is a diagram showing another example of a chatbot screen on which the restaurant information providing server according to one embodiment of the present invention responds to a question from a user terminal about a menu. [Figure 7] FIG. 13 is a diagram showing another example of a chatbot screen on which the restaurant information providing server according to one embodiment of the present invention responds to a question from a user terminal about a menu. [Figure 8] FIG. 13 is a diagram showing another example of a chatbot screen on which the restaurant information providing server according to one embodiment of the present invention responds to a question from a user terminal about a menu. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0027] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
[0028] [System configuration] FIG. 1 is a diagram showing the configuration of a restaurant menu suggestion system according to this embodiment.
[0029] As shown in the figure, this system includes a restaurant information providing server 100 on the Internet 50, user terminals 200 for users who use tables T of a plurality of restaurants, and restaurant terminals 300 installed in each restaurant.
[0030] The restaurant information providing server 100 is a web server that operates a portal site that lists information about restaurants. The restaurant information providing server 100 is connected to a plurality of user terminals 200 and a plurality of restaurant terminals 300 of restaurants via the Internet 50.
[0031] The restaurant information providing server 100 provides a restaurant information search system for users of user terminals 200 on the portal site. Specifically, the restaurant information providing server 100 searches for restaurant information that matches search conditions based on a search request from the user terminal 200, generates a Web page listing the search results, and transmits it to the user terminal 200. The restaurant information providing server 100 also performs reservation acceptance processing for any restaurant from the user terminal 200 of the user who viewed the restaurant information.
[0032] Furthermore, restaurant information providing server 100 provides a management screen (web page) of restaurant information for restaurants (affiliated restaurants) listed on the portal site. Through the management screen, users of restaurant terminals 300 can edit and update restaurant information on a web page provided to general users as the search results, and upload the web page to the portal site.
[0033] The user terminals 200 (200A, 200B, 200C...) are terminals used by users, such as smartphones, mobile phones, tablet PCs, etc. The user terminals 200 access the restaurant information providing server 100, receive the above-mentioned web page, and display it on a screen using a browser or the like.
[0034] The user terminal 200 determines search conditions for restaurants on the portal site based on a user operation, and transmits a restaurant search request based on the search conditions to the restaurant information providing server 100.
[0035] Meanwhile, restaurant information providing server 100 can receive orders for food and drink at each table of each restaurant from a user using each table T of the restaurant via user terminal 200. When user terminal 200 reads, for example, a two-dimensional barcode C placed as a card, plate, stand, or the like on each table of the restaurant and accesses restaurant information providing server 100 based on the read information (URL), restaurant information providing server 100 displays an order screen for the restaurant's menu on the display unit of user terminal 200.
[0036] Then, when restaurant information providing server 100 accepts an order input from the user via the order screen, it transmits the accepted order information to restaurant terminal 300 of the restaurant. The order information is transferred from restaurant terminal 300 to a terminal (not shown) in the kitchen. Restaurant information providing server 100 also stores the accepted order information, and when it accepts a request for accounting from the user, it executes the accounting in cooperation with a POS terminal (not shown) of the restaurant via restaurant terminal 300.
[0037] In this case, the user terminal 200 also functions as a mobile order terminal for receiving orders from users using each table T of the restaurant and transmitting order information to the restaurant information providing server 100. The user terminal 200 may have installed therein an application (hereinafter also referred to as an MO app) capable of executing processing related to mobile orders in cooperation with the restaurant information providing server 100.
[0038] Restaurant terminal 300 is a terminal installed in each restaurant, and may be a tablet PC, notebook PC, desktop PC, or the like. Restaurant terminal 300 can execute processes related to its own restaurant information, such as editing and updating the restaurant information, based on the operation of an administrator, by communicating with restaurant information providing server 100. Restaurant terminal 300 also mediates communication between restaurant information providing server 100 and a kitchen terminal or a POS terminal in the mobile order process. Of course, restaurant terminal 300 may also function as the kitchen terminal or POS terminal and communicate directly with restaurant information providing server 100.
[0039] In addition, in this embodiment, the restaurant information providing server 100 is connected to a generation AI such as ChatGPT, which is a learning model based on a large-scale language model (LLM), via an API.
[0040] The restaurant information providing server 100 has the generation AI learn the menu information of each restaurant in advance, and while the user is using the table T, accepts a question from the user terminal 200 via the chatbot screen in the form of voice or text about the menu desired by the user, and uses the generation AI to generate response information suggesting a menu in response to the question and transmits it to the user terminal 200 via the chatbot screen. Details of the menu suggestion process via the chatbot screen will be described later.
[0041] [Hardware configuration of restaurant information server] Fig. 2 is a diagram showing a hardware configuration of the restaurant information providing server 100. As shown in the figure, the restaurant information providing server 100 includes a CPU (Central Processing Unit) 11 (server control unit), a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an input / output interface 15, and a bus 14 connecting these to each other.
[0042] The CPU 11 appropriately accesses the RAM 13 etc. as necessary, and performs various arithmetic processing while controlling all the blocks of the restaurant information providing server 100. The ROM 12 is a non-volatile memory in which the OS, programs, various parameters, and other firmware to be executed by the CPU 11 are fixedly stored. The RAM 13 is used as a working area for the CPU 11, and temporarily stores the OS, various applications being executed, and various data being processed.
[0043] The input / output interface 15 is connected to a display unit 16, an operation reception unit 17, a storage unit 18, a communication unit 19, and the like.
[0044] The display unit 16 is a display device using, for example, a Liquid Crystal Display (LCD), an Organic ElectroLuminescence Display (OLED), a Cathode Ray Tube (CRT), or the like.
[0045] The operation reception unit 17 is, for example, a pointing device such as a mouse, a keyboard, a touch panel, or other input device. When the operation reception unit 17 is a touch panel, the touch panel can be integrated with the display unit 16.
[0046] The storage unit 18 is, for example, a non-volatile memory such as a hard disk drive (HDD), a flash memory (SSD; Solid State Drive), or other solid-state memory. The storage unit 18 stores the OS, various applications, and various data.
[0047] Particularly in this embodiment, the storage unit 18 stores data, applications, and other programs for the restaurant information providing server 100 to execute a mobile order process and a menu suggestion process via a chatbot screen, which will be described later. As will be described later, the storage unit 18 has a restaurant information database, a user information database, a learning information database, and a POS (Point Of Sale) information database as databases containing such data.
[0048] The communication unit 19 is, for example, a NIC (Network Interface Card) for Ethernet or various modules for wireless communication such as wireless LAN, and is responsible for communication processing between the user terminal 200 and the restaurant terminal 300.
[0049] [Database configuration of restaurant information server] FIG. 3 is a diagram showing the configuration of a database included in the restaurant information providing server 100. As shown in FIG.
[0050] As shown in the figure, restaurant information providing server 100 has restaurant information database 31, user information database 32, learning information database 33, and POS information database in storage unit 18. These databases are mutually referenced and used as necessary in the learning process of the generation AI and the menu suggestion process by restaurant information providing server 100.
[0051] Restaurant information database 31 stores, for each restaurant, the restaurant's name, location (address or latitude and longitude) information, area information, access information (nearest station information, walking distance information from the nearest station), telephone number, ID (store ID) identifying the restaurant, information on the restaurant's business type and service genre, information introducing the restaurant (information showing the characteristics of the restaurant, such as store PR text, information on events held by the restaurant, etc.), image data related to the restaurant (introducing the restaurant), menu information related to the menu offered by the restaurant, average budget information, business hours, website URL, etc. This information is input from restaurant terminal 300 of each restaurant via a management screen provided by restaurant information providing server 100.
[0052] The menu information corresponds to the menus posted on the website of each restaurant on the portal site, and stores the names of multiple menus that each restaurant can provide for each restaurant. The menu information may be stored for each menu category, such as appetizer / main / meat dish / fish dish / carbohydrate / side dish, lunch / dinner / course, etc. The menu information may also include text information indicating the menu name, price, description, etc., as well as image information such as a photograph of the menu, which is stored in association with the menu. Furthermore, information regarding the calories, amount, and ingredients of each menu may also be stored.
[0053] A menu book ID is assigned to the menu information of each restaurant to identify the menu information for each restaurant. The menu book ID is used in the learning process of the generation AI, and when generating response information to a question from the user terminal 200 via the chatbot screen, the generation AI uses the menu book ID to determine which restaurant's menu information to refer to.
[0054] The genre information may include main categories such as Japanese, Chinese, Italian, French, and grilled meat, as well as more detailed subcategories such as yakitori and tempura in Japanese cuisine, and pasta and pizza in Italian cuisine.
[0055] The user information database 32 stores information about users who own user terminals 200 and are users (members) of the restaurant information service via the portal site provided by the restaurant information providing server 100. Specifically, the user information database 32 stores information such as a user ID, password, name, email address (and other information to be used as a message destination), phone number, address, age (group), sex, and date of birth for each user.
[0056] The learning information database 33 stores training data for the generation AI to learn from. The training data includes (general / individual) knowledge information on products (and brands / names) in specific fields, such as menu information of each restaurant included in the restaurant information database 31 and other restaurant information, and ingredients of dishes and drinks (sake, wine, shochu, etc.) included in the menu information, and is added sequentially as necessary.
[0057] The learning information database 33 also stores setting information for the chatbot characters, information on prohibited questions asked by the chatbot, rule information, and the like.
[0058] As character setting information, training data on a number of different tones of speech, personalities, genders, ages, etc. are stored so that the tone of speech can be changed to match the atmosphere of each restaurant.
[0059] The prohibited items information prohibits questions other than those related to the menu and ordering at each restaurant, so training data is stored to train the generation AI to respond that such questions cannot be answered.
[0060] The rule information includes training data for training the generation AI to reset the conversation on a ticket-by-ticket basis for each table (from when a user starts ordering to when they pay), and training data for training the generation AI to avoid confusing a conversation on the chatbot screen with a customer at one table with a conversation with a customer at another table.
[0061] The POS information database 34 stores POS data based on orders received from the user terminal 200 in each table T. The POS data includes, for example, a table ID, time of entry, number of customers, ordered menu items, number of orders (number of items served), sales amount, order time, etc. The POS information database 34 also stores inventory information for each menu item and its raw materials (information on sold-out menu items, inventory amount of raw materials, etc.).
[0062] The information in the POS information database 34 is also used in the learning process of the generation AI, and when generating response information to a question from the user terminal 200 via the chatbot screen, the generation AI can refer to the POS information (sales status, inventory status, etc.) at that time.
[0063] [System Operation] Next, the operation of the system (mainly the restaurant information providing server 100) configured as above will be described. The operation is executed by the cooperation of hardware such as the CPU 1 of the restaurant information providing server 100 and software stored in the storage unit 18. For convenience, in the following description, the CPU 11 is the subject of operation.
[0064] (Generative AI learning process) First, the learning process of the generation AI will be described.
[0065] The CPU 11 of the restaurant information providing server 100 first makes the generation AI learn the menu information for each of the menu book IDs of each restaurant. Specifically, the CPU 11 registers the menu information of each restaurant in (the database used by) the generation AI via the API, and then transmits information likely to correspond to a question about the menu desired by the user and appropriate information (menu suggestion information) as a response to the generation AI, and repeatedly executes the process of having the generation AI respond.
[0066] In addition, the CPU 11 uses a combination of the above menu information and knowledge information on products and brands (companies, brands, ingredients) stored in the learning information database 33 to train the generation AI by combining various training data so that it can respond to questions such as, for example, "What is a bottle of white wine that goes well with grilled salted saury that costs around 5,000 yen?", "What is a good Japanese sake that is strong-flavored and spicy?", which request a menu that matches a specific dish, a menu that matches taste preferences, a menu that is large in portions (feels filling) / refreshing (low in calories), and other complex questions.
[0067] At this time, the CPU 11 may train the generation AI so that, when there are multiple menu options that can be suggested in response to the question information, the CPU 11 selects from the multiple menu options that satisfies a predetermined condition regarding sales performance (e.g., highest sales / lowest sales) based on the POS information of each restaurant stored in the POS information database 34, and generates response information.
[0068] This makes it possible to suggest menu items that meet the user's needs, including menu items with high sales (popular menu items) and menu items with low sales (menus that the restaurant is more interested in selling).
[0069] In addition, when there are multiple menus that can be suggested in response to the same question information, the CPU 11 may train the learning model to preferentially select from the multiple menus a menu that uses ingredients with a large inventory based on the inventory information of ingredients for the menus of each restaurant stored in the restaurant information database 31, and generate the response information.
[0070] This makes it possible to propose a menu that meets the user's needs and requires less stock of ingredients.
[0071] Furthermore, CPU 11 may train the generation AI so that, when a question from a user includes information unrelated to the restaurant or menu order the user is using, the AI does not generate response information for the unrelated information (so that the AI responds that it cannot answer), based on the prohibited item information stored in the learning information database 33. Furthermore, CPU 11 trains the generation AI to generate response information on a table-by-table or account-by-account basis, for example, by having the AI learn a set of table ID and question information based on the rule information.
[0072] This can prevent unnecessary exchanges between the user and the restaurant that are unrelated to ordering a menu item, and can also prevent conversations with customers at other tables from becoming confused.
[0073] In addition, the CPU 11 makes the generation AI learn training data such as tone of voice, personality, gender, age, etc., which differ for each menu book ID based on the setting information of the character, thereby making it possible to generate response information as if a character that matches the atmosphere of each restaurant is speaking.
[0074] In addition, when the question information from the user includes a request to combine multiple menu items within a given budget, such as "Please suggest an order configuration for four people for 3,000 yen," the CPU 11 may train the generation AI to generate response information by combining multiple menu items from different categories (e.g., appetizers, main courses, desserts, etc.) using the menu category information from the menu information of the restaurant.
[0075] This allows a balanced combination of menus from a plurality of different categories to be proposed to the user, and prevents menus from being disproportionately proposed to the user from the same category.
[0076] Furthermore, the CPU 11 may train the generation AI so that, when there is no menu that can be suggested in response to the question information from the user, the generation AI generates response information indicating a menu that can be substituted for the menu indicated by the question information. Substitutable menus are, for example, menus in the same category or using similar ingredients, menus with similar prices, menus with similar portions, menus with similar calories, menus with similar sales, etc., depending on the content of the question.
[0077] This allows the system to suggest an alternative menu item even if a menu item that meets the user's request is not available, leading to an order.
[0078] The above learning process is executed appropriately each time the menu and product information serving as training data is updated, or each time the above prohibitions and rules are changed.
[0079] (Menu suggestion processing) FIG. 4 is a flowchart showing the flow of the menu suggestion process performed by the restaurant information providing server 100.
[0080] First, the CPU 11 of the restaurant information providing server 100 determines whether or not a request to start using table T (a request to start mobile order) has been received from the user terminal 200 (step 41).
[0081] The request to start using table T includes information about the number of users who will use the table T. When the CPU 11 determines that the request to start using the table T has been accepted (Yes in step 41), the CPU 11 stores the information about the number of users in the POS information database 34 of the storage unit 18, and transmits an order screen for mobile ordering to the user terminal 200 (step 42).
[0082] Next, the CPU 11 determines whether or not the chatbot screen has been accessed from the user terminal 200 via, for example, a predetermined icon or button on the order screen (step 43).
[0083] When it is determined that the chatbot screen has been accessed (Yes in step 43), the CPU 11 causes the chatbot screen to be displayed on the user terminal 200 (step 44).
[0084] Next, the CPU 11 determines whether question information has been accepted by text or voice on the chatbot screen (step 45).
[0085] If it is determined that the question information has been accepted (Yes in step 45), the CPU 11 transmits the question information and the menu book ID to the generation AI, and requests the AI to return reply information (step 46).
[0086] The generation AI refers to the menu information corresponding to the menu book ID, extracts from the menu information a menu that meets the request in the question information, and generates text response information (as well as voice response information).
[0087] Next, the CPU 11 determines whether or not response information has been received from the generating AI (step 47).
[0088] When it is determined that the response information has been received from the generation AI (Yes in step 47), the CPU 11 displays the response information and a menu order button on the chat pod screen (step 48). The response information may also be displayed as a text and a voice output.
[0089] 5, 6, 7, and 8 are diagrams showing examples of chatbot screens in which the restaurant information providing server 100 uses a generation AI to respond to questions from the user terminal 200.
[0090] As shown in these figures, on the chatbot screen, for example, question information 52 is displayed on the left side of the screen, and reply information 53 is displayed on the right side. The question information 52 and reply information 53 are displayed, for example, in a rectangular frame or a speech bubble. The question information 52 is displayed on the chatbot screen, for example, when the user inputs the question information into an input box 51 at the bottom of the screen and performs a send operation such as pressing a send button (not shown), or when a voice input from the user is recognized via a microphone.
[0091] The response information 53 also includes information indicating a menu suggested in response to the question information 52, and an order button 54 for accepting an order operation for the menu is displayed near each suggested menu. This allows the user who has confirmed the response to the question to intuitively order the menu.
[0092] Fig. 5 shows a chatbot screen in a situation where a user wants a wine that matches a specific product within a budget to be suggested. In this screen, three white wines from different regions are suggested as response information 53 to question information 52 requesting a bottle of wine for 5,000 yen that matches grilled salted saury, along with explanations of the taste and the reasons for the suggestion, and corresponding order buttons 54 are displayed near each menu.
[0093] As mentioned above, the compatibility of specific dishes and drinks is learned using training data of general and specific knowledge of products in a specific domain.
[0094] Fig. 6 shows a chatbot screen in a situation where a user expresses taste preferences and has the product selected. In response to question information 52 requesting a strong, spicy sake, three sakes from different regions are suggested as response information 53 along with descriptions of the taste and food that goes well with the sake, and order buttons 54 corresponding to each sake are displayed.
[0095] This output is thought to be the result of not just answering questions about taste preferences, but also learning about which dishes go well with them.
[0096] Fig. 7 shows a chatbot screen in a situation where a user is asked to suggest an ordering combination of in-store menu items within a budget. In response to question information 52 requesting an ordering configuration that will fill four people up for under 3,000 yen, the screen displays multiple menu items for different categories such as pizza, pasta, doria, and side dishes as response information 53, suggests a combination of the menu items, and displays an order button 54 for ordering the menu items in the combination all at once.
[0097] In this case, it is not necessary for all the proposed menu items to belong to different categories, and it is sufficient to select from menu items of as different categories as possible. In other words, a configuration in which the same menu item is selected multiple times, such as spaghetti bolognese and fries in the figure, due to budgetary constraints, is not excluded.
[0098] In response to the request to be "full," the AI selected pizza and pasta, as well as French fries from the side menu instead of salad or cheese.
[0099] FIG. 8 shows a chatbot screen when a question unrelated to the restaurant used by the user or the order on the menu is received.
[0100] As shown in FIG. 5A, for example, in response to a question comparing the menu of the currently used restaurant (fried chicken at Restaurant B) with the menu of another restaurant (fried chicken at Restaurant A), response information 53 is output indicating that a response explaining the menu of the currently used restaurant is given but that a response cannot be given about the menu of other restaurants.
[0101] Furthermore, as shown in FIG. 1B, for questions that are completely unrelated to the restaurant currently being used, reply information 53 is output indicating that no reply can be made.
[0102] Returning to FIG. 4, the CPU 11 then determines whether or not a menu order has been accepted from the user terminal 200 via the order button 54 (step 49).
[0103] If it is determined that the order has been accepted (Yes in step 49), the CPU 11 transmits order information, including a menu ID identifying the menu to be ordered, the order quantity, and a category ID indicating the category of the menu, to the kitchen terminal via (or without) the restaurant terminal 300, and stores the information in the POS information database 34 (step 50).
[0104] Next, the CPU 11 determines whether or not the end of use of table T (a billing request) has been received from the user via a billing button or the like displayed on the order screen or chatbot screen (step 51).
[0105] If it is determined that the end of use has been accepted (Yes in step 51), CPU 11 executes the transaction process in cooperation with the POS terminal via (or without) restaurant terminal 300 (step 52).
[0106] As long as the end of use is not accepted (No in step 51), the CPU 11 returns to step 45 and repeats the subsequent processes.
[0107] [summary] As described above, according to this embodiment, the restaurant information providing server 100 accepts questions about the menu requested by the user via the chatbot screen each time, and returns answers using a generation AI that has learned about the menu information of each restaurant, so that when the user orders a menu at a restaurant via mobile order, a menu that meets the user's needs at that time can be suggested.
[0108] [Variations] The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and scope of the present invention.
[0109] In the above embodiment, the content of the question information 52 input by the user on the chatbot screen is useful customer opinion, so by visualizing the input content, it can be utilized for in-store POS, banners, and campaigns. For example, the restaurant information providing server 100 can analyze the content of the question information 52 input in multiple tables T, and display the menu proposed by the generation AI for the menu (and menu configuration) that was most frequently requested by users as promotional information on the above order page, or deliver a message to the user terminal 200.
[0110] In the above embodiment, an example was shown in which a chatbot character is set in advance for each restaurant and learned by the generation AI. However, the character settings may be changeable by the user on the chatbot screen, for example. In this case, the restaurant information providing server 100 may have the generation AI learn the characteristics of multiple different characters together with the character ID in advance, and the character ID may be sent to the generation AI in response to a user's setting change operation, thereby changing the character settings in the generation AI.
[0111] In the above-described embodiment, the mobile order processing via the user terminal 200 is started by the user terminal 200 reading the two-dimensional barcode C placed on the table T, but the trigger for starting the mobile order is not limited to this, and the trigger for starting the mobile order may be, for example, the MO app successfully achieving a specified communication with a communication device such as an IC tag placed on the table T.
[0112] The processing of the restaurant information providing server 100 in the above-described embodiment may be distributed and executed by a plurality of servers. For example, the processing related to the mobile order and the processing for generating response information (menu suggestion information) to a question from a user may be executed by different servers.
[0113] In the above embodiment, an example is shown in which a cloud-based restaurant information providing server 100 executes restaurant search processing for multiple user terminals 200, but a server having similar functions to the restaurant information providing server 100 and capable of linking with the generation AI may be installed for each restaurant to execute the menu suggestion processing.
[0114] In the above embodiment, an example in which the present invention is applied to a mobile ordering system that allows ordering from a user terminal 200 has been shown, but the present invention is not limited to this as long as the system allows ordering without the intervention of a staff member. The present invention can also be applied to a tabletop ordering system that allows ordering from an ordering terminal installed in each restaurant, i.e., a tabletop terminal installed at each table T, a smart speaker ordering system that allows ordering from a smart speaker (with microphone) installed at each table T, and other systems that allow ordering from a floor robot that patrols the floor of a restaurant. In addition, when ordering from a smart speaker, the chatbot screen is not presented to the user, and the exchange of question information and response information (and ordering processing of the proposed menu) via the chatbot is performed only by voice. In addition, in a system in which ordering is performed from a floor robot, a process of identifying each table by, for example, user input or sensing is required when ordering.
[0115] Among the inventions described in the claims of this application, the invention described as an "information processing method" is one in which each step is automatically performed by at least one device such as a computer through information processing by software, and is not performed by a human using a device such as a computer. In other words, the "information processing method" is an information processing method by computer software, and is not a method in which a human operates a calculation tool called a computer. [Explanation of symbols]
[0116] 11...CPU 18...Storage section 19…Communications Department 31…Restaurant information database 32...User information database 33…Educational information database 34…POS information database 52...Question information 53...Response information 54…Order button 100: Restaurant information server 200...User terminal 300…Restaurant terminal C…Code (2D barcode)
Claims
1. A control unit receives question information from the user's terminal or an ordering terminal installed in the restaurant the user is using, indicating a question about a menu item requested by the user from the restaurant's menu. It receives response information indicating a menu item to be suggested in response to the question from a learning model that has been pre-trained with information about the restaurant's menu, and transmits the response information to the user's terminal or the ordering terminal. An information processing system equipped with the following features.
2. The information processing system according to claim 1, The control unit, in response to the question information, trains the learning model to select a menu item that satisfies predetermined conditions regarding sales performance from among the multiple menu items, based on the restaurant's POS information, if there are multiple menu items that can be suggested, and generates the response information. Information processing system.
3. The information processing system according to claim 1, The control unit, in response to the question information, trains the learning model to generate the response information by selecting the menu item that uses the most readily available raw material from among the multiple menu items, based on the raw material inventory information of the restaurant's menu items, when there are multiple menu items that can be suggested. Information processing system.
4. The information processing system according to claim 1, The control unit instructs the learning model not to generate response information for irrelevant information if the question information includes information unrelated to the restaurant. Information processing system.
5. The information processing system according to claim 4, The control unit causes the learning model to generate the response information for the relevant information if the question information includes both information unrelated to the restaurant and information related to the restaurant. Information processing system.
6. The information processing system according to claim 1, The control unit, when the question information includes a request to combine multiple menus within a predetermined budget, trains the learning model to generate the response information by combining multiple menus from different categories. Information processing system.
7. The information processing system according to claim 6, The control unit causes the user terminal or the ordering terminal to display a chatbot screen containing the question information and the response information, and displays an order button on the chatbot screen to accept an order operation for ordering the combined menu items included in the response information. Information processing system.
8. The information processing system according to claim 1, The control unit causes the user terminal or ordering terminal to display a chatbot screen containing the question information and the response information, and if the response information includes multiple menus, it causes the user terminal to display an order button for each of the multiple menus in the vicinity of each of the menus on the chatbot screen to accept the order operation for each menu. Information processing system.
9. The system receives question information from the user's user terminal or an ordering terminal installed in the restaurant the user is visiting, indicating a question about the menu item the user requests from the restaurant's menu. The response information indicating the menu to be proposed in response to the aforementioned question is received from a machine learning model that has been pre-trained with information about the restaurant's menu. The response information is sent to the user terminal or the ordering terminal. Information processing methods.
10. In an information processing device, The steps include receiving question information from the user's user terminal or an ordering terminal installed in the restaurant the user is using, indicating a question about the menu item the user requests from the restaurant's menu, The steps include receiving response information indicating a menu to be proposed in response to the aforementioned question from a machine learning model that has been pre-trained with information about the restaurant's menu, The steps include: sending the aforementioned response information to the user terminal or the order terminal; A program that executes the command.