system

Abstract

Provide a system. [Solution] A system including: a means for a user to input the type of beverage they wish to drink, their preferred flavor, budget, and meal details; a means for transmitting the user's input to a server; a means for the server to format the user's input and perform any necessary preprocessing; a means for the server to search a database for beverage candidates that meet the criteria based on the formatted data; a means for the server to evaluate the candidates using a generation AI and select the most suitable beverage; a means for the server to transmit information about the selected beverage to the user's terminal; a means for the terminal to display information about the beverage to the user; a means for the user to provide feedback; a means for transmitting the feedback to a server; and a means for the server to store the feedback in a database.

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Describe the "problem that the invention aims to solve" and the "means for solving the problem."

[0005] Traditionally, beverage selection for customers has often been performed by top sommeliers, which requires specialized knowledge and experience. However, not all restaurants or consumers have access to top sommeliers, making it difficult for ordinary employees and consumers to select the appropriate beverage. Furthermore, it has been difficult to provide individualized support that precisely addresses the user's preferences, budget, and meal contents. The present invention aims to solve these problems and provide a system that allows ordinary employees and consumers to easily select beverages. [Means for solving the problem]

[0006] The present invention provides a system including a means for a user to input the type of beverage they wish to drink, their preferred flavor, their budget, and their meal details, a means for transmitting the user's input to a server, a means for the server to format the user's input and perform any necessary preprocessing, a means for the server to search a database for beverage candidates that meet the formatted data, a means for the server to evaluate the candidates using a generation AI and select the optimal beverage, a means for the server to transmit information about the selected beverage to the user's terminal, a means for the terminal to display the beverage information to the user, a means for the user to provide feedback, a means for transmitting the feedback to the server, and a means for the server to store the feedback in a database, thereby enabling even ordinary employees and consumers to easily select an appropriate beverage.

[0007] "User" refers to consumers and general employees who use this system.

[0008] "Terminal" refers to a computer device such as a smartphone, tablet, or PC, which allows a user to access the System and input or receive information.

[0009] "Server" refers to the central processing unit that receives and processes data sent by users and performs the necessary database searches and evaluations by the generating AI.

[0010] "Database" refers to a collection of accumulated information that stores information about beverages and word-of-mouth data and is searchable by the server.

[0011] "Generative AI" refers to an artificial intelligence engine that uses machine learning algorithms to evaluate and select the beverage that best suits the user's requirements based on information in a database.

[0012] "Input" refers to the act of the user providing information to the terminal, such as the type of beverage they want to drink, their preferred flavor, their budget, and what they will eat.

[0013] "Formatting" refers to the process in which the server performs format checks, string normalization, category assignment, etc. on the data received from the user to convert it into a format suitable for searching and evaluation.

[0014] "Search" refers to the process in which the server extracts candidate beverages that match the criteria from the database based on the formatted data.

[0015] "Selection" refers to the process in which the generative AI evaluates and determines the beverage that best suits the user's criteria from the candidates obtained through the search.

[0016] "Feedback" refers to opinions such as ratings and comments provided by users after experiencing a beverage.

[0017] "Storage" refers to the process by which the server records the feedback received from users in a database and uses it to improve the system and the accuracy of recommendations in the future.

[0018] By defining it in this way, the role of each element and process in the system becomes clear. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] System Overview

[0041] This system recommends the perfect drink based on the type of drink the user wants, their preferred flavor, budget, and meal content. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers. The system accepts user input, searches a database based on that input, and uses generative AI to select the best drink candidates and provide information.

[0042] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] 1. Accepting user input

[0044] User: Accesses the system and inputs the type of beverage they want to drink (e.g., wine, beer, sake, etc.), their preferred flavor (e.g., fruity, sweet, dry, etc.), their budget (e.g., under 3,000 yen), and their meal plan (e.g., pasta, sushi, etc.). This input is done via a device such as a smartphone, tablet, or PC.

[0045] Example user input:

[0046] Type of drink: Wine

[0047] Favorite flavor: Fruity and sweet

[0048] Budget: Under 3,000 yen

[0049] Meal: Pasta

[0050] 2. Data transmission and formatting

[0051] Terminal: Sends input data from the user to the server. The data is converted into an appropriate format (e.g., JSON) and sent to the server.

[0052] Server: Formats the received data, performs any necessary format checks, normalizes strings, assigns categories, and converts it into a form suitable for searching and evaluation.

[0053] 3. Database Search

[0054] Server: Based on the formatted data, the server searches the database (stored information about beverages and reviews) for beverage candidates that match the criteria. Methods such as SQL queries are used for the search process.

[0055] SQL query example:

[0056] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[0057] 4. Candidate evaluation and selection

[0058] Server: The search results are passed to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[0059] AI evaluation process example:

[0060] Candidates evaluated by AI:

[0061] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0062] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[0063] 5. Providing information about alcohol

[0064] Server: Sends information about the drink that is determined to be the most suitable (such as name, rating, price, and detailed description) to the user's device.

[0065] Examples of information to send:

[0066] {

[0067] "name": "Japanese fruity wine",

[0068] "rating": 4.5,

[0069] "price": 2800,

[0070] "description": "A very fruity and sweet wine that pairs well with pasta."

[0071] }

[0072] Terminal: Displays the received information to the user, allowing the user to check the details of the recommended beverage.

[0073] 6. User Feedback

[0074] User: After experiencing the purchased beverage, provide a rating and comment. Enter your feedback in a dedicated form.

[0075] Feedback input example:

[0076] Rating: 4 (out of 5)

[0077] Comment: "Very satisfied. Just as described, perfect with pasta."

[0078] Terminal: Sends user feedback to the server.

[0079] Server: The received feedback is stored in a database and used to improve the accuracy of recommendations next time.

[0080] Specific examples

[0081] For example, if a user wants to drink a "fruity and sweet wine" with "pasta" for dinner for "less than 3,000 yen," the system operates as follows.

[0082] 1. User: Accesses the system app on a smartphone, selects a wine, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta."

[0083] 2. Terminal: Converts the input data into JSON format or similar and sends it to the server.

[0084] 3. Server: Formats the received data and performs any necessary preprocessing.

[0085] 4. Server: Searches the database for wines that meet the above criteria.

[0086] 5. Server: Passes the search results to the generation AI, which evaluates and selects the most suitable wine.

[0087] 6. Server: Sends information about the selected wine to the user's device so that the user can check the information.

[0088] 7. User: Purchases the recommended wine, drinks it, and then provides feedback with a rating and comments.

[0089] 8. Terminal: Sends the provided feedback to the server.

[0090] 9. Server: The feedback is stored in a database and used to improve future recommendations.

[0091] This allows users to easily select the beverage that best suits their tastes and also allows them to easily rate and provide feedback.

[0092] The processing flow will be explained below.

[0093] Step 1: Accepting User Input

[0094] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and meal details. This can be done using a device such as a smartphone, tablet, or PC.

[0095] example:

[0096] Type of drink: Wine

[0097] Favorite flavor: Fruity and sweet

[0098] Budget: Under 3,000 yen

[0099] Meal: Pasta

[0100] Step 2: Send data

[0101] Terminal: Converts user input data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input for errors.

[0102] Example of transmitted data:

[0103] {

[0104] "type": "wine",

[0105] "preferences": "Fruity and sweet",

[0106] "budget": 3000,

[0107] "meal": "pasta"

[0108] }

[0109] Step 3: Data Shaping

[0110] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment).

[0111] Formatting example:

[0112] Check the format

[0113] String normalization

[0114] Category assignment

[0115] Step 4: Database Search

[0116] Server: Queries the database based on the formatted data to find beverage candidates that match the criteria. SQL is used for query processing.

[0117] SQL query example:

[0118] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[0119] Step 5: Evaluate and select candidates

[0120] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best meets the user's criteria. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[0121] Example of evaluation process:

[0122] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0123] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[0124] Step 6: Submit your information

[0125] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[0126] Examples of information to send:

[0127] {

[0128] "name": "Japanese fruity wine",

[0129] "rating": 4.5,

[0130] "price": 2800,

[0131] "description": "A very fruity and sweet wine that pairs well with pasta."

[0132] }

[0133] Step 7: Viewing information

[0134] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[0135] Display example:

[0136] Drink Name: Japanese Fruity Wine

[0137] Rating: 4.5

[0138] Price: 2800 yen

[0139] Description: A very fruity and sweet wine that pairs well with pasta.

[0140] Step 8: Provide feedback

[0141] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[0142] Feedback input example:

[0143] Rating: 4 (out of 5)

[0144] Comment: "Very satisfied. Just as described, perfect with pasta."

[0145] Step 9: Submit your feedback

[0146] Terminal: Sends the feedback information entered by the user to the server.

[0147] Example of submitted feedback:

[0148] {

[0149] "rating": 4,

[0150] "comment": "Very satisfied. Just as described, it was a perfect match for pasta."

[0151] }

[0152] Step 10: Save your feedback

[0153] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[0154] Save example:

[0155] Feedback data stored in a database

[0156] Use it for your next recommendation

[0157] Through these steps, users can easily choose the best beverage that suits their tastes and rate and provide feedback on their experience.

[0158] Example 1

[0159] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0160] Conventional beverage recommendation systems have difficulty in selecting the optimal beverage for a user's individual preferences, budget, and meal content, and it has been particularly difficult to achieve highly accurate recommendations comparable to those made by a top sommelier. Furthermore, they lacked a mechanism for utilizing user feedback in future recommendations. This resulted in a high risk of users choosing a beverage that did not suit their preferences, often resulting in dissatisfaction with the beverage selection.

[0161] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0162] In this invention, the server includes a means for the user to input the type of beverage they want to drink, their preferred flavor, their budget, and their meal details; a means for transmitting the user's input to the information processing device; and a means for the information processing device to format the user's input and perform any necessary preprocessing. This enables the information processing device to select the optimal beverage based on the user's input data using a highly accurate database search and a generative AI model. Furthermore, by utilizing user feedback to improve the accuracy of the next recommendation, it is possible to continuously provide highly satisfying selection results.

[0163] A "user" is a person who accesses the system to input the type of beverage they wish to drink, their preferred taste, budget, and dietary needs to receive recommendations.

[0164] The "information processing device" is a device that receives user input data, formats it and performs necessary preprocessing, and then evaluates and selects candidate beverages using a generative AI model.

[0165] The "storage device" is a device that stores information about beverages, word-of-mouth data, past evaluation data, etc., and stores a database that the information processing device uses for searches and evaluations.

[0166] A "generative AI model" is an artificial intelligence model that evaluates potential beverages based on past word-of-mouth and rating data, and selects the beverage that best suits the user's requirements.

[0167] A "terminal device" is a device that transmits user input to an information processing device and displays information about the selected drink to the user. Examples of such devices include smartphones, tablets, and personal computers.

[0168] "Feedback" refers to the ratings and comments provided by users after trying a recommended beverage, and is information used to improve the accuracy of recommendations next time.

[0169] The following describes in detail an embodiment of the present invention, specifically showing what hardware and software are used to process and calculate data.

[0170] 1. Accepting user input

[0171] User:

[0172] First, users access the system using a smartphone, tablet, or PC. Through the system's interface (web application or native application), they input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs. This allows users to easily select a beverage that suits their preferences.

[0173] Example of input contents:

[0174] Type of drink: Wine

[0175] Favorite flavor: Fruity and sweet

[0176] Budget: Under 3,000 yen

[0177] Meal: Pasta

[0178] 2. Data transmission and formatting

[0179] Device:

[0180] The data entered by the user is first converted into an appropriate format (e.g., JSON) and then sent to the information processing device, so that the input data is delivered to the server in a unified format.

[0181] server:

[0182] The information processing device (server) formats the received data, performs necessary format checks and normalizes character strings, and corrects any incomplete or invalid data. For example, it normalizes a taste preference such as "fruity and sweet" and performs preprocessing to facilitate database searches.

[0183] 3. Database Search

[0184] server:

[0185] Based on the formatted data, a search is performed on the storage device (database) to find beverage candidates that match the criteria. To do this, a search algorithm such as an SQL query is used. The database stores detailed information about the beverages, word-of-mouth data, and past evaluation data.

[0186] Example search queries:

[0187] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[0188] 4. Candidate evaluation and selection

[0189] server:

[0190] The candidate list obtained from the search process is passed to a generative AI model for evaluation. The generative AI model takes into account past reviews and evaluation data to select the drink that best suits the user's criteria. This enables highly accurate recommendations comparable to those made by a top sommelier.

[0191] Examples of AI evaluation processes:

[0192] candidate:

[0193] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0194] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[0195] AI Selection: Wine A

[0196] 5. Providing information about alcohol

[0197] server:

[0198] Information about the drink that is judged to be the most suitable, such as its name, rating, price, and detailed description, is sent to the user's device. The information sent is standardized in JSON format.

[0199] Examples of information to send:

[0200] {

[0201] "name": "Japanese fruity wine",

[0202] "rating": 4.5,

[0203] "price": 2800,

[0204] "description": "A very fruity and sweet wine that pairs well with pasta."

[0205] }

[0206] Device:

[0207] The received information is displayed to the user, who can then view the details of the recommended beverage.

[0208] 6. User Feedback

[0209] User:

[0210] After tasting the purchased beverage, users can enter their rating and comments in a dedicated feedback form. This feedback will be important data that will help improve the accuracy of the system.

[0211] Examples of feedback include:

[0212] Rating: 4 (out of 5)

[0213] Comment: "Very satisfied. Just as described, perfect with pasta."

[0214] Device:

[0215] The input feedback data is transmitted again to the information processing device.

[0216] server:

[0217] The received feedback is stored in a storage device and used to improve the accuracy of the next beverage recommendation.

[0218] As a result, the system of the present invention allows users to enjoy the perfect beverage that suits their tastes. Furthermore, by continuously collecting feedback, the system's recommendation accuracy continues to improve.

[0219] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0220] Step 1:

[0221] Users access the system using a smartphone, tablet, or computer, and input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs through the interface.

[0222] Input: type of drink, preferred flavor, budget, meal contents

[0223] What happens: A user enters data into a web form or in-app input field and presses the submit button.

[0224] Step 2:

[0225] Terminal: Converts the data entered by the user into JSON format or similar and sends it to the information processing device (server).

[0226] Input: User-entered data

[0227] Output: Formatted data such as JSON

[0228] Specific operation: The data obtained from the input field is encoded in JSON format and sent to the server as an HTTP request.

[0229] Step 3:

[0230] Server: Formats the received data, performs any necessary format checks and string normalization, and checks for incomplete or invalid data.

[0231] Input: JSON format data sent from the terminal

[0232] Output: Normalized and formatted data

[0233] Specific operation: For example, the user input "fruity and sweet" is broken down into individual conditions such as "fruity" and "sweet."

[0234] Step 4:

[0235] Server: Searches for beverage candidates that match the criteria from the storage device (database) based on the normalized data.

[0236] Input: Normalized user-entered data

[0237] Output: A list of drinks that match the criteria

[0238] Specific operation: Using SQL queries against the database, candidates are extracted based on criteria such as "wine," "fruity," "sweet," and "under 3,000 yen."

[0239] Step 5:

[0240] Server: The candidate list obtained as a search result is passed to the generative AI model, which evaluates it.

[0241] Input: A list of beverage candidates extracted from the database

[0242] Output: Evaluation by the generative AI model and selection of the optimal beverage

[0243] Specific operation: Using an evaluation algorithm, each candidate is scored, taking into account word-of-mouth data and evaluation data.

[0244] Step 6:

[0245] Server: Sends information about the beverage that has been determined to be optimal to the user's terminal device.

[0246] Input: Selection results by generative AI model

[0247] Output: Detailed drink information sent to user device

[0248] Specific operation: Information including the name, rating, price, and detailed description of the selected drink is encoded in JSON format and sent to the user's device as an HTTP response.

[0249] Step 7:

[0250] Terminal: Displays the received information to the user, who can then check the details of the recommended beverage.

[0251] Input: detailed drink information sent from the server

[0252] Output: Detailed drink information displayed to the user

[0253] Specific behavior: Display detailed information about the drink on a webpage or app screen for the user to review.

[0254] Step 8:

[0255] User: Tastes the purchased beverage and enters a rating and comments in the feedback form.

[0256] Input: User ratings and comments

[0257] Output: Feedback data

[0258] Specific actions: Enter your ratings and comments in the feedback form and press the submit button.

[0259] Step 9:

[0260] Terminal: Transmits the input feedback data to the information processing device.

[0261] Input: User feedback data

[0262] Output: Feedback data sent to the server

[0263] Specific operation: The feedback data is encoded in JSON format and sent to the server as an HTTP request.

[0264] Step 10:

[0265] Server: The received feedback is stored in a storage device and used to improve the accuracy of recommendations next time.

[0266] Input: Feedback data sent from the device

[0267] Output: Feedback data stored in memory device

[0268] Specific operation: The feedback data is inserted into a database and used as training data for future algorithms.

[0269] (Application example 1)

[0270] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0271] Conventional beverage recommendation systems struggle to easily and effectively find beverages that match users' preferences. They also lack a visual and interactive experience, making it difficult to improve user satisfaction. Furthermore, they lack a means to utilize feedback to improve recommendation accuracy.

[0272] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0273] In this invention, the server includes a means for visually inputting information from the user using a VR head-mounted display, a means for searching for beverage candidates that meet the criteria from a database based on the formatted data, and a means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling the user to easily and effectively find a beverage that suits their preferences through a visual and interactive experience.

[0274] "User" means a consumer who uses the system to receive beverage recommendations.

[0275] "Type of beverage" refers to the category of beverage consumed, such as wine, beer, or sake.

[0276] "Preferred taste" refers to the flavor or characteristics of a beverage desired by a user, including taste attributes such as fruity, sweet, dry, etc.

[0277] "Budget" indicates the maximum amount the user is willing to spend, and the price of the selected drink must not exceed this amount.

[0278] "Meal content" refers to the type of meal the user consumes with the beverage, including specific dish names such as pasta or sushi.

[0279] A "VR head-mounted display" is a display device that allows users to visually experience a virtual reality environment by wearing it.

[0280] The "server" is a computer system that receives user input, formats the data, and uses database search and generative AI to recommend beverages.

[0281] "Formatting" is the process of converting received user input data into an appropriate format, making it suitable for searching and evaluation.

[0282] A "database" is a collection of information that stores information about beverages and word-of-mouth data, and is a target that can be searched based on user criteria.

[0283] "Generative AI" is a type of artificial intelligence that has the ability to evaluate and select the most suitable beverage based on input data.

[0284] "Feedback" refers to the ratings and comments provided by users after their experience, which are used to improve the system's recommendation accuracy.

[0285] "VR space" is a virtual reality world that users experience through a VR head-mounted display.

[0286] System Overview

[0287] In this embodiment, a system is provided in which a user wears a VR head-mounted display and inputs the type of beverage they want to drink, their preferred flavor, their budget, and their meal plan in a virtual space, and the system recommends the most suitable beverage. The system sends the user's input to a server, which performs a database search and evaluation using a generation AI, selects the most suitable beverage, and displays the result in the VR space. The user then provides feedback, which the system stores in a database.

[0288] 1. Accepting user input

[0289] Users put on a VR head-mounted display and log in to the virtual space, where they can use the interface to input the type of drink, their preferred flavor, budget, and meal details.

[0290] 2. Data transmission and formatting

[0291] The device converts the user's input data into JSON format and sends it to the server, which then formats the received data and converts it into the appropriate format.

[0292] 3. Database Search

[0293] The server then searches the database for beverage candidates that match the criteria based on the formatted data, using SQL queries.

[0294] 4. Candidate evaluation and selection

[0295] The server passes the search results to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[0296] 5. Providing information about beverages

[0297] The server sends information about the best drink to the user's device, which then displays the information visually in the VR space.

[0298] 6. User Feedback

[0299] After trying the beverage, users enter their ratings and comments in a dedicated form in the virtual space, and the device sends the feedback to the server.

[0300] 7. Storage and Use of Feedback

[0301] The server stores the received feedback in a database and uses it to improve the accuracy of recommendations next time.

[0302] Hardware and software used

[0303] This system is implemented using the following hardware and software:

[0304] VR head-mounted display: Oculus Quest 2, etc.

[0305] Frontend: VR application

[0306] Server: Flask (Python web framework), SQLite (database)

[0307] Generative AI: OpenAI® GPT-3® 5 API

[0308] Specific examples

[0309] If a user wants to pair a "fruity and sweet wine" with pasta for dinner and under 3,000 yen, the system works as follows: The user puts on the Oculus Quest 2, selects a wine in the VR space, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta." The device converts the input data into JSON format and sends it to the server. The server formats the received data and performs any necessary preprocessing. The server searches its database for wines that meet the criteria. The server passes the search results to the generation AI, which evaluates and selects the most suitable wine. The server then sends information about the selected wine to the user's device, where the user can view it in the VR space. The user purchases the recommended wine and, after tasting it, provides feedback with a rating and comments. The device then sends the provided feedback to the server. The server stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[0310] Prompt Sentence Examples

[0311] In a beverage recommendation system, please recommend the best wine for the user based on the following criteria:

[0312] Type of drink: Wine

[0313] Favorite flavor: Fruity and sweet

[0314] Budget: Under 3,000 yen

[0315] Meal: Pasta

[0316] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0317] Step 1:

[0318] The user wears a VR head-mounted display and accesses the virtual space interface. The user inputs the type of drink, preferred flavor, budget, and meal contents. This input information is received by the terminal via a form in the VR space.

[0319] Step 2:

[0320] The device converts the user's input data into JSON format and sends the input information to the server. The input information includes "type of drink," "preferred flavor," "budget," and "meal details." The input data is formatted in JSON format and sent to the server as an HTTP request.

[0321] Step 3:

[0322] The server then formats the received JSON data, converting it into a format suitable for database queries. This formatting process includes format checks and string normalization for the input data. The formatted data is then in a format that can be applied to database queries.

[0323] Step 4:

[0324] The server uses the formatted data to search the database for drink candidates that match the criteria, using SQL queries to extract drinks that fit the drink type, flavor requirements, and budget. This search result may contain multiple candidates.

[0325] Step 5:

[0326] The server passes the search results to a generative AI model, which evaluates and selects the beverage that best suits the user's criteria. The generative AI model selects the optimal beverage based on the evaluation and word-of-mouth data for each beverage included in the search results.

[0327] Step 6:

[0328] The server converts the information about the optimal drink selected by the generative AI model into JSON format and sends it to the device. The transmitted data includes the drink name, rating, price, and description. The device receives this information and interprets the JSON data.

[0329] Step 7:

[0330] The device visually displays the received beverage information in the VR space. Detailed information about the selected beverage is visually presented to the user in the virtual space interface. The user can review this information and make a purchase decision.

[0331] Step 8:

[0332] After purchasing and experiencing the beverage, users can provide their rating and feedback through a virtual interface, which is entered into a dedicated form.

[0333] Step 9:

[0334] The device converts the feedback provided by the user into JSON format and sends it to the server. The feedback data includes rating values ​​and comments. The server interprets the received feedback data and stores it in a database.

[0335] Step 10:

[0336] The server stores the feedback information in a database and refers to it the next time the server recommends a beverage. The stored feedback is used to improve the accuracy of future recommendations.

[0337] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0338] System Overview

[0339] This system recommends the best drink based on the type of drink a user wants, their preferred flavor, their budget, and their meal. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system can select drinks that take into account the user's emotional information. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers, providing more personalized and precise recommendations.

[0340] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0341] 1. Accepting user input

[0342] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[0343] Example user input:

[0344] Type of drink: Wine

[0345] Favorite flavor: Fruity and sweet

[0346] Budget: Under 3,000 yen

[0347] Meal: Pasta

[0348] Emotional data: facial photos, audio recordings, text comments

[0349] 2. Data transmission and formatting

[0350] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[0351] Example of transmitted data:

[0352] {

[0353] "type": "wine",

[0354] "preferences": "Fruity and sweet",

[0355] "budget": 3000,

[0356] "meal": "pasta",

[0357] "emotion_data": {

[0358] "photo": "image data",

[0359] "voice": "voice data",

[0360] "text": "I'm tired today and would like a relaxing drink"

[0361] }

[0362] }

[0363] 3. Data Formatting

[0364] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[0365] Formatting example:

[0366] Check the format

[0367] String normalization

[0368] Category assignment

[0369] 4. Emotion Data Analysis

[0370] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[0371] Emotion analysis example:

[0372] Emotion: Tired

[0373] 5. Database Search

[0374] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[0375] SQL query example:

[0376] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[0377] 6. Evaluation and selection of candidates

[0378] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[0379] Example of evaluation process:

[0380] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0381] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[0382] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[0383] 7. Transmission of Information

[0384] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[0385] Examples of information to send:

[0386] {

[0387] "name": "Japanese fruity wine",

[0388] "rating": 4.7,

[0389] "price": 2800,

[0390] "description": "A very fruity, sweet and relaxing wine."

[0391] }

[0392] 8. Display of Information

[0393] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[0394] Display example:

[0395] Drink Name: Japanese Fruity Wine

[0396] Rating: 4.7

[0397] Price: 2800 yen

[0398] Description: A very fruity, sweet and relaxing wine.

[0399] 9. Providing Feedback

[0400] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[0401] Feedback input example:

[0402] Rating: 4 (out of 5)

[0403] Comment: "Very satisfied. It was relaxing just as described."

[0404] 10. Submitting Feedback

[0405] Terminal: Sends the feedback information entered by the user to the server.

[0406] Example of submitted feedback:

[0407] {

[0408] "rating": 4,

[0409] "comment": "Very satisfied. It was as relaxing as described."

[0410] }

[0411] 11. Saving Feedback

[0412] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[0413] Save example:

[0414] Feedback data stored in a database

[0415] Use it for your next recommendation

[0416] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[0417] The processing flow will be explained below.

[0418] Step 1: Accepting User Input

[0419] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[0420] example:

[0421] Type of drink: Wine

[0422] Favorite flavor: Fruity and sweet

[0423] Budget: Under 3,000 yen

[0424] Meal: Pasta

[0425] Emotional data: facial photos, audio recordings, text comments

[0426] Step 2: Send data

[0427] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[0428] Example of transmitted data:

[0429] {

[0430] "type": "wine",

[0431] "preferences": "Fruity and sweet",

[0432] "budget": 3000,

[0433] "meal": "pasta",

[0434] "emotion_data": {

[0435] "photo": "image data",

[0436] "voice": "voice data",

[0437] "text": "I'm tired today and would like a relaxing drink"

[0438] }

[0439] }

[0440] Step 3: Data Shaping

[0441] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[0442] Formatting example:

[0443] Check the format

[0444] String normalization

[0445] Category assignment

[0446] Step 4: Analyze the sentiment data

[0447] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[0448] Emotion analysis example:

[0449] Emotion: Tired

[0450] Step 5: Database Search

[0451] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[0452] SQL query example:

[0453] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[0454] Step 6: Evaluate and select candidates

[0455] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[0456] Example of evaluation process:

[0457] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0458] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[0459] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[0460] Step 7: Submit your information

[0461] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[0462] Examples of information to send:

[0463] {

[0464] "name": "Japanese fruity wine",

[0465] "rating": 4.7,

[0466] "price": 2800,

[0467] "description": "A very fruity, sweet and relaxing wine."

[0468] }

[0469] Step 8: Viewing Information

[0470] Terminal: The received information is displayed to the user, who can then check the details of the selected beverage based on the displayed information.

[0471] Display example:

[0472] Drink Name: Japanese Fruity Wine

[0473] Rating: 4.7

[0474] Price: 2800 yen

[0475] Description: A very fruity, sweet and relaxing wine.

[0476] Step 9: Provide feedback

[0477] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[0478] Feedback input example:

[0479] Rating: 4 (out of 5)

[0480] Comment: "Very satisfied. It was relaxing just as described."

[0481] Step 10: Submit your feedback

[0482] Terminal: Sends the feedback information entered by the user to the server.

[0483] Example of submitted feedback:

[0484] {

[0485] "rating": 4,

[0486] "comment": "Very satisfied. It was as relaxing as described."

[0487] }

[0488] Step 11: Save your feedback

[0489] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[0490] Save example:

[0491] Feedback data stored in a database

[0492] Use it for your next recommendation

[0493] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[0494] Example 2

[0495] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0496] Conventional beverage recommendation systems recommend drinks based on a user's preferred taste, budget, and dietary preferences, but because they are unable to consider the user's emotional information, it is difficult to recommend the optimal drink for the user's psychological state. Furthermore, there is a lack of technology to effectively utilize feedback to improve recommendation accuracy.

[0497] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0498] In this invention, the server includes means for extracting emotion data using an emotion engine that analyzes user emotions, means for searching a database for beverage candidates that meet the criteria based on the formatted data and analyzed emotion data, and means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling sophisticated beverage recommendations based on the user's preferences and emotion information.

[0499] "User" means an individual or entity that intends to use the System to select a beverage.

[0500] "Beverage type" refers to the category of beverage selected, such as wine, beer, juice, etc.

[0501] "Preferred taste" refers to the flavor or characteristics of the beverage that the user prefers, such as fruity, sweet, dry, etc.

[0502] "Budget" refers to the monetary limit a User is willing to pay for a beverage.

[0503] "Meal content" refers to the type and menu of food that the user consumes along with the beverage they drink.

[0504] "Emotional Data" refers to data that indicates a user's emotional state analyzed from facial expressions, voice, and text.

[0505] "Terminal" means a device used by a User to access the System, including a smartphone, tablet, or PC.

[0506] "Server" refers to the central processing unit that processes, stores, and analyzes data submitted by users.

[0507] "Formatting" refers to the process of processing data received from a user and converting it into a usable format.

[0508] "Emotion engine" refers to software or hardware for analyzing a user's facial expressions, voice, and text to extract emotional data.

[0509] "Database" refers to a data management system that stores information about beverages and makes it accessible by query.

[0510] "Generative AI" refers to an artificial intelligence model that evaluates candidates based on past data and reviews and selects the most suitable beverage.

[0511] "Feedback" refers to the ratings and comments that users make about the beverages they have experienced.

[0512] The present invention is a system that recommends optimal drinks based on the user's input of the type of drink they want to drink, their preferred taste, budget, dietary content, and emotional data. Detailed explanations and concrete examples necessary for implementing this system are provided below.

[0513] Accepting user input

[0514] Users access the system from devices such as smartphones, tablets, and PCs and enter the type of beverage they want to drink, their preferred flavor, budget, dietary preferences, and emotional data. For example, users can enter the following data:

[0515] Example prompt sentence:

[0516] Type of drink: Wine

[0517] Favorite flavor: Fruity and sweet

[0518] Budget: Under 3,000 yen

[0519] Meal: Pasta

[0520] Emotional data: "I'm tired today and would like a drink to help me relax."

[0521] Data transmission and formatting

[0522] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input content and checks for errors. For example, the device generates the following JSON data and sends it to the server.

[0523] {

[0524] "type": "wine",

[0525] "preferences": "Fruity and sweet",

[0526] "budget": 3000,

[0527] "meal": "pasta",

[0528] "emotion_data": {

[0529] "photo": "image data",

[0530] "voice": "voice data",

[0531] "text": "I'm tired today and would like a relaxing drink"

[0532] }

[0533] }

[0534] Data Shaping and Sentiment Analysis

[0535] The server formats the data received from the device. This process includes format checks, string normalization, and category assignment. It also uses an emotion engine to analyze emotion data, extracting emotion data from the user's facial expressions, voice, and text. For example, it can detect fatigue levels from the text "I'm tired today."

[0536] Database search

[0537] The server queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that meet the criteria. SQL is used for query processing. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[0538] Candidate evaluation and selection

[0539] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the best wine from candidate wines A, B, and C.

[0540] Sending and Displaying Information

[0541] The server sends the information about the selected optimal drink to the user's terminal. The terminal displays the received information to the user. For example, the following information is displayed:

[0542] Display example:

[0543] Drink Name: Japanese Fruity Wine

[0544] Rating: 4.7

[0545] Price: 2800 yen

[0546] Description: A very fruity, sweet and relaxing wine.

[0547] Providing and storing feedback

[0548] After trying the beverage, the user provides feedback such as a rating and comments. The device sends the feedback information entered by the user to the server. The server receives the feedback and stores it in a database. This feedback will be used to improve the accuracy of the recommendation system in the future.

[0549] This allows users to easily select the perfect drink based on their preferences and emotions, and rate and provide feedback on their experience, allowing for a more personalized and refined selection of drinks.

[0550] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0551] Step 1:

[0552] Users access the system using devices such as smartphones, tablets, and PCs and input the type of beverage they want, their preferred flavor, budget, dietary preferences, and emotional data. This input includes a facial photo, audio recording, and text comments. An example of a prompt sentence is, "Please choose your favorite wine. I like fruity and sweet flavors, and I'd like something that goes well with pasta, within a budget of 3,000 yen or less. I'm tired after a long day today, so I'd like a drink that helps me relax."

[0553] Input: type of drink, preferred flavor, budget, meal contents, emotional data

[0554] Output: Information entered by the user

[0555] Step 2:

[0556] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input and performs error checks. For example, it checks whether all required fields are filled in. Data that passes the error check is composed as JSON and sent.

[0557] Input: User-entered data

[0558] Output: JSON format data

[0559] Step 3:

[0560] The server formats the JSON data received from the device. This includes formatting, string normalization, and category assignment. For example, it might verify that the type of drink entered is correct and correct any typos in the string. The formatted data is then converted into a standard format for internal processing.

[0561] Input: JSON format data

[0562] Output: Formatted user data

[0563] Step 4:

[0564] The server uses an emotion engine to analyze emotion data from the user's facial expressions, voice, and text. For example, the emotion engine detects the user's fatigue state from text such as "I'm tired." Based on the results of this analysis, the server quantifies or categorizes the user's emotional state.

[0565] Input: User's emotional data (face photo, voice, text)

[0566] Output: Analyzed emotion data (e.g., fatigue level)

[0567] Step 5:

[0568] The server queries the database based on the formatted beverage data and the analyzed emotion data to find beverage candidates that meet the criteria. This query uses SQL. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[0569] Input: formatted beverage data, parsed emotion data

[0570] Output: List of beverage candidates

[0571] Step 6:

[0572] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the candidate with the highest evaluation score.

[0573] Input: Beverage candidate list, review data, past evaluation data

[0574] Output: Best beverage candidates

[0575] Step 7:

[0576] The server then sends information about the selected optimal drink to the user's device. The information sent includes the name of the drink, its rating, price, and a detailed description. For example, "Japanese fruity wine, rating 4.7, price 2,800 yen, very fruity, sweet, and relaxing."

[0577] Input: Best beverage candidates

[0578] Output: detailed information about the beverage

[0579] Step 8:

[0580] The device then displays the received information to the user, allowing the user to check the details of the selected beverage based on the displayed information. For example, the name of the beverage, its rating, price, and detailed description are visually displayed.

[0581] Input: Beverage details

[0582] Output: Beverage information displayed to the user

[0583] Step 9:

[0584] After trying the beverage, users provide feedback by rating and commenting on it. They enter their rating in a special feedback form and press the "Submit" button. For example, a rating of 4 (out of 5) and a comment like "I was very satisfied. It was relaxing, just as described."

[0585] Input: Feedback after the experience

[0586] Output: Feedback information

[0587] Step 10:

[0588] The terminal sends the feedback information entered by the user to the server. The sending process includes converting the feedback data into JSON format and sending it.

[0589] Input: User feedback information

[0590] Output: Feedback data sent to the server

[0591] Step 11:

[0592] The server receives the feedback and stores it in a database, which can be used to improve the accuracy of the recommendation system in the future, for example, by improving the recommendation algorithm next time.

[0593] Input: Feedback data sent to the server

[0594] Output: Feedback stored in a database

[0595] (Application example 2)

[0596] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0597] In today's food and beverage selection process, it is difficult to make recommendations that accurately reflect a user's individual preferences and current emotional state. Food delivery services also require recommendation systems that can quickly respond to diverse user needs. While existing systems can recommend beverages based on a user's preferences and budget, they are unable to take into account the user's emotions. This makes it difficult for users to select a beverage that satisfies them. Furthermore, there has been a lack of systems that can improve the accuracy of recommendations based on feedback.

[0598] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0599] In this invention, the server includes means for formatting the type, preferences, budget, meal details, and emotion data entered by the user and performing the necessary preprocessing, means for analyzing emotions using an emotion engine based on the emotion data, and means for searching a database for beverage candidates that meet the conditions based on the analyzed emotion data, thereby making it possible to recommend optimal beverages that match the user's individual preferences and emotional state.

[0600] "User" refers to a person who uses the system to receive beverage recommendations.

[0601] "Beverage type" refers to the classification of different beverages, such as wine, beer, juice, etc.

[0602] "Preferred taste" refers to the taste characteristics of a beverage that a user prefers, such as fruity, sweet, or dry.

[0603] "Budget" refers to the maximum amount a user is willing to pay for a beverage.

[0604] "Meal Content" refers to the type and details of the meal the user plans to consume with the beverage.

[0605] "Emotional Data" refers to data about a user's current emotional state, including a photo of the user's face, audio recordings, text comments, etc.

[0606] "Formatting" refers to the process of converting user-entered data into a format suitable for analysis and search.

[0607] "Preprocessing" refers to preparing data to make it easier to analyze, such as normalizing the data or assigning categories.

[0608] An "emotion engine" refers to a system that analyzes a user's emotions from data such as facial photos, voice, and text.

[0609] A "database" refers to a collection of information that systematically stores information such as types of beverages, flavors, prices, and ratings.

[0610] "Candidates" refers to a list of beverages extracted based on the user's conditions and emotions.

[0611] "Generative AI" refers to an artificial intelligence algorithm that selects the optimal beverage based on given data.

[0612] "Terminal" refers to a device that a user uses to operate the system, such as a smartphone, tablet, or PC.

[0613] "Feedback" refers to ratings and comments provided by users based on their usage experience.

[0614] "Storing in a database" refers to storing feedback and other data in a database so that it can be reused for later retrieval and analysis.

[0615] The present invention relates to a system for recommending optimal beverages based on a user's individual preferences and emotional state, which is implemented using a set of hardware and software, such as:

[0616] System Overview

[0617] 1. Accepting user input

[0618] Users use devices such as smartphones, tablets, and computers to input the type of drink they want, their preferred flavor, budget, meal contents, and emotional data, which includes a photo of their face, audio recordings, and text comments.

[0619] 2. Data transmission and formatting

[0620] The device converts the user's input data and emotion data into JSON format and sends it to the server, which then performs formatting processes such as format checks and normalization on the received data.

[0621] 3. Emotion Data Analysis

[0622] The server uses an emotion engine (e.g., IBM Watson®, Microsoft® Azure® Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format and include emotional states such as "fatigue."

[0623] 4. Search for candidates

[0624] The server uses an SQL query to search the database for beverage candidates that meet the analyzed emotional data and the user's formatted conditions.

[0625] 5. Candidate evaluation and selection

[0626] The server uses generative AI (e.g., OpenAI's GPT model) to evaluate beverage candidates retrieved from the database and select the beverage that best suits the user's conditions and emotions.

[0627] 6. Transmission and Display of Information

[0628] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which then displays this information to the user.

[0629] 7. Providing and Saving Feedback

[0630] After tasting the beverage, the user sends their rating and comments to the server, which stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[0631] Specific examples

[0632] Prompt Sentence Examples

[0633] Choose the best beverage from the following options: Emotional information: Fatigue, Options: [Wine A (Rating: 4.5, Review: "Very fruity and sweet, perfect with pasta"), Wine B (Rating: 4.2, Review: "Well-balanced, with a touch of sweetness"), Wine C (Rating: 4.7, Review: "Relaxing taste")]

[0634] By inputting this prompt into a generative AI model, the optimal beverage can be selected based on the user's emotions and review information. This invention enables more personalized and precise beverage recommendations based on the user's emotional state. This allows users to easily select the beverage that best suits their specific emotional state, improving the quality of their dining experience.

[0635] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0636] Step 1:

[0637] Users use devices such as smartphones, tablets, and PCs to input the type of drink they want, their preferred flavor, budget, dietary preferences, and emotional data, including facial photos, audio recordings, and text comments. This data is collected through an input form on the device.

[0638] Input: User's drink type, preferred flavor, budget, dietary details, emotional data

[0639] Output: User input data in JSON format

[0640] Step 2:

[0641] The device converts the user's input data and emotion data into JSON format and sends it to the server, checking the data format and checking for errors before sending.

[0642] Input: User-entered data

[0643] Output: HTTP request with data in JSON format

[0644] Step 3:

[0645] The server performs format checks and normalization on the received data, including normalizing strings and assigning categories, and then sends the resulting data to a sentiment analysis engine.

[0646] Input: User-supplied data in JSON format

[0647] Output: Formatted data and data for sentiment analysis

[0648] Step 4:

[0649] The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format, and include emotional states such as "fatigue."

[0650] Input: Formatted data and sentiment data

[0651] Output: Parsed emotional state (e.g. "fatigue")

[0652] Step 5:

[0653] The server uses SQL queries to search for beverage candidates that match the analyzed emotion data and the user's preferences from a database containing information such as the type of beverage, flavor characteristics, price, and suitable meals.

[0654] Input: A search query for a beverage that matches your criteria (type, preference, budget, diet, emotional state)

[0655] Output: A list of beverage candidates as search results

[0656] Step 6:

[0657] The server evaluates and prioritizes potential drinks using a generative AI (e.g., OpenAI's GPT model). The model selects the best drink based on the user's sentiment and reviews based on the input prompt.

[0658] Input: Drink candidate list and emotional state

[0659] Output: Best drink information (name, rating, price, detailed description)

[0660] Step 7:

[0661] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which receives this information and displays it in a user-friendly format.

[0662] Input: Optimal beverage information

[0663] Output: Beverage information displayed on the terminal

[0664] Step 8:

[0665] After tasting the beverage, the user sends their rating and comments to the server from their device using a feedback form provided in the UI.

[0666] Input: User ratings and comments

[0667] Output: Feedback data in JSON format

[0668] Step 9:

[0669] The server verifies the received feedback and stores it in a database, which then serves as data to improve the accuracy of future recommendations.

[0670] Input: Feedback data in JSON format

[0671] Output: Ratings and comments stored in the database

[0672] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0673] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0674] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0675] [Second embodiment]

[0676] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0677] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0678] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0679] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0680] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0681] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0682] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0683] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0684] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0685] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0686] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0687] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0688] System Overview

[0689] This system recommends the perfect drink based on the type of drink the user wants, their preferred flavor, budget, and meal content. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers. The system accepts user input, searches a database based on that input, and uses generative AI to select the best drink candidates and provide information.

[0690] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0691] 1. Accepting user input

[0692] User: Accesses the system and inputs the type of beverage they want to drink (e.g., wine, beer, sake, etc.), their preferred flavor (e.g., fruity, sweet, dry, etc.), their budget (e.g., under 3,000 yen), and their meal plan (e.g., pasta, sushi, etc.). This input is done via a device such as a smartphone, tablet, or PC.

[0693] Example user input:

[0694] Type of drink: Wine

[0695] Favorite flavor: Fruity and sweet

[0696] Budget: Under 3,000 yen

[0697] Meal: Pasta

[0698] 2. Data transmission and formatting

[0699] Terminal: Sends input data from the user to the server. The data is converted into an appropriate format (e.g., JSON) and sent to the server.

[0700] Server: Formats the received data, performs any necessary format checks, normalizes strings, assigns categories, and converts it into a form suitable for searching and evaluation.

[0701] 3. Database Search

[0702] Server: Based on the formatted data, the server searches the database (stored information about beverages and reviews) for beverage candidates that match the criteria. Methods such as SQL queries are used for the search process.

[0703] SQL query example:

[0704] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[0705] 4. Candidate evaluation and selection

[0706] Server: The search results are passed to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[0707] AI evaluation process example:

[0708] Candidates evaluated by AI:

[0709] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0710] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[0711] 5. Providing information about alcohol

[0712] Server: Sends information about the drink that is determined to be the most suitable (such as name, rating, price, and detailed description) to the user's device.

[0713] Examples of information to send:

[0714] {

[0715] "name": "Japanese fruity wine",

[0716] "rating": 4.5,

[0717] "price": 2800,

[0718] "description": "A very fruity and sweet wine that pairs well with pasta."

[0719] }

[0720] Terminal: Displays the received information to the user, allowing the user to check the details of the recommended beverage.

[0721] 6. User Feedback

[0722] User: After experiencing the purchased beverage, provide a rating and comment. Enter your feedback in a dedicated form.

[0723] Feedback input example:

[0724] Rating: 4 (out of 5)

[0725] Comment: "Very satisfied. Just as described, perfect with pasta."

[0726] Terminal: Sends user feedback to the server.

[0727] Server: The received feedback is stored in a database and used to improve the accuracy of recommendations next time.

[0728] Specific examples

[0729] For example, if a user wants to drink a "fruity and sweet wine" with "pasta" for dinner for "less than 3,000 yen," the system operates as follows.

[0730] 1. User: Accesses the system app on a smartphone, selects a wine, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta."

[0731] 2. Terminal: Converts the input data into JSON format or similar and sends it to the server.

[0732] 3. Server: Formats the received data and performs any necessary preprocessing.

[0733] 4. Server: Searches the database for wines that meet the above criteria.

[0734] 5. Server: Passes the search results to the generation AI, which evaluates and selects the most suitable wine.

[0735] 6. Server: Sends information about the selected wine to the user's device so that the user can check the information.

[0736] 7. User: Purchases the recommended wine, drinks it, and then provides feedback with a rating and comments.

[0737] 8. Terminal: Sends the provided feedback to the server.

[0738] 9. Server: The feedback is stored in a database and used to improve future recommendations.

[0739] This allows users to easily select the beverage that best suits their tastes and also allows them to easily rate and provide feedback.

[0740] The processing flow will be explained below.

[0741] Step 1: Accepting User Input

[0742] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and meal details. This can be done using a device such as a smartphone, tablet, or PC.

[0743] example:

[0744] Type of drink: Wine

[0745] Favorite flavor: Fruity and sweet

[0746] Budget: Under 3,000 yen

[0747] Meal: Pasta

[0748] Step 2: Send data

[0749] Terminal: Converts user input data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input for errors.

[0750] Example of transmitted data:

[0751] {

[0752] "type": "wine",

[0753] "preferences": "Fruity and sweet",

[0754] "budget": 3000,

[0755] "meal": "pasta"

[0756] }

[0757] Step 3: Data Shaping

[0758] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment).

[0759] Formatting example:

[0760] Check the format

[0761] String normalization

[0762] Category assignment

[0763] Step 4: Database Search

[0764] Server: Queries the database based on the formatted data to find beverage candidates that match the criteria. SQL is used for query processing.

[0765] SQL query example:

[0766] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[0767] Step 5: Evaluate and select candidates

[0768] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best meets the user's criteria. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[0769] Example of evaluation process:

[0770] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0771] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[0772] Step 6: Submit your information

[0773] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[0774] Examples of information to send:

[0775] {

[0776] "name": "Japanese fruity wine",

[0777] "rating": 4.5,

[0778] "price": 2800,

[0779] "description": "A very fruity and sweet wine that pairs well with pasta."

[0780] }

[0781] Step 7: Viewing information

[0782] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[0783] Display example:

[0784] Drink Name: Japanese Fruity Wine

[0785] Rating: 4.5

[0786] Price: 2800 yen

[0787] Description: A very fruity and sweet wine that pairs well with pasta.

[0788] Step 8: Provide feedback

[0789] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[0790] Feedback input example:

[0791] Rating: 4 (out of 5)

[0792] Comment: "Very satisfied. Just as described, perfect with pasta."

[0793] Step 9: Submit your feedback

[0794] Terminal: Sends the feedback information entered by the user to the server.

[0795] Example of submitted feedback:

[0796] {

[0797] "rating": 4,

[0798] "comment": "Very satisfied. Just as described, it was a perfect match for pasta."

[0799] }

[0800] Step 10: Save your feedback

[0801] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[0802] Save example:

[0803] Feedback data stored in a database

[0804] Use it for your next recommendation

[0805] Through these steps, users can easily choose the best beverage that suits their tastes and rate and provide feedback on their experience.

[0806] Example 1

[0807] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0808] Conventional beverage recommendation systems have difficulty in selecting the optimal beverage for a user's individual preferences, budget, and meal content, and it has been particularly difficult to achieve highly accurate recommendations comparable to those made by a top sommelier. Furthermore, they lacked a mechanism for utilizing user feedback in future recommendations. This resulted in a high risk of users choosing a beverage that did not suit their preferences, often resulting in dissatisfaction with the beverage selection.

[0809] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0810] In this invention, the server includes a means for the user to input the type of beverage they want to drink, their preferred flavor, their budget, and their meal details; a means for transmitting the user's input to the information processing device; and a means for the information processing device to format the user's input and perform any necessary preprocessing. This enables the information processing device to select the optimal beverage based on the user's input data using a highly accurate database search and a generative AI model. Furthermore, by utilizing user feedback to improve the accuracy of the next recommendation, it is possible to continuously provide highly satisfying selection results.

[0811] A "user" is a person who accesses the system to input the type of beverage they wish to drink, their preferred taste, budget, and dietary needs to receive recommendations.

[0812] The "information processing device" is a device that receives user input data, formats it and performs necessary preprocessing, and then evaluates and selects candidate beverages using a generative AI model.

[0813] The "storage device" is a device that stores information about beverages, word-of-mouth data, past evaluation data, etc., and stores a database that the information processing device uses for searches and evaluations.

[0814] A "generative AI model" is an artificial intelligence model that evaluates potential beverages based on past word-of-mouth and rating data, and selects the beverage that best suits the user's requirements.

[0815] A "terminal device" is a device that transmits user input to an information processing device and displays information about the selected drink to the user. Examples of such devices include smartphones, tablets, and personal computers.

[0816] "Feedback" refers to the ratings and comments provided by users after trying a recommended beverage, and is information used to improve the accuracy of recommendations next time.

[0817] The following describes in detail an embodiment of the present invention, specifically showing what hardware and software are used to process and calculate data.

[0818] 1. Accepting user input

[0819] User:

[0820] First, users access the system using a smartphone, tablet, or PC. Through the system's interface (web application or native application), they input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs. This allows users to easily select a beverage that suits their preferences.

[0821] Example of input contents:

[0822] Type of drink: Wine

[0823] Favorite flavor: Fruity and sweet

[0824] Budget: Under 3,000 yen

[0825] Meal: Pasta

[0826] 2. Data transmission and formatting

[0827] Device:

[0828] The data entered by the user is first converted into an appropriate format (e.g., JSON) and then sent to the information processing device, so that the input data is delivered to the server in a unified format.

[0829] server:

[0830] The information processing device (server) formats the received data, performs necessary format checks and normalizes character strings, and corrects any incomplete or invalid data. For example, it normalizes a taste preference such as "fruity and sweet" and performs preprocessing to facilitate database searches.

[0831] 3. Database Search

[0832] server:

[0833] Based on the formatted data, a search is performed on the storage device (database) to find beverage candidates that match the criteria. To do this, a search algorithm such as an SQL query is used. The database stores detailed information about the beverages, word-of-mouth data, and past evaluation data.

[0834] Example search queries:

[0835] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[0836] 4. Candidate evaluation and selection

[0837] server:

[0838] The candidate list obtained from the search process is passed to a generative AI model for evaluation. The generative AI model takes into account past reviews and evaluation data to select the drink that best suits the user's criteria. This enables highly accurate recommendations comparable to those made by a top sommelier.

[0839] Examples of AI evaluation processes:

[0840] candidate:

[0841] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[0842] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[0843] AI Selection: Wine A

[0844] 5. Providing information about alcohol

[0845] server:

[0846] Information about the drink that is judged to be the most suitable, such as its name, rating, price, and detailed description, is sent to the user's device. The information sent is standardized in JSON format.

[0847] Examples of information to send:

[0848] {

[0849] "name": "Japanese fruity wine",

[0850] "rating": 4.5,

[0851] "price": 2800,

[0852] "description": "A very fruity and sweet wine that pairs well with pasta."

[0853] }

[0854] Device:

[0855] The received information is displayed to the user, who can then view the details of the recommended beverage.

[0856] 6. User Feedback

[0857] User:

[0858] After tasting the purchased beverage, users can enter their rating and comments in a dedicated feedback form. This feedback will be important data that will help improve the accuracy of the system.

[0859] Examples of feedback include:

[0860] Rating: 4 (out of 5)

[0861] Comment: "Very satisfied. Just as described, perfect with pasta."

[0862] Device:

[0863] The input feedback data is transmitted again to the information processing device.

[0864] server:

[0865] The received feedback is stored in a storage device and used to improve the accuracy of the next beverage recommendation.

[0866] As a result, the system of the present invention allows users to enjoy the perfect beverage that suits their tastes. Furthermore, by continuously collecting feedback, the system's recommendation accuracy continues to improve.

[0867] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0868] Step 1:

[0869] Users access the system using a smartphone, tablet, or computer, and input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs through the interface.

[0870] Input: type of drink, preferred flavor, budget, meal contents

[0871] What happens: A user enters data into a web form or in-app input field and presses the submit button.

[0872] Step 2:

[0873] Terminal: Converts the data entered by the user into JSON format or similar and sends it to the information processing device (server).

[0874] Input: User-entered data

[0875] Output: Formatted data such as JSON

[0876] Specific operation: The data obtained from the input field is encoded in JSON format and sent to the server as an HTTP request.

[0877] Step 3:

[0878] Server: Formats the received data, performs any necessary format checks and string normalization, and checks for incomplete or invalid data.

[0879] Input: JSON format data sent from the terminal

[0880] Output: Normalized and formatted data

[0881] Specific operation: For example, the user input "fruity and sweet" is broken down into individual conditions such as "fruity" and "sweet."

[0882] Step 4:

[0883] Server: Searches for beverage candidates that match the criteria from the storage device (database) based on the normalized data.

[0884] Input: Normalized user-entered data

[0885] Output: A list of drinks that match the criteria

[0886] Specific operation: Using SQL queries against the database, candidates are extracted based on criteria such as "wine," "fruity," "sweet," and "under 3,000 yen."

[0887] Step 5:

[0888] Server: The candidate list obtained as a search result is passed to the generative AI model, which evaluates it.

[0889] Input: A list of beverage candidates extracted from the database

[0890] Output: Evaluation by the generative AI model and selection of the optimal beverage

[0891] Specific operation: Using an evaluation algorithm, each candidate is scored, taking into account word-of-mouth data and evaluation data.

[0892] Step 6:

[0893] Server: Sends information about the beverage that has been determined to be optimal to the user's terminal device.

[0894] Input: Selection results by generative AI model

[0895] Output: Detailed drink information sent to user device

[0896] Specific operation: Information including the name, rating, price, and detailed description of the selected drink is encoded in JSON format and sent to the user's device as an HTTP response.

[0897] Step 7:

[0898] Terminal: Displays the received information to the user, who can then check the details of the recommended beverage.

[0899] Input: detailed drink information sent from the server

[0900] Output: Detailed drink information displayed to the user

[0901] Specific behavior: Display detailed information about the drink on a webpage or app screen for the user to review.

[0902] Step 8:

[0903] User: Tastes the purchased beverage and enters a rating and comments in the feedback form.

[0904] Input: User ratings and comments

[0905] Output: Feedback data

[0906] Specific actions: Enter your ratings and comments in the feedback form and press the submit button.

[0907] Step 9:

[0908] Terminal: Transmits the input feedback data to the information processing device.

[0909] Input: User feedback data

[0910] Output: Feedback data sent to the server

[0911] Specific operation: The feedback data is encoded in JSON format and sent to the server as an HTTP request.

[0912] Step 10:

[0913] Server: The received feedback is stored in a storage device and used to improve the accuracy of recommendations next time.

[0914] Input: Feedback data sent from the device

[0915] Output: Feedback data stored in memory device

[0916] Specific operation: The feedback data is inserted into a database and used as training data for future algorithms.

[0917] (Application example 1)

[0918] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0919] Conventional beverage recommendation systems struggle to easily and effectively find beverages that match users' preferences. They also lack a visual and interactive experience, making it difficult to improve user satisfaction. Furthermore, they lack a means to utilize feedback to improve recommendation accuracy.

[0920] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0921] In this invention, the server includes a means for visually inputting information from the user using a VR head-mounted display, a means for searching for beverage candidates that meet the criteria from a database based on the formatted data, and a means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling the user to easily and effectively find a beverage that suits their preferences through a visual and interactive experience.

[0922] "User" means a consumer who uses the system to receive beverage recommendations.

[0923] "Type of beverage" refers to the category of beverage consumed, such as wine, beer, or sake.

[0924] "Preferred taste" refers to the flavor or characteristics of a beverage desired by a user, including taste attributes such as fruity, sweet, dry, etc.

[0925] "Budget" indicates the maximum amount the user is willing to spend, and the price of the selected drink must not exceed this amount.

[0926] "Meal content" refers to the type of meal the user consumes with the beverage, including specific dish names such as pasta or sushi.

[0927] A "VR head-mounted display" is a display device that allows users to visually experience a virtual reality environment by wearing it.

[0928] The "server" is a computer system that receives user input, formats the data, and uses database search and generative AI to recommend beverages.

[0929] "Formatting" is the process of converting received user input data into an appropriate format, making it suitable for searching and evaluation.

[0930] A "database" is a collection of information that stores information about beverages and word-of-mouth data, and is a target that can be searched based on user criteria.

[0931] "Generative AI" is a type of artificial intelligence that has the ability to evaluate and select the most suitable beverage based on input data.

[0932] "Feedback" refers to the ratings and comments provided by users after their experience, which are used to improve the system's recommendation accuracy.

[0933] "VR space" is a virtual reality world that users experience through a VR head-mounted display.

[0934] System Overview

[0935] In this embodiment, a system is provided in which a user wears a VR head-mounted display and inputs the type of beverage they want to drink, their preferred flavor, their budget, and their meal plan in a virtual space, and the system recommends the most suitable beverage. The system sends the user's input to a server, which performs a database search and evaluation using a generation AI, selects the most suitable beverage, and displays the result in the VR space. The user then provides feedback, which the system stores in a database.

[0936] 1. Accepting user input

[0937] Users put on a VR head-mounted display and log in to the virtual space, where they can use the interface to input the type of drink, their preferred flavor, budget, and meal details.

[0938] 2. Data transmission and formatting

[0939] The device converts the user's input data into JSON format and sends it to the server, which then formats the received data and converts it into the appropriate format.

[0940] 3. Database Search

[0941] The server then searches the database for beverage candidates that match the criteria based on the formatted data, using SQL queries.

[0942] 4. Candidate evaluation and selection

[0943] The server passes the search results to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[0944] 5. Providing information about beverages

[0945] The server sends information about the best drink to the user's device, which then displays the information visually in the VR space.

[0946] 6. User Feedback

[0947] After trying the beverage, users enter their ratings and comments in a dedicated form in the virtual space, and the device sends the feedback to the server.

[0948] 7. Storage and Use of Feedback

[0949] The server stores the received feedback in a database and uses it to improve the accuracy of recommendations next time.

[0950] Hardware and software used

[0951] This system is implemented using the following hardware and software:

[0952] VR head-mounted display: Oculus Quest 2, etc.

[0953] Frontend: VR application

[0954] Server: Flask (Python web framework), SQLite (database)

[0955] Generation AI: OpenAI GPT-3.5 API

[0956] Specific examples

[0957] If a user wants to pair a "fruity and sweet wine" with pasta for dinner and under 3,000 yen, the system works as follows: The user puts on the Oculus Quest 2, selects a wine in the VR space, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta." The device converts the input data into JSON format and sends it to the server. The server formats the received data and performs any necessary preprocessing. The server searches its database for wines that meet the criteria. The server passes the search results to the generation AI, which evaluates and selects the most suitable wine. The server then sends information about the selected wine to the user's device, where the user can view it in the VR space. The user purchases the recommended wine and, after tasting it, provides feedback with a rating and comments. The device then sends the provided feedback to the server. The server stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[0958] Prompt Sentence Examples

[0959] In a beverage recommendation system, please recommend the best wine for the user based on the following criteria:

[0960] Type of drink: Wine

[0961] Favorite flavor: Fruity and sweet

[0962] Budget: Under 3,000 yen

[0963] Meal: Pasta

[0964] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0965] Step 1:

[0966] The user wears a VR head-mounted display and accesses the virtual space interface. The user inputs the type of drink, preferred flavor, budget, and meal contents. This input information is received by the terminal via a form in the VR space.

[0967] Step 2:

[0968] The device converts the user's input data into JSON format and sends the input information to the server. The input information includes "type of drink," "preferred flavor," "budget," and "meal details." The input data is formatted in JSON format and sent to the server as an HTTP request.

[0969] Step 3:

[0970] The server then formats the received JSON data, converting it into a format suitable for database queries. This formatting process includes format checks and string normalization for the input data. The formatted data is then in a format that can be applied to database queries.

[0971] Step 4:

[0972] The server uses the formatted data to search the database for drink candidates that match the criteria, using SQL queries to extract drinks that fit the drink type, flavor requirements, and budget. This search result may contain multiple candidates.

[0973] Step 5:

[0974] The server passes the search results to a generative AI model, which evaluates and selects the beverage that best suits the user's criteria. The generative AI model selects the optimal beverage based on the evaluation and word-of-mouth data for each beverage included in the search results.

[0975] Step 6:

[0976] The server converts the information about the optimal drink selected by the generative AI model into JSON format and sends it to the device. The transmitted data includes the drink name, rating, price, and description. The device receives this information and interprets the JSON data.

[0977] Step 7:

[0978] The device visually displays the received beverage information in the VR space. Detailed information about the selected beverage is visually presented to the user in the virtual space interface. The user can review this information and make a purchase decision.

[0979] Step 8:

[0980] After purchasing and experiencing the beverage, users can provide their rating and feedback through a virtual interface, which is entered into a dedicated form.

[0981] Step 9:

[0982] The device converts the feedback provided by the user into JSON format and sends it to the server. The feedback data includes rating values ​​and comments. The server interprets the received feedback data and stores it in a database.

[0983] Step 10:

[0984] The server stores the feedback information in a database and refers to it the next time the server recommends a beverage. The stored feedback is used to improve the accuracy of future recommendations.

[0985] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0986] System Overview

[0987] This system recommends the best drink based on the type of drink a user wants, their preferred flavor, their budget, and their meal. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system can select drinks that take into account the user's emotional information. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers, providing more personalized and precise recommendations.

[0988] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0989] 1. Accepting user input

[0990] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[0991] Example user input:

[0992] Type of drink: Wine

[0993] Favorite flavor: Fruity and sweet

[0994] Budget: Under 3,000 yen

[0995] Meal: Pasta

[0996] Emotional data: facial photos, audio recordings, text comments

[0997] 2. Data transmission and formatting

[0998] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[0999] Example of transmitted data:

[1000] {

[1001] "type": "wine",

[1002] "preferences": "Fruity and sweet",

[1003] "budget": 3000,

[1004] "meal": "pasta",

[1005] "emotion_data": {

[1006] "photo": "image data",

[1007] "voice": "voice data",

[1008] "text": "I'm tired today and would like a relaxing drink"

[1009] }

[1010] }

[1011] 3. Data Formatting

[1012] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[1013] Formatting example:

[1014] Check the format

[1015] String normalization

[1016] Category assignment

[1017] 4. Emotion Data Analysis

[1018] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[1019] Emotion analysis example:

[1020] Emotion: Tired

[1021] 5. Database Search

[1022] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[1023] SQL query example:

[1024] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[1025] 6. Evaluation and selection of candidates

[1026] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[1027] Example of evaluation process:

[1028] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[1029] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[1030] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[1031] 7. Transmission of Information

[1032] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[1033] Examples of information to send:

[1034] {

[1035] "name": "Japanese fruity wine",

[1036] "rating": 4.7,

[1037] "price": 2800,

[1038] "description": "A very fruity, sweet and relaxing wine."

[1039] }

[1040] 8. Display of Information

[1041] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[1042] Display example:

[1043] Drink Name: Japanese Fruity Wine

[1044] Rating: 4.7

[1045] Price: 2800 yen

[1046] Description: A very fruity, sweet and relaxing wine.

[1047] 9. Providing Feedback

[1048] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[1049] Feedback input example:

[1050] Rating: 4 (out of 5)

[1051] Comment: "Very satisfied. It was relaxing just as described."

[1052] 10. Submitting Feedback

[1053] Terminal: Sends the feedback information entered by the user to the server.

[1054] Example of submitted feedback:

[1055] {

[1056] "rating": 4,

[1057] "comment": "Very satisfied. It was as relaxing as described."

[1058] }

[1059] 11. Saving Feedback

[1060] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[1061] Save example:

[1062] Feedback data stored in a database

[1063] Use it for your next recommendation

[1064] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[1065] The processing flow will be explained below.

[1066] Step 1: Accepting User Input

[1067] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[1068] example:

[1069] Type of drink: Wine

[1070] Favorite flavor: Fruity and sweet

[1071] Budget: Under 3,000 yen

[1072] Meal: Pasta

[1073] Emotional data: facial photos, audio recordings, text comments

[1074] Step 2: Send data

[1075] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[1076] Example of transmitted data:

[1077] {

[1078] "type": "wine",

[1079] "preferences": "Fruity and sweet",

[1080] "budget": 3000,

[1081] "meal": "pasta",

[1082] "emotion_data": {

[1083] "photo": "image data",

[1084] "voice": "voice data",

[1085] "text": "I'm tired today and would like a relaxing drink"

[1086] }

[1087] }

[1088] Step 3: Data Shaping

[1089] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[1090] Formatting example:

[1091] Check the format

[1092] String normalization

[1093] Category assignment

[1094] Step 4: Analyze the sentiment data

[1095] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[1096] Emotion analysis example:

[1097] Emotion: Tired

[1098] Step 5: Database Search

[1099] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[1100] SQL query example:

[1101] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[1102] Step 6: Evaluate and select candidates

[1103] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[1104] Example of evaluation process:

[1105] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[1106] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[1107] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[1108] Step 7: Submit your information

[1109] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[1110] Examples of information to send:

[1111] {

[1112] "name": "Japanese fruity wine",

[1113] "rating": 4.7,

[1114] "price": 2800,

[1115] "description": "A very fruity, sweet and relaxing wine."

[1116] }

[1117] Step 8: Viewing Information

[1118] Terminal: The received information is displayed to the user, who can then check the details of the selected beverage based on the displayed information.

[1119] Display example:

[1120] Drink Name: Japanese Fruity Wine

[1121] Rating: 4.7

[1122] Price: 2800 yen

[1123] Description: A very fruity, sweet and relaxing wine.

[1124] Step 9: Provide feedback

[1125] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[1126] Feedback input example:

[1127] Rating: 4 (out of 5)

[1128] Comment: "Very satisfied. It was relaxing just as described."

[1129] Step 10: Submit your feedback

[1130] Terminal: Sends the feedback information entered by the user to the server.

[1131] Example of submitted feedback:

[1132] {

[1133] "rating": 4,

[1134] "comment": "Very satisfied. It was as relaxing as described."

[1135] }

[1136] Step 11: Save your feedback

[1137] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[1138] Save example:

[1139] Feedback data stored in a database

[1140] Use it for your next recommendation

[1141] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[1142] Example 2

[1143] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1144] Conventional beverage recommendation systems recommend drinks based on a user's preferred taste, budget, and dietary preferences, but because they are unable to consider the user's emotional information, it is difficult to recommend the optimal drink for the user's psychological state. Furthermore, there is a lack of technology to effectively utilize feedback to improve recommendation accuracy.

[1145] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1146] In this invention, the server includes means for extracting emotion data using an emotion engine that analyzes user emotions, means for searching a database for beverage candidates that meet the criteria based on the formatted data and analyzed emotion data, and means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling sophisticated beverage recommendations based on the user's preferences and emotion information.

[1147] "User" means an individual or entity that intends to use the System to select a beverage.

[1148] "Beverage type" refers to the category of beverage selected, such as wine, beer, juice, etc.

[1149] "Preferred taste" refers to the flavor or characteristics of the beverage that the user prefers, such as fruity, sweet, dry, etc.

[1150] "Budget" refers to the monetary limit a User is willing to pay for a beverage.

[1151] "Meal content" refers to the type and menu of food that the user consumes along with the beverage they drink.

[1152] "Emotional Data" refers to data that indicates a user's emotional state analyzed from facial expressions, voice, and text.

[1153] "Terminal" means a device used by a User to access the System, including a smartphone, tablet, or PC.

[1154] "Server" refers to the central processing unit that processes, stores, and analyzes data submitted by users.

[1155] "Formatting" refers to the process of processing data received from a user and converting it into a usable format.

[1156] "Emotion engine" refers to software or hardware for analyzing a user's facial expressions, voice, and text to extract emotional data.

[1157] "Database" refers to a data management system that stores information about beverages and makes it accessible by query.

[1158] "Generative AI" refers to an artificial intelligence model that evaluates candidates based on past data and reviews and selects the most suitable beverage.

[1159] "Feedback" refers to the ratings and comments that users make about the beverages they have experienced.

[1160] The present invention is a system that recommends optimal drinks based on the user's input of the type of drink they want to drink, their preferred taste, budget, dietary content, and emotional data. Detailed explanations and concrete examples necessary for implementing this system are provided below.

[1161] Accepting user input

[1162] Users access the system from devices such as smartphones, tablets, and PCs and enter the type of beverage they want to drink, their preferred flavor, budget, dietary preferences, and emotional data. For example, users can enter the following data:

[1163] Example prompt sentence:

[1164] Type of drink: Wine

[1165] Favorite flavor: Fruity and sweet

[1166] Budget: Under 3,000 yen

[1167] Meal: Pasta

[1168] Emotional data: "I'm tired today and would like a drink to help me relax."

[1169] Data transmission and formatting

[1170] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input content and checks for errors. For example, the device generates the following JSON data and sends it to the server.

[1171] {

[1172] "type": "wine",

[1173] "preferences": "Fruity and sweet",

[1174] "budget": 3000,

[1175] "meal": "pasta",

[1176] "emotion_data": {

[1177] "photo": "image data",

[1178] "voice": "voice data",

[1179] "text": "I'm tired today and would like a relaxing drink"

[1180] }

[1181] }

[1182] Data Shaping and Sentiment Analysis

[1183] The server formats the data received from the device. This process includes format checks, string normalization, and category assignment. It also uses an emotion engine to analyze emotion data, extracting emotion data from the user's facial expressions, voice, and text. For example, it can detect fatigue levels from the text "I'm tired today."

[1184] Database search

[1185] The server queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that meet the criteria. SQL is used for query processing. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[1186] Candidate evaluation and selection

[1187] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the best wine from candidate wines A, B, and C.

[1188] Sending and Displaying Information

[1189] The server sends the information about the selected optimal drink to the user's terminal. The terminal displays the received information to the user. For example, the following information is displayed:

[1190] Display example:

[1191] Drink Name: Japanese Fruity Wine

[1192] Rating: 4.7

[1193] Price: 2800 yen

[1194] Description: A very fruity, sweet and relaxing wine.

[1195] Providing and storing feedback

[1196] After trying the beverage, the user provides feedback such as a rating and comments. The device sends the feedback information entered by the user to the server. The server receives the feedback and stores it in a database. This feedback will be used to improve the accuracy of the recommendation system in the future.

[1197] This allows users to easily select the perfect drink based on their preferences and emotions, and rate and provide feedback on their experience, allowing for a more personalized and refined selection of drinks.

[1198] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1199] Step 1:

[1200] Users access the system using devices such as smartphones, tablets, and PCs and input the type of beverage they want, their preferred flavor, budget, dietary preferences, and emotional data. This input includes a facial photo, audio recording, and text comments. An example of a prompt sentence is, "Please choose your favorite wine. I like fruity and sweet flavors, and I'd like something that goes well with pasta, within a budget of 3,000 yen or less. I'm tired after a long day today, so I'd like a drink that helps me relax."

[1201] Input: type of drink, preferred flavor, budget, meal contents, emotional data

[1202] Output: Information entered by the user

[1203] Step 2:

[1204] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input and performs error checks. For example, it checks whether all required fields are filled in. Data that passes the error check is composed as JSON and sent.

[1205] Input: User-entered data

[1206] Output: JSON format data

[1207] Step 3:

[1208] The server formats the JSON data received from the device. This includes formatting, string normalization, and category assignment. For example, it might verify that the type of drink entered is correct and correct any typos in the string. The formatted data is then converted into a standard format for internal processing.

[1209] Input: JSON format data

[1210] Output: Formatted user data

[1211] Step 4:

[1212] The server uses an emotion engine to analyze emotion data from the user's facial expressions, voice, and text. For example, the emotion engine detects the user's fatigue state from text such as "I'm tired." Based on the results of this analysis, the server quantifies or categorizes the user's emotional state.

[1213] Input: User's emotional data (face photo, voice, text)

[1214] Output: Analyzed emotion data (e.g., fatigue level)

[1215] Step 5:

[1216] The server queries the database based on the formatted beverage data and the analyzed emotion data to find beverage candidates that meet the criteria. This query uses SQL. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[1217] Input: formatted beverage data, parsed emotion data

[1218] Output: List of beverage candidates

[1219] Step 6:

[1220] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the candidate with the highest evaluation score.

[1221] Input: Beverage candidate list, review data, past evaluation data

[1222] Output: Best beverage candidates

[1223] Step 7:

[1224] The server then sends information about the selected optimal drink to the user's device. The information sent includes the name of the drink, its rating, price, and a detailed description. For example, "Japanese fruity wine, rating 4.7, price 2,800 yen, very fruity, sweet, and relaxing."

[1225] Input: Best beverage candidates

[1226] Output: detailed information about the beverage

[1227] Step 8:

[1228] The device then displays the received information to the user, allowing the user to check the details of the selected beverage based on the displayed information. For example, the name of the beverage, its rating, price, and detailed description are visually displayed.

[1229] Input: Beverage details

[1230] Output: Beverage information displayed to the user

[1231] Step 9:

[1232] After trying the beverage, users provide feedback by rating and commenting on it. They enter their rating in a special feedback form and press the "Submit" button. For example, a rating of 4 (out of 5) and a comment like "I was very satisfied. It was relaxing, just as described."

[1233] Input: Feedback after the experience

[1234] Output: Feedback information

[1235] Step 10:

[1236] The terminal sends the feedback information entered by the user to the server. The sending process includes converting the feedback data into JSON format and sending it.

[1237] Input: User feedback information

[1238] Output: Feedback data sent to the server

[1239] Step 11:

[1240] The server receives the feedback and stores it in a database, which can be used to improve the accuracy of the recommendation system in the future, for example, by improving the recommendation algorithm next time.

[1241] Input: Feedback data sent to the server

[1242] Output: Feedback stored in a database

[1243] (Application example 2)

[1244] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1245] In today's food and beverage selection process, it is difficult to make recommendations that accurately reflect a user's individual preferences and current emotional state. Food delivery services also require recommendation systems that can quickly respond to diverse user needs. While existing systems can recommend beverages based on a user's preferences and budget, they are unable to take into account the user's emotions. This makes it difficult for users to select a beverage that satisfies them. Furthermore, there has been a lack of systems that can improve the accuracy of recommendations based on feedback.

[1246] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1247] In this invention, the server includes means for formatting the type, preferences, budget, meal details, and emotion data entered by the user and performing the necessary preprocessing, means for analyzing emotions using an emotion engine based on the emotion data, and means for searching a database for beverage candidates that meet the conditions based on the analyzed emotion data, thereby making it possible to recommend optimal beverages that match the user's individual preferences and emotional state.

[1248] "User" refers to a person who uses the system to receive beverage recommendations.

[1249] "Beverage type" refers to the classification of different beverages, such as wine, beer, juice, etc.

[1250] "Preferred taste" refers to the taste characteristics of a beverage that a user prefers, such as fruity, sweet, or dry.

[1251] "Budget" refers to the maximum amount a user is willing to pay for a beverage.

[1252] "Meal Content" refers to the type and details of the meal the user plans to consume with the beverage.

[1253] "Emotional Data" refers to data about a user's current emotional state, including a photo of the user's face, audio recordings, text comments, etc.

[1254] "Formatting" refers to the process of converting user-entered data into a format suitable for analysis and search.

[1255] "Preprocessing" refers to preparing data to make it easier to analyze, such as normalizing the data or assigning categories.

[1256] An "emotion engine" refers to a system that analyzes a user's emotions from data such as facial photos, voice, and text.

[1257] A "database" refers to a collection of information that systematically stores information such as types of beverages, flavors, prices, and ratings.

[1258] "Candidates" refers to a list of beverages extracted based on the user's conditions and emotions.

[1259] "Generative AI" refers to an artificial intelligence algorithm that selects the optimal beverage based on given data.

[1260] "Terminal" refers to a device that a user uses to operate the system, such as a smartphone, tablet, or PC.

[1261] "Feedback" refers to ratings and comments provided by users based on their usage experience.

[1262] "Storing in a database" refers to storing feedback and other data in a database so that it can be reused for later retrieval and analysis.

[1263] The present invention relates to a system for recommending optimal beverages based on a user's individual preferences and emotional state, which is implemented using a set of hardware and software, such as:

[1264] System Overview

[1265] 1. Accepting user input

[1266] Users use devices such as smartphones, tablets, and computers to input the type of drink they want, their preferred flavor, budget, meal contents, and emotional data, which includes a photo of their face, audio recordings, and text comments.

[1267] 2. Data transmission and formatting

[1268] The device converts the user's input data and emotion data into JSON format and sends it to the server, which then performs formatting processes such as format checks and normalization on the received data.

[1269] 3. Emotion Data Analysis

[1270] The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format, and include emotional states such as "fatigue."

[1271] 4. Search for candidates

[1272] The server uses an SQL query to search the database for beverage candidates that meet the analyzed emotional data and the user's formatted conditions.

[1273] 5. Candidate evaluation and selection

[1274] The server uses generative AI (e.g., OpenAI's GPT model) to evaluate beverage candidates retrieved from the database and select the beverage that best suits the user's conditions and emotions.

[1275] 6. Transmission and Display of Information

[1276] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which then displays this information to the user.

[1277] 7. Providing and Saving Feedback

[1278] After tasting the beverage, the user sends their rating and comments to the server, which stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[1279] Specific examples

[1280] Prompt Sentence Examples

[1281] Choose the best beverage from the following options: Emotional information: Fatigue, Options: [Wine A (Rating: 4.5, Review: "Very fruity and sweet, perfect with pasta"), Wine B (Rating: 4.2, Review: "Well-balanced, with a touch of sweetness"), Wine C (Rating: 4.7, Review: "Relaxing taste")]

[1282] By inputting this prompt into a generative AI model, the optimal beverage can be selected based on the user's emotions and review information. This invention enables more personalized and precise beverage recommendations based on the user's emotional state. This allows users to easily select the beverage that best suits their specific emotional state, improving the quality of their dining experience.

[1283] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1284] Step 1:

[1285] Users use devices such as smartphones, tablets, and PCs to input the type of drink they want, their preferred flavor, budget, dietary preferences, and emotional data, including facial photos, audio recordings, and text comments. This data is collected through an input form on the device.

[1286] Input: User's drink type, preferred flavor, budget, dietary details, emotional data

[1287] Output: User input data in JSON format

[1288] Step 2:

[1289] The device converts the user's input data and emotion data into JSON format and sends it to the server, checking the data format and checking for errors before sending.

[1290] Input: User-entered data

[1291] Output: HTTP request with data in JSON format

[1292] Step 3:

[1293] The server performs format checks and normalization on the received data, including normalizing strings and assigning categories, and then sends the resulting data to a sentiment analysis engine.

[1294] Input: User-supplied data in JSON format

[1295] Output: Formatted data and data for sentiment analysis

[1296] Step 4:

[1297] The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format, and include emotional states such as "fatigue."

[1298] Input: Formatted data and sentiment data

[1299] Output: Parsed emotional state (e.g. "fatigue")

[1300] Step 5:

[1301] The server uses SQL queries to search for beverage candidates that match the analyzed emotion data and the user's preferences from a database containing information such as the type of beverage, flavor characteristics, price, and suitable meals.

[1302] Input: A search query for a beverage that matches your criteria (type, preference, budget, diet, emotional state)

[1303] Output: A list of beverage candidates as search results

[1304] Step 6:

[1305] The server evaluates and prioritizes potential drinks using a generative AI (e.g., OpenAI's GPT model). The model selects the best drink based on the user's sentiment and reviews based on the input prompt.

[1306] Input: Drink candidate list and emotional state

[1307] Output: Best drink information (name, rating, price, detailed description)

[1308] Step 7:

[1309] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which receives this information and displays it in a user-friendly format.

[1310] Input: Optimal beverage information

[1311] Output: Beverage information displayed on the terminal

[1312] Step 8:

[1313] After tasting the beverage, the user sends their rating and comments to the server from their device using a feedback form provided in the UI.

[1314] Input: User ratings and comments

[1315] Output: Feedback data in JSON format

[1316] Step 9:

[1317] The server verifies the received feedback and stores it in a database, which then serves as data to improve the accuracy of future recommendations.

[1318] Input: Feedback data in JSON format

[1319] Output: Ratings and comments stored in the database

[1320] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1321] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1322] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1323] [Third embodiment]

[1324] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1325] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1326] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1327] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1328] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1329] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1330] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1331] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1332] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1333] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1334] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1335] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1336] System Overview

[1337] This system recommends the perfect drink based on the type of drink the user wants, their preferred flavor, budget, and meal content. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers. The system accepts user input, searches a database based on that input, and uses generative AI to select the best drink candidates and provide information.

[1338] DETAILED DESCRIPTION OF THE EMBODIMENTS

[1339] 1. Accepting user input

[1340] User: Accesses the system and inputs the type of beverage they want to drink (e.g., wine, beer, sake, etc.), their preferred flavor (e.g., fruity, sweet, dry, etc.), their budget (e.g., under 3,000 yen), and their meal plan (e.g., pasta, sushi, etc.). This input is done via a device such as a smartphone, tablet, or PC.

[1341] Example user input:

[1342] Type of drink: Wine

[1343] Favorite flavor: Fruity and sweet

[1344] Budget: Under 3,000 yen

[1345] Meal: Pasta

[1346] 2. Data transmission and formatting

[1347] Terminal: Sends input data from the user to the server. The data is converted into an appropriate format (e.g., JSON) and sent to the server.

[1348] Server: Formats the received data, performs any necessary format checks, normalizes strings, assigns categories, and converts it into a form suitable for searching and evaluation.

[1349] 3. Database Search

[1350] Server: Based on the formatted data, the server searches the database (stored information about beverages and reviews) for beverage candidates that match the criteria. Methods such as SQL queries are used for the search process.

[1351] SQL query example:

[1352] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[1353] 4. Candidate evaluation and selection

[1354] Server: The search results are passed to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[1355] AI evaluation process example:

[1356] Candidates evaluated by AI:

[1357] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[1358] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[1359] 5. Providing information about alcohol

[1360] Server: Sends information about the drink that is determined to be the most suitable (such as name, rating, price, and detailed description) to the user's device.

[1361] Examples of information to send:

[1362] {

[1363] "name": "Japanese fruity wine",

[1364] "rating": 4.5,

[1365] "price": 2800,

[1366] "description": "A very fruity and sweet wine that pairs well with pasta."

[1367] }

[1368] Terminal: Displays the received information to the user, allowing the user to check the details of the recommended beverage.

[1369] 6. User Feedback

[1370] User: After experiencing the purchased beverage, provide a rating and comment. Enter your feedback in a dedicated form.

[1371] Feedback input example:

[1372] Rating: 4 (out of 5)

[1373] Comment: "Very satisfied. Just as described, perfect with pasta."

[1374] Terminal: Sends user feedback to the server.

[1375] Server: The received feedback is stored in a database and used to improve the accuracy of recommendations next time.

[1376] Specific examples

[1377] For example, if a user wants to drink a "fruity and sweet wine" with "pasta" for dinner for "less than 3,000 yen," the system operates as follows.

[1378] 1. User: Accesses the system app on a smartphone, selects a wine, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta."

[1379] 2. Terminal: Converts the input data into JSON format or similar and sends it to the server.

[1380] 3. Server: Formats the received data and performs any necessary preprocessing.

[1381] 4. Server: Searches the database for wines that meet the above criteria.

[1382] 5. Server: Passes the search results to the generation AI, which evaluates and selects the most suitable wine.

[1383] 6. Server: Sends information about the selected wine to the user's device so that the user can check the information.

[1384] 7. User: Purchases the recommended wine, drinks it, and then provides feedback with a rating and comments.

[1385] 8. Terminal: Sends the provided feedback to the server.

[1386] 9. Server: The feedback is stored in a database and used to improve future recommendations.

[1387] This allows users to easily select the beverage that best suits their tastes and also allows them to easily rate and provide feedback.

[1388] The processing flow will be explained below.

[1389] Step 1: Accepting User Input

[1390] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and meal details. This can be done using a device such as a smartphone, tablet, or PC.

[1391] example:

[1392] Type of drink: Wine

[1393] Favorite flavor: Fruity and sweet

[1394] Budget: Under 3,000 yen

[1395] Meal: Pasta

[1396] Step 2: Send data

[1397] Terminal: Converts user input data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input for errors.

[1398] Example of transmitted data:

[1399] {

[1400] "type": "wine",

[1401] "preferences": "Fruity and sweet",

[1402] "budget": 3000,

[1403] "meal": "pasta"

[1404] }

[1405] Step 3: Data Shaping

[1406] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment).

[1407] Formatting example:

[1408] Check the format

[1409] String normalization

[1410] Category assignment

[1411] Step 4: Database Search

[1412] Server: Queries the database based on the formatted data to find beverage candidates that match the criteria. SQL is used for query processing.

[1413] SQL query example:

[1414] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[1415] Step 5: Evaluate and select candidates

[1416] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best meets the user's criteria. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[1417] Example of evaluation process:

[1418] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[1419] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[1420] Step 6: Submit your information

[1421] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[1422] Examples of information to send:

[1423] {

[1424] "name": "Japanese fruity wine",

[1425] "rating": 4.5,

[1426] "price": 2800,

[1427] "description": "A very fruity and sweet wine that pairs well with pasta."

[1428] }

[1429] Step 7: Viewing information

[1430] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[1431] Display example:

[1432] Drink Name: Japanese Fruity Wine

[1433] Rating: 4.5

[1434] Price: 2800 yen

[1435] Description: A very fruity and sweet wine that pairs well with pasta.

[1436] Step 8: Provide feedback

[1437] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[1438] Feedback input example:

[1439] Rating: 4 (out of 5)

[1440] Comment: "Very satisfied. Just as described, perfect with pasta."

[1441] Step 9: Submit your feedback

[1442] Terminal: Sends the feedback information entered by the user to the server.

[1443] Example of submitted feedback:

[1444] {

[1445] "rating": 4,

[1446] "comment": "Very satisfied. Just as described, it was a perfect match for pasta."

[1447] }

[1448] Step 10: Save your feedback

[1449] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[1450] Save example:

[1451] Feedback data stored in a database

[1452] Use it for your next recommendation

[1453] Through these steps, users can easily choose the best beverage that suits their tastes and rate and provide feedback on their experience.

[1454] Example 1

[1455] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1456] Conventional beverage recommendation systems have difficulty in selecting the optimal beverage for a user's individual preferences, budget, and meal content, and it has been particularly difficult to achieve highly accurate recommendations comparable to those made by a top sommelier. Furthermore, they lacked a mechanism for utilizing user feedback in future recommendations. This resulted in a high risk of users choosing a beverage that did not suit their preferences, often resulting in dissatisfaction with the beverage selection.

[1457] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1458] In this invention, the server includes a means for the user to input the type of beverage they want to drink, their preferred flavor, their budget, and their meal details; a means for transmitting the user's input to the information processing device; and a means for the information processing device to format the user's input and perform any necessary preprocessing. This enables the information processing device to select the optimal beverage based on the user's input data using a highly accurate database search and a generative AI model. Furthermore, by utilizing user feedback to improve the accuracy of the next recommendation, it is possible to continuously provide highly satisfying selection results.

[1459] A "user" is a person who accesses the system to input the type of beverage they wish to drink, their preferred taste, budget, and dietary needs to receive recommendations.

[1460] The "information processing device" is a device that receives user input data, formats it and performs necessary preprocessing, and then evaluates and selects candidate beverages using a generative AI model.

[1461] The "storage device" is a device that stores information about beverages, word-of-mouth data, past evaluation data, etc., and stores a database that the information processing device uses for searches and evaluations.

[1462] A "generative AI model" is an artificial intelligence model that evaluates potential beverages based on past word-of-mouth and rating data, and selects the beverage that best suits the user's requirements.

[1463] A "terminal device" is a device that transmits user input to an information processing device and displays information about the selected drink to the user. Examples of such devices include smartphones, tablets, and personal computers.

[1464] "Feedback" refers to the ratings and comments provided by users after trying a recommended beverage, and is information used to improve the accuracy of recommendations next time.

[1465] The following describes in detail an embodiment of the present invention, specifically showing what hardware and software are used to process and calculate data.

[1466] 1. Accepting user input

[1467] User:

[1468] First, users access the system using a smartphone, tablet, or PC. Through the system's interface (web application or native application), they input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs. This allows users to easily select a beverage that suits their preferences.

[1469] Example of input contents:

[1470] Type of drink: Wine

[1471] Favorite flavor: Fruity and sweet

[1472] Budget: Under 3,000 yen

[1473] Meal: Pasta

[1474] 2. Data transmission and formatting

[1475] Device:

[1476] The data entered by the user is first converted into an appropriate format (e.g., JSON) and then sent to the information processing device, so that the input data is delivered to the server in a unified format.

[1477] server:

[1478] The information processing device (server) formats the received data, performs necessary format checks and normalizes character strings, and corrects any incomplete or invalid data. For example, it normalizes a taste preference such as "fruity and sweet" and performs preprocessing to facilitate database searches.

[1479] 3. Database Search

[1480] server:

[1481] Based on the formatted data, a search is performed on the storage device (database) to find beverage candidates that match the criteria. To do this, a search algorithm such as an SQL query is used. The database stores detailed information about the beverages, word-of-mouth data, and past evaluation data.

[1482] Example search queries:

[1483] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[1484] 4. Candidate evaluation and selection

[1485] server:

[1486] The candidate list obtained from the search process is passed to a generative AI model for evaluation. The generative AI model takes into account past reviews and evaluation data to select the drink that best suits the user's criteria. This enables highly accurate recommendations comparable to those made by a top sommelier.

[1487] Examples of AI evaluation processes:

[1488] candidate:

[1489] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[1490] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[1491] AI Selection: Wine A

[1492] 5. Providing information about alcohol

[1493] server:

[1494] Information about the drink that is judged to be the most suitable, such as its name, rating, price, and detailed description, is sent to the user's device. The information sent is standardized in JSON format.

[1495] Examples of information to send:

[1496] {

[1497] "name": "Japanese fruity wine",

[1498] "rating": 4.5,

[1499] "price": 2800,

[1500] "description": "A very fruity and sweet wine that pairs well with pasta."

[1501] }

[1502] Device:

[1503] The received information is displayed to the user, who can then view the details of the recommended beverage.

[1504] 6. User Feedback

[1505] User:

[1506] After tasting the purchased beverage, users can enter their rating and comments in a dedicated feedback form. This feedback will be important data that will help improve the accuracy of the system.

[1507] Examples of feedback include:

[1508] Rating: 4 (out of 5)

[1509] Comment: "Very satisfied. Just as described, perfect with pasta."

[1510] Device:

[1511] The input feedback data is transmitted again to the information processing device.

[1512] server:

[1513] The received feedback is stored in a storage device and used to improve the accuracy of the next beverage recommendation.

[1514] As a result, the system of the present invention allows users to enjoy the perfect beverage that suits their tastes. Furthermore, by continuously collecting feedback, the system's recommendation accuracy continues to improve.

[1515] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1516] Step 1:

[1517] Users access the system using a smartphone, tablet, or computer, and input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs through the interface.

[1518] Input: type of drink, preferred flavor, budget, meal contents

[1519] What happens: A user enters data into a web form or in-app input field and presses the submit button.

[1520] Step 2:

[1521] Terminal: Converts the data entered by the user into JSON format or similar and sends it to the information processing device (server).

[1522] Input: User-entered data

[1523] Output: Formatted data such as JSON

[1524] Specific operation: The data obtained from the input field is encoded in JSON format and sent to the server as an HTTP request.

[1525] Step 3:

[1526] Server: Formats the received data, performs any necessary format checks and string normalization, and checks for incomplete or invalid data.

[1527] Input: JSON format data sent from the terminal

[1528] Output: Normalized and formatted data

[1529] Specific operation: For example, the user input "fruity and sweet" is broken down into individual conditions such as "fruity" and "sweet."

[1530] Step 4:

[1531] Server: Searches for beverage candidates that match the criteria from the storage device (database) based on the normalized data.

[1532] Input: Normalized user-entered data

[1533] Output: A list of drinks that match the criteria

[1534] Specific operation: Using SQL queries against the database, candidates are extracted based on criteria such as "wine," "fruity," "sweet," and "under 3,000 yen."

[1535] Step 5:

[1536] Server: The candidate list obtained as a search result is passed to the generative AI model, which evaluates it.

[1537] Input: A list of beverage candidates extracted from the database

[1538] Output: Evaluation by the generative AI model and selection of the optimal beverage

[1539] Specific operation: Using an evaluation algorithm, each candidate is scored, taking into account word-of-mouth data and evaluation data.

[1540] Step 6:

[1541] Server: Sends information about the beverage that has been determined to be optimal to the user's terminal device.

[1542] Input: Selection results by generative AI model

[1543] Output: Detailed drink information sent to user device

[1544] Specific operation: Information including the name, rating, price, and detailed description of the selected drink is encoded in JSON format and sent to the user's device as an HTTP response.

[1545] Step 7:

[1546] Terminal: Displays the received information to the user, who can then check the details of the recommended beverage.

[1547] Input: detailed drink information sent from the server

[1548] Output: Detailed drink information displayed to the user

[1549] Specific behavior: Display detailed information about the drink on a webpage or app screen for the user to review.

[1550] Step 8:

[1551] User: Tastes the purchased beverage and enters a rating and comments in the feedback form.

[1552] Input: User ratings and comments

[1553] Output: Feedback data

[1554] Specific actions: Enter your ratings and comments in the feedback form and press the submit button.

[1555] Step 9:

[1556] Terminal: Transmits the input feedback data to the information processing device.

[1557] Input: User feedback data

[1558] Output: Feedback data sent to the server

[1559] Specific operation: The feedback data is encoded in JSON format and sent to the server as an HTTP request.

[1560] Step 10:

[1561] Server: The received feedback is stored in a storage device and used to improve the accuracy of recommendations next time.

[1562] Input: Feedback data sent from the device

[1563] Output: Feedback data stored in memory device

[1564] Specific operation: The feedback data is inserted into a database and used as training data for future algorithms.

[1565] (Application example 1)

[1566] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1567] Conventional beverage recommendation systems struggle to easily and effectively find beverages that match users' preferences. They also lack a visual and interactive experience, making it difficult to improve user satisfaction. Furthermore, they lack a means to utilize feedback to improve recommendation accuracy.

[1568] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1569] In this invention, the server includes a means for visually inputting information from the user using a VR head-mounted display, a means for searching for beverage candidates that meet the criteria from a database based on the formatted data, and a means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling the user to easily and effectively find a beverage that suits their preferences through a visual and interactive experience.

[1570] "User" means a consumer who uses the system to receive beverage recommendations.

[1571] "Type of beverage" refers to the category of beverage consumed, such as wine, beer, or sake.

[1572] "Preferred taste" refers to the flavor or characteristics of a beverage desired by a user, including taste attributes such as fruity, sweet, dry, etc.

[1573] "Budget" indicates the maximum amount the user is willing to spend, and the price of the selected drink must not exceed this amount.

[1574] "Meal content" refers to the type of meal the user consumes with the beverage, including specific dish names such as pasta or sushi.

[1575] A "VR head-mounted display" is a display device that allows users to visually experience a virtual reality environment by wearing it.

[1576] The "server" is a computer system that receives user input, formats the data, and uses database search and generative AI to recommend beverages.

[1577] "Formatting" is the process of converting received user input data into an appropriate format, making it suitable for searching and evaluation.

[1578] A "database" is a collection of information that stores information about beverages and word-of-mouth data, and is a target that can be searched based on user criteria.

[1579] "Generative AI" is a type of artificial intelligence that has the ability to evaluate and select the most suitable beverage based on input data.

[1580] "Feedback" refers to the ratings and comments provided by users after their experience, which are used to improve the system's recommendation accuracy.

[1581] "VR space" is a virtual reality world that users experience through a VR head-mounted display.

[1582] System Overview

[1583] In this embodiment, a system is provided in which a user wears a VR head-mounted display and inputs the type of beverage they want to drink, their preferred flavor, their budget, and their meal plan in a virtual space, and the system recommends the most suitable beverage. The system sends the user's input to a server, which performs a database search and evaluation using a generation AI, selects the most suitable beverage, and displays the result in the VR space. The user then provides feedback, which the system stores in a database.

[1584] 1. Accepting user input

[1585] Users put on a VR head-mounted display and log in to the virtual space, where they can use the interface to input the type of drink, their preferred flavor, budget, and meal details.

[1586] 2. Data transmission and formatting

[1587] The device converts the user's input data into JSON format and sends it to the server, which then formats the received data and converts it into the appropriate format.

[1588] 3. Database Search

[1589] The server then searches the database for beverage candidates that match the criteria based on the formatted data, using SQL queries.

[1590] 4. Candidate evaluation and selection

[1591] The server passes the search results to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[1592] 5. Providing information about beverages

[1593] The server sends information about the best drink to the user's device, which then displays the information visually in the VR space.

[1594] 6. User Feedback

[1595] After trying the beverage, users enter their ratings and comments in a dedicated form in the virtual space, and the device sends the feedback to the server.

[1596] 7. Storage and Use of Feedback

[1597] The server stores the received feedback in a database and uses it to improve the accuracy of recommendations next time.

[1598] Hardware and software used

[1599] This system is implemented using the following hardware and software:

[1600] VR head-mounted display: Oculus Quest 2, etc.

[1601] Frontend: VR application

[1602] Server: Flask (Python web framework), SQLite (database)

[1603] Generation AI: OpenAI GPT-3.5 API

[1604] Specific examples

[1605] If a user wants to pair a "fruity and sweet wine" with pasta for dinner and under 3,000 yen, the system works as follows: The user puts on the Oculus Quest 2, selects a wine in the VR space, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta." The device converts the input data into JSON format and sends it to the server. The server formats the received data and performs any necessary preprocessing. The server searches its database for wines that meet the criteria. The server passes the search results to the generation AI, which evaluates and selects the most suitable wine. The server then sends information about the selected wine to the user's device, where the user can view it in the VR space. The user purchases the recommended wine and, after tasting it, provides feedback with a rating and comments. The device then sends the provided feedback to the server. The server stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[1606] Prompt Sentence Examples

[1607] In a beverage recommendation system, please recommend the best wine for the user based on the following criteria:

[1608] Type of drink: Wine

[1609] Favorite flavor: Fruity and sweet

[1610] Budget: Under 3,000 yen

[1611] Meal: Pasta

[1612] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1613] Step 1:

[1614] The user wears a VR head-mounted display and accesses the virtual space interface. The user inputs the type of drink, preferred flavor, budget, and meal contents. This input information is received by the terminal via a form in the VR space.

[1615] Step 2:

[1616] The device converts the user's input data into JSON format and sends the input information to the server. The input information includes "type of drink," "preferred flavor," "budget," and "meal details." The input data is formatted in JSON format and sent to the server as an HTTP request.

[1617] Step 3:

[1618] The server then formats the received JSON data, converting it into a format suitable for database queries. This formatting process includes format checks and string normalization for the input data. The formatted data is then in a format that can be applied to database queries.

[1619] Step 4:

[1620] The server uses the formatted data to search the database for drink candidates that match the criteria, using SQL queries to extract drinks that fit the drink type, flavor requirements, and budget. This search result may contain multiple candidates.

[1621] Step 5:

[1622] The server passes the search results to a generative AI model, which evaluates and selects the beverage that best suits the user's criteria. The generative AI model selects the optimal beverage based on the evaluation and word-of-mouth data for each beverage included in the search results.

[1623] Step 6:

[1624] The server converts the information about the optimal drink selected by the generative AI model into JSON format and sends it to the device. The transmitted data includes the drink name, rating, price, and description. The device receives this information and interprets the JSON data.

[1625] Step 7:

[1626] The device visually displays the received beverage information in the VR space. Detailed information about the selected beverage is visually presented to the user in the virtual space interface. The user can review this information and make a purchase decision.

[1627] Step 8:

[1628] After purchasing and experiencing the beverage, users can provide their rating and feedback through a virtual interface, which is entered into a dedicated form.

[1629] Step 9:

[1630] The device converts the feedback provided by the user into JSON format and sends it to the server. The feedback data includes rating values ​​and comments. The server interprets the received feedback data and stores it in a database.

[1631] Step 10:

[1632] The server stores the feedback information in a database and refers to it the next time the server recommends a beverage. The stored feedback is used to improve the accuracy of future recommendations.

[1633] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1634] System Overview

[1635] This system recommends the best drink based on the type of drink a user wants, their preferred flavor, their budget, and their meal. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system can select drinks that take into account the user's emotional information. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers, providing more personalized and precise recommendations.

[1636] DETAILED DESCRIPTION OF THE EMBODIMENTS

[1637] 1. Accepting user input

[1638] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[1639] Example user input:

[1640] Type of drink: Wine

[1641] Favorite flavor: Fruity and sweet

[1642] Budget: Under 3,000 yen

[1643] Meal: Pasta

[1644] Emotional data: facial photos, audio recordings, text comments

[1645] 2. Data transmission and formatting

[1646] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[1647] Example of transmitted data:

[1648] {

[1649] "type": "wine",

[1650] "preferences": "Fruity and sweet",

[1651] "budget": 3000,

[1652] "meal": "pasta",

[1653] "emotion_data": {

[1654] "photo": "image data",

[1655] "voice": "voice data",

[1656] "text": "I'm tired today and would like a relaxing drink"

[1657] }

[1658] }

[1659] 3. Data Formatting

[1660] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[1661] Formatting example:

[1662] Check the format

[1663] String normalization

[1664] Category assignment

[1665] 4. Emotion Data Analysis

[1666] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[1667] Emotion analysis example:

[1668] Emotion: Tired

[1669] 5. Database Search

[1670] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[1671] SQL query example:

[1672] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[1673] 6. Evaluation and selection of candidates

[1674] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[1675] Example of evaluation process:

[1676] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[1677] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[1678] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[1679] 7. Transmission of Information

[1680] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[1681] Examples of information to send:

[1682] {

[1683] "name": "Japanese fruity wine",

[1684] "rating": 4.7,

[1685] "price": 2800,

[1686] "description": "A very fruity, sweet and relaxing wine."

[1687] }

[1688] 8. Display of Information

[1689] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[1690] Display example:

[1691] Drink Name: Japanese Fruity Wine

[1692] Rating: 4.7

[1693] Price: 2800 yen

[1694] Description: A very fruity, sweet and relaxing wine.

[1695] 9. Providing Feedback

[1696] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[1697] Feedback input example:

[1698] Rating: 4 (out of 5)

[1699] Comment: "Very satisfied. It was relaxing just as described."

[1700] 10. Submitting Feedback

[1701] Terminal: Sends the feedback information entered by the user to the server.

[1702] Example of submitted feedback:

[1703] {

[1704] "rating": 4,

[1705] "comment": "Very satisfied. It was as relaxing as described."

[1706] }

[1707] 11. Saving Feedback

[1708] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[1709] Save example:

[1710] Feedback data stored in a database

[1711] Use it for your next recommendation

[1712] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[1713] The processing flow will be explained below.

[1714] Step 1: Accepting User Input

[1715] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[1716] example:

[1717] Type of drink: Wine

[1718] Favorite flavor: Fruity and sweet

[1719] Budget: Under 3,000 yen

[1720] Meal: Pasta

[1721] Emotional data: facial photos, audio recordings, text comments

[1722] Step 2: Send data

[1723] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[1724] Example of transmitted data:

[1725] {

[1726] "type": "wine",

[1727] "preferences": "Fruity and sweet",

[1728] "budget": 3000,

[1729] "meal": "pasta",

[1730] "emotion_data": {

[1731] "photo": "image data",

[1732] "voice": "voice data",

[1733] "text": "I'm tired today and would like a relaxing drink"

[1734] }

[1735] }

[1736] Step 3: Data Shaping

[1737] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[1738] Formatting example:

[1739] Check the format

[1740] String normalization

[1741] Category assignment

[1742] Step 4: Analyze the sentiment data

[1743] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[1744] Emotion analysis example:

[1745] Emotion: Tired

[1746] Step 5: Database Search

[1747] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[1748] SQL query example:

[1749] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[1750] Step 6: Evaluate and select candidates

[1751] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[1752] Example of evaluation process:

[1753] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[1754] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[1755] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[1756] Step 7: Submit your information

[1757] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[1758] Examples of information to send:

[1759] {

[1760] "name": "Japanese fruity wine",

[1761] "rating": 4.7,

[1762] "price": 2800,

[1763] "description": "A very fruity, sweet and relaxing wine."

[1764] }

[1765] Step 8: Viewing Information

[1766] Terminal: The received information is displayed to the user, who can then check the details of the selected beverage based on the displayed information.

[1767] Display example:

[1768] Drink Name: Japanese Fruity Wine

[1769] Rating: 4.7

[1770] Price: 2800 yen

[1771] Description: A very fruity, sweet and relaxing wine.

[1772] Step 9: Provide feedback

[1773] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[1774] Feedback input example:

[1775] Rating: 4 (out of 5)

[1776] Comment: "Very satisfied. It was relaxing just as described."

[1777] Step 10: Submit your feedback

[1778] Terminal: Sends the feedback information entered by the user to the server.

[1779] Example of submitted feedback:

[1780] {

[1781] "rating": 4,

[1782] "comment": "Very satisfied. It was as relaxing as described."

[1783] }

[1784] Step 11: Save your feedback

[1785] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[1786] Save example:

[1787] Feedback data stored in a database

[1788] Use it for your next recommendation

[1789] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[1790] Example 2

[1791] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1792] Conventional beverage recommendation systems recommend drinks based on a user's preferred taste, budget, and dietary preferences, but because they are unable to consider the user's emotional information, it is difficult to recommend the optimal drink for the user's psychological state. Furthermore, there is a lack of technology to effectively utilize feedback to improve recommendation accuracy.

[1793] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1794] In this invention, the server includes means for extracting emotion data using an emotion engine that analyzes user emotions, means for searching a database for beverage candidates that meet the criteria based on the formatted data and analyzed emotion data, and means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling sophisticated beverage recommendations based on the user's preferences and emotion information.

[1795] "User" means an individual or entity that intends to use the System to select a beverage.

[1796] "Beverage type" refers to the category of beverage selected, such as wine, beer, juice, etc.

[1797] "Preferred taste" refers to the flavor or characteristics of the beverage that the user prefers, such as fruity, sweet, dry, etc.

[1798] "Budget" refers to the monetary limit a User is willing to pay for a beverage.

[1799] "Meal content" refers to the type and menu of food that the user consumes along with the beverage they drink.

[1800] "Emotional Data" refers to data that indicates a user's emotional state analyzed from facial expressions, voice, and text.

[1801] "Terminal" means a device used by a User to access the System, including a smartphone, tablet, or PC.

[1802] "Server" refers to the central processing unit that processes, stores, and analyzes data submitted by users.

[1803] "Formatting" refers to the process of processing data received from a user and converting it into a usable format.

[1804] "Emotion engine" refers to software or hardware for analyzing a user's facial expressions, voice, and text to extract emotional data.

[1805] "Database" refers to a data management system that stores information about beverages and makes it accessible by query.

[1806] "Generative AI" refers to an artificial intelligence model that evaluates candidates based on past data and reviews and selects the most suitable beverage.

[1807] "Feedback" refers to the ratings and comments that users make about the beverages they have experienced.

[1808] The present invention is a system that recommends optimal drinks based on the user's input of the type of drink they want to drink, their preferred taste, budget, dietary content, and emotional data. Detailed explanations and concrete examples necessary for implementing this system are provided below.

[1809] Accepting user input

[1810] Users access the system from devices such as smartphones, tablets, and PCs and enter the type of beverage they want to drink, their preferred flavor, budget, dietary preferences, and emotional data. For example, users can enter the following data:

[1811] Example prompt sentence:

[1812] Type of drink: Wine

[1813] Favorite flavor: Fruity and sweet

[1814] Budget: Under 3,000 yen

[1815] Meal: Pasta

[1816] Emotional data: "I'm tired today and would like a drink to help me relax."

[1817] Data transmission and formatting

[1818] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input content and checks for errors. For example, the device generates the following JSON data and sends it to the server.

[1819] {

[1820] "type": "wine",

[1821] "preferences": "Fruity and sweet",

[1822] "budget": 3000,

[1823] "meal": "pasta",

[1824] "emotion_data": {

[1825] "photo": "image data",

[1826] "voice": "voice data",

[1827] "text": "I'm tired today and would like a relaxing drink"

[1828] }

[1829] }

[1830] Data Shaping and Sentiment Analysis

[1831] The server formats the data received from the device. This process includes format checks, string normalization, and category assignment. It also uses an emotion engine to analyze emotion data, extracting emotion data from the user's facial expressions, voice, and text. For example, it can detect fatigue levels from the text "I'm tired today."

[1832] Database search

[1833] The server queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that meet the criteria. SQL is used for query processing. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[1834] Candidate evaluation and selection

[1835] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the best wine from candidate wines A, B, and C.

[1836] Sending and Displaying Information

[1837] The server sends the information about the selected optimal drink to the user's terminal. The terminal displays the received information to the user. For example, the following information is displayed:

[1838] Display example:

[1839] Drink Name: Japanese Fruity Wine

[1840] Rating: 4.7

[1841] Price: 2800 yen

[1842] Description: A very fruity, sweet and relaxing wine.

[1843] Providing and storing feedback

[1844] After trying the beverage, the user provides feedback such as a rating and comments. The device sends the feedback information entered by the user to the server. The server receives the feedback and stores it in a database. This feedback will be used to improve the accuracy of the recommendation system in the future.

[1845] This allows users to easily select the perfect drink based on their preferences and emotions, and rate and provide feedback on their experience, allowing for a more personalized and refined selection of drinks.

[1846] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1847] Step 1:

[1848] Users access the system using devices such as smartphones, tablets, and PCs and input the type of beverage they want, their preferred flavor, budget, dietary preferences, and emotional data. This input includes a facial photo, audio recording, and text comments. An example of a prompt sentence is, "Please choose your favorite wine. I like fruity and sweet flavors, and I'd like something that goes well with pasta, within a budget of 3,000 yen or less. I'm tired after a long day today, so I'd like a drink that helps me relax."

[1849] Input: type of drink, preferred flavor, budget, meal contents, emotional data

[1850] Output: Information entered by the user

[1851] Step 2:

[1852] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input and performs error checks. For example, it checks whether all required fields are filled in. Data that passes the error check is composed as JSON and sent.

[1853] Input: User-entered data

[1854] Output: JSON format data

[1855] Step 3:

[1856] The server formats the JSON data received from the device. This includes formatting, string normalization, and category assignment. For example, it might verify that the type of drink entered is correct and correct any typos in the string. The formatted data is then converted into a standard format for internal processing.

[1857] Input: JSON format data

[1858] Output: Formatted user data

[1859] Step 4:

[1860] The server uses an emotion engine to analyze emotion data from the user's facial expressions, voice, and text. For example, the emotion engine detects the user's fatigue state from text such as "I'm tired." Based on the results of this analysis, the server quantifies or categorizes the user's emotional state.

[1861] Input: User's emotional data (face photo, voice, text)

[1862] Output: Analyzed emotion data (e.g., fatigue level)

[1863] Step 5:

[1864] The server queries the database based on the formatted beverage data and the analyzed emotion data to find beverage candidates that meet the criteria. This query uses SQL. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[1865] Input: formatted beverage data, parsed emotion data

[1866] Output: List of beverage candidates

[1867] Step 6:

[1868] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the candidate with the highest evaluation score.

[1869] Input: Beverage candidate list, review data, past evaluation data

[1870] Output: Best beverage candidates

[1871] Step 7:

[1872] The server then sends information about the selected optimal drink to the user's device. The information sent includes the name of the drink, its rating, price, and a detailed description. For example, "Japanese fruity wine, rating 4.7, price 2,800 yen, very fruity, sweet, and relaxing."

[1873] Input: Best beverage candidates

[1874] Output: detailed information about the beverage

[1875] Step 8:

[1876] The device then displays the received information to the user, allowing the user to check the details of the selected beverage based on the displayed information. For example, the name of the beverage, its rating, price, and detailed description are visually displayed.

[1877] Input: Beverage details

[1878] Output: Beverage information displayed to the user

[1879] Step 9:

[1880] After trying the beverage, users provide feedback by rating and commenting on it. They enter their rating in a special feedback form and press the "Submit" button. For example, a rating of 4 (out of 5) and a comment like "I was very satisfied. It was relaxing, just as described."

[1881] Input: Feedback after the experience

[1882] Output: Feedback information

[1883] Step 10:

[1884] The terminal sends the feedback information entered by the user to the server. The sending process includes converting the feedback data into JSON format and sending it.

[1885] Input: User feedback information

[1886] Output: Feedback data sent to the server

[1887] Step 11:

[1888] The server receives the feedback and stores it in a database, which can be used to improve the accuracy of the recommendation system in the future, for example, by improving the recommendation algorithm next time.

[1889] Input: Feedback data sent to the server

[1890] Output: Feedback stored in a database

[1891] (Application example 2)

[1892] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1893] In today's food and beverage selection process, it is difficult to make recommendations that accurately reflect a user's individual preferences and current emotional state. Food delivery services also require recommendation systems that can quickly respond to diverse user needs. While existing systems can recommend beverages based on a user's preferences and budget, they are unable to take into account the user's emotions. This makes it difficult for users to select a beverage that satisfies them. Furthermore, there has been a lack of systems that can improve the accuracy of recommendations based on feedback.

[1894] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1895] In this invention, the server includes means for formatting the type, preferences, budget, meal details, and emotion data entered by the user and performing the necessary preprocessing, means for analyzing emotions using an emotion engine based on the emotion data, and means for searching a database for beverage candidates that meet the conditions based on the analyzed emotion data, thereby making it possible to recommend optimal beverages that match the user's individual preferences and emotional state.

[1896] "User" refers to a person who uses the system to receive beverage recommendations.

[1897] "Beverage type" refers to the classification of different beverages, such as wine, beer, juice, etc.

[1898] "Preferred taste" refers to the taste characteristics of a beverage that a user prefers, such as fruity, sweet, or dry.

[1899] "Budget" refers to the maximum amount a user is willing to pay for a beverage.

[1900] "Meal Content" refers to the type and details of the meal the user plans to consume with the beverage.

[1901] "Emotional Data" refers to data about a user's current emotional state, including a photo of the user's face, audio recordings, text comments, etc.

[1902] "Formatting" refers to the process of converting user-entered data into a format suitable for analysis and search.

[1903] "Preprocessing" refers to preparing data to make it easier to analyze, such as normalizing the data or assigning categories.

[1904] An "emotion engine" refers to a system that analyzes a user's emotions from data such as facial photos, voice, and text.

[1905] A "database" refers to a collection of information that systematically stores information such as types of beverages, flavors, prices, and ratings.

[1906] "Candidates" refers to a list of beverages extracted based on the user's conditions and emotions.

[1907] "Generative AI" refers to an artificial intelligence algorithm that selects the optimal beverage based on given data.

[1908] "Terminal" refers to a device that a user uses to operate the system, such as a smartphone, tablet, or PC.

[1909] "Feedback" refers to ratings and comments provided by users based on their usage experience.

[1910] "Storing in a database" refers to storing feedback and other data in a database so that it can be reused for later retrieval and analysis.

[1911] The present invention relates to a system for recommending optimal beverages based on a user's individual preferences and emotional state, which is implemented using a set of hardware and software, such as:

[1912] System Overview

[1913] 1. Accepting user input

[1914] Users use devices such as smartphones, tablets, and computers to input the type of drink they want, their preferred flavor, budget, meal contents, and emotional data, which includes a photo of their face, audio recordings, and text comments.

[1915] 2. Data transmission and formatting

[1916] The device converts the user's input data and emotion data into JSON format and sends it to the server, which then performs formatting processes such as format checks and normalization on the received data.

[1917] 3. Emotion Data Analysis

[1918] The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format, and include emotional states such as "fatigue."

[1919] 4. Search for candidates

[1920] The server uses an SQL query to search the database for beverage candidates that meet the analyzed emotional data and the user's formatted conditions.

[1921] 5. Candidate evaluation and selection

[1922] The server uses generative AI (e.g., OpenAI's GPT model) to evaluate beverage candidates retrieved from the database and select the beverage that best suits the user's conditions and emotions.

[1923] 6. Transmission and Display of Information

[1924] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which then displays this information to the user.

[1925] 7. Providing and Saving Feedback

[1926] After tasting the beverage, the user sends their rating and comments to the server, which stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[1927] Specific examples

[1928] Prompt Sentence Examples

[1929] Choose the best beverage from the following options: Emotional information: Fatigue, Options: [Wine A (Rating: 4.5, Review: "Very fruity and sweet, perfect with pasta"), Wine B (Rating: 4.2, Review: "Well-balanced, with a touch of sweetness"), Wine C (Rating: 4.7, Review: "Relaxing taste")]

[1930] By inputting this prompt into a generative AI model, the optimal beverage can be selected based on the user's emotions and review information. This invention enables more personalized and precise beverage recommendations based on the user's emotional state. This allows users to easily select the beverage that best suits their specific emotional state, improving the quality of their dining experience.

[1931] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1932] Step 1:

[1933] Users use devices such as smartphones, tablets, and PCs to input the type of drink they want, their preferred flavor, budget, dietary preferences, and emotional data, including facial photos, audio recordings, and text comments. This data is collected through an input form on the device.

[1934] Input: User's drink type, preferred flavor, budget, dietary details, emotional data

[1935] Output: User input data in JSON format

[1936] Step 2:

[1937] The device converts the user's input data and emotion data into JSON format and sends it to the server, checking the data format and checking for errors before sending.

[1938] Input: User-entered data

[1939] Output: HTTP request with data in JSON format

[1940] Step 3:

[1941] The server performs format checks and normalization on the received data, including normalizing strings and assigning categories, and then sends the resulting data to a sentiment analysis engine.

[1942] Input: User-supplied data in JSON format

[1943] Output: Formatted data and data for sentiment analysis

[1944] Step 4:

[1945] The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format, and include emotional states such as "fatigue."

[1946] Input: Formatted data and sentiment data

[1947] Output: Parsed emotional state (e.g. "fatigue")

[1948] Step 5:

[1949] The server uses SQL queries to search for beverage candidates that match the analyzed emotion data and the user's preferences from a database containing information such as the type of beverage, flavor characteristics, price, and suitable meals.

[1950] Input: A search query for a beverage that matches your criteria (type, preference, budget, diet, emotional state)

[1951] Output: A list of beverage candidates as search results

[1952] Step 6:

[1953] The server evaluates and prioritizes potential drinks using a generative AI (e.g., OpenAI's GPT model). The model selects the best drink based on the user's sentiment and reviews based on the input prompt.

[1954] Input: Drink candidate list and emotional state

[1955] Output: Best drink information (name, rating, price, detailed description)

[1956] Step 7:

[1957] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which receives this information and displays it in a user-friendly format.

[1958] Input: Optimal beverage information

[1959] Output: Beverage information displayed on the terminal

[1960] Step 8:

[1961] After tasting the beverage, the user sends their rating and comments to the server from their device using a feedback form provided in the UI.

[1962] Input: User ratings and comments

[1963] Output: Feedback data in JSON format

[1964] Step 9:

[1965] The server verifies the received feedback and stores it in a database, which then serves as data to improve the accuracy of future recommendations.

[1966] Input: Feedback data in JSON format

[1967] Output: Ratings and comments stored in the database

[1968] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1969] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1970] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1971] [Fourth embodiment]

[1972] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1973] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1974] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1975] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1976] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1977] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1978] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1979] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1980] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1981] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1982] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1983] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1984] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1985] System Overview

[1986] This system recommends the perfect drink based on the type of drink the user wants, their preferred flavor, budget, and meal content. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers. The system accepts user input, searches a database based on that input, and uses generative AI to select the best drink candidates and provide information.

[1987] DETAILED DESCRIPTION OF THE EMBODIMENTS

[1988] 1. Accepting user input

[1989] User: Accesses the system and inputs the type of beverage they want to drink (e.g., wine, beer, sake, etc.), their preferred flavor (e.g., fruity, sweet, dry, etc.), their budget (e.g., under 3,000 yen), and their meal plan (e.g., pasta, sushi, etc.). This input is done via a device such as a smartphone, tablet, or PC.

[1990] Example user input:

[1991] Type of drink: Wine

[1992] Favorite flavor: Fruity and sweet

[1993] Budget: Under 3,000 yen

[1994] Meal: Pasta

[1995] 2. Data transmission and formatting

[1996] Terminal: Sends input data from the user to the server. The data is converted into an appropriate format (e.g., JSON) and sent to the server.

[1997] Server: Formats the received data, performs any necessary format checks, normalizes strings, assigns categories, and converts it into a form suitable for searching and evaluation.

[1998] 3. Database Search

[1999] Server: Based on the formatted data, the server searches the database (stored information about beverages and reviews) for beverage candidates that match the criteria. Methods such as SQL queries are used for the search process.

[2000] SQL query example:

[2001] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[2002] 4. Candidate evaluation and selection

[2003] Server: The search results are passed to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[2004] AI evaluation process example:

[2005] Candidates evaluated by AI:

[2006] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[2007] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[2008] 5. Providing information about alcohol

[2009] Server: Sends information about the drink that is determined to be the most suitable (such as name, rating, price, and detailed description) to the user's device.

[2010] Examples of information to send:

[2011] {

[2012] "name": "Japanese fruity wine",

[2013] "rating": 4.5,

[2014] "price": 2800,

[2015] "description": "A very fruity and sweet wine that pairs well with pasta."

[2016] }

[2017] Terminal: Displays the received information to the user, allowing the user to check the details of the recommended beverage.

[2018] 6. User Feedback

[2019] User: After experiencing the purchased beverage, provide a rating and comment. Enter your feedback in a dedicated form.

[2020] Feedback input example:

[2021] Rating: 4 (out of 5)

[2022] Comment: "Very satisfied. Just as described, perfect with pasta."

[2023] Terminal: Sends user feedback to the server.

[2024] Server: The received feedback is stored in a database and used to improve the accuracy of recommendations next time.

[2025] Specific examples

[2026] For example, if a user wants to drink a "fruity and sweet wine" with "pasta" for dinner for "less than 3,000 yen," the system operates as follows.

[2027] 1. User: Accesses the system app on a smartphone, selects a wine, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta."

[2028] 2. Terminal: Converts the input data into JSON format or similar and sends it to the server.

[2029] 3. Server: Formats the received data and performs any necessary preprocessing.

[2030] 4. Server: Searches the database for wines that meet the above criteria.

[2031] 5. Server: Passes the search results to the generation AI, which evaluates and selects the most suitable wine.

[2032] 6. Server: Sends information about the selected wine to the user's device so that the user can check the information.

[2033] 7. User: Purchases the recommended wine, drinks it, and then provides feedback with a rating and comments.

[2034] 8. Terminal: Sends the provided feedback to the server.

[2035] 9. Server: The feedback is stored in a database and used to improve future recommendations.

[2036] This allows users to easily select the beverage that best suits their tastes and also allows them to easily rate and provide feedback.

[2037] The processing flow will be explained below.

[2038] Step 1: Accepting User Input

[2039] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and meal details. This can be done using a device such as a smartphone, tablet, or PC.

[2040] example:

[2041] Type of drink: Wine

[2042] Favorite flavor: Fruity and sweet

[2043] Budget: Under 3,000 yen

[2044] Meal: Pasta

[2045] Step 2: Send data

[2046] Terminal: Converts user input data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input for errors.

[2047] Example of transmitted data:

[2048] {

[2049] "type": "wine",

[2050] "preferences": "Fruity and sweet",

[2051] "budget": 3000,

[2052] "meal": "pasta"

[2053] }

[2054] Step 3: Data Shaping

[2055] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment).

[2056] Formatting example:

[2057] Check the format

[2058] String normalization

[2059] Category assignment

[2060] Step 4: Database Search

[2061] Server: Queries the database based on the formatted data to find beverage candidates that match the criteria. SQL is used for query processing.

[2062] SQL query example:

[2063] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[2064] Step 5: Evaluate and select candidates

[2065] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best meets the user's criteria. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[2066] Example of evaluation process:

[2067] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[2068] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[2069] Step 6: Submit your information

[2070] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[2071] Examples of information to send:

[2072] {

[2073] "name": "Japanese fruity wine",

[2074] "rating": 4.5,

[2075] "price": 2800,

[2076] "description": "A very fruity and sweet wine that pairs well with pasta."

[2077] }

[2078] Step 7: Viewing information

[2079] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[2080] Display example:

[2081] Drink Name: Japanese Fruity Wine

[2082] Rating: 4.5

[2083] Price: 2800 yen

[2084] Description: A very fruity and sweet wine that pairs well with pasta.

[2085] Step 8: Provide feedback

[2086] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[2087] Feedback input example:

[2088] Rating: 4 (out of 5)

[2089] Comment: "Very satisfied. Just as described, perfect with pasta."

[2090] Step 9: Submit your feedback

[2091] Terminal: Sends the feedback information entered by the user to the server.

[2092] Example of submitted feedback:

[2093] {

[2094] "rating": 4,

[2095] "comment": "Very satisfied. Just as described, it was a perfect match for pasta."

[2096] }

[2097] Step 10: Save your feedback

[2098] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[2099] Save example:

[2100] Feedback data stored in a database

[2101] Use it for your next recommendation

[2102] Through these steps, users can easily choose the best beverage that suits their tastes and rate and provide feedback on their experience.

[2103] Example 1

[2104] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2105] Conventional beverage recommendation systems have difficulty in selecting the optimal beverage for a user's individual preferences, budget, and meal content, and it has been particularly difficult to achieve highly accurate recommendations comparable to those made by a top sommelier. Furthermore, they lacked a mechanism for utilizing user feedback in future recommendations. This resulted in a high risk of users choosing a beverage that did not suit their preferences, often resulting in dissatisfaction with the beverage selection.

[2106] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[2107] In this invention, the server includes a means for the user to input the type of beverage they want to drink, their preferred flavor, their budget, and their meal details; a means for transmitting the user's input to the information processing device; and a means for the information processing device to format the user's input and perform any necessary preprocessing. This enables the information processing device to select the optimal beverage based on the user's input data using a highly accurate database search and a generative AI model. Furthermore, by utilizing user feedback to improve the accuracy of the next recommendation, it is possible to continuously provide highly satisfying selection results.

[2108] A "user" is a person who accesses the system to input the type of beverage they wish to drink, their preferred taste, budget, and dietary needs to receive recommendations.

[2109] The "information processing device" is a device that receives user input data, formats it and performs necessary preprocessing, and then evaluates and selects candidate beverages using a generative AI model.

[2110] The "storage device" is a device that stores information about beverages, word-of-mouth data, past evaluation data, etc., and stores a database that the information processing device uses for searches and evaluations.

[2111] A "generative AI model" is an artificial intelligence model that evaluates potential beverages based on past word-of-mouth and rating data, and selects the beverage that best suits the user's requirements.

[2112] A "terminal device" is a device that transmits user input to an information processing device and displays information about the selected drink to the user. Examples of such devices include smartphones, tablets, and personal computers.

[2113] "Feedback" refers to the ratings and comments provided by users after trying a recommended beverage, and is information used to improve the accuracy of recommendations next time.

[2114] The following describes in detail an embodiment of the present invention, specifically showing what hardware and software are used to process and calculate data.

[2115] 1. Accepting user input

[2116] User:

[2117] First, users access the system using a smartphone, tablet, or PC. Through the system's interface (web application or native application), they input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs. This allows users to easily select a beverage that suits their preferences.

[2118] Example of input contents:

[2119] Type of drink: Wine

[2120] Favorite flavor: Fruity and sweet

[2121] Budget: Under 3,000 yen

[2122] Meal: Pasta

[2123] 2. Data transmission and formatting

[2124] Device:

[2125] The data entered by the user is first converted into an appropriate format (e.g., JSON) and then sent to the information processing device, so that the input data is delivered to the server in a unified format.

[2126] server:

[2127] The information processing device (server) formats the received data, performs necessary format checks and normalizes character strings, and corrects any incomplete or invalid data. For example, it normalizes a taste preference such as "fruity and sweet" and performs preprocessing to facilitate database searches.

[2128] 3. Database Search

[2129] server:

[2130] Based on the formatted data, a search is performed on the storage device (database) to find beverage candidates that match the criteria. To do this, a search algorithm such as an SQL query is used. The database stores detailed information about the beverages, word-of-mouth data, and past evaluation data.

[2131] Example search queries:

[2132] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000;

[2133] 4. Candidate evaluation and selection

[2134] server:

[2135] The candidate list obtained from the search process is passed to a generative AI model for evaluation. The generative AI model takes into account past reviews and evaluation data to select the drink that best suits the user's criteria. This enables highly accurate recommendations comparable to those made by a top sommelier.

[2136] Examples of AI evaluation processes:

[2137] candidate:

[2138] 1. Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[2139] 2. Wine B - Rating 4.2 (Review: "Well-balanced and slightly sweet.")

[2140] AI Selection: Wine A

[2141] 5. Providing information about alcohol

[2142] server:

[2143] Information about the drink that is judged to be the most suitable, such as its name, rating, price, and detailed description, is sent to the user's device. The information sent is standardized in JSON format.

[2144] Examples of information to send:

[2145] {

[2146] "name": "Japanese fruity wine",

[2147] "rating": 4.5,

[2148] "price": 2800,

[2149] "description": "A very fruity and sweet wine that pairs well with pasta."

[2150] }

[2151] Device:

[2152] The received information is displayed to the user, who can then view the details of the recommended beverage.

[2153] 6. User Feedback

[2154] User:

[2155] After tasting the purchased beverage, users can enter their rating and comments in a dedicated feedback form. This feedback will be important data that will help improve the accuracy of the system.

[2156] Examples of feedback include:

[2157] Rating: 4 (out of 5)

[2158] Comment: "Very satisfied. Just as described, perfect with pasta."

[2159] Device:

[2160] The input feedback data is transmitted again to the information processing device.

[2161] server:

[2162] The received feedback is stored in a storage device and used to improve the accuracy of the next beverage recommendation.

[2163] As a result, the system of the present invention allows users to enjoy the perfect beverage that suits their tastes. Furthermore, by continuously collecting feedback, the system's recommendation accuracy continues to improve.

[2164] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2165] Step 1:

[2166] Users access the system using a smartphone, tablet, or computer, and input criteria such as the type of beverage they want, their preferred flavor, budget, and dietary needs through the interface.

[2167] Input: type of drink, preferred flavor, budget, meal contents

[2168] What happens: A user enters data into a web form or in-app input field and presses the submit button.

[2169] Step 2:

[2170] Terminal: Converts the data entered by the user into JSON format or similar and sends it to the information processing device (server).

[2171] Input: User-entered data

[2172] Output: Formatted data such as JSON

[2173] Specific operation: The data obtained from the input field is encoded in JSON format and sent to the server as an HTTP request.

[2174] Step 3:

[2175] Server: Formats the received data, performs any necessary format checks and string normalization, and checks for incomplete or invalid data.

[2176] Input: JSON format data sent from the terminal

[2177] Output: Normalized and formatted data

[2178] Specific operation: For example, the user input "fruity and sweet" is broken down into individual conditions such as "fruity" and "sweet."

[2179] Step 4:

[2180] Server: Searches for beverage candidates that match the criteria from the storage device (database) based on the normalized data.

[2181] Input: Normalized user-entered data

[2182] Output: A list of drinks that match the criteria

[2183] Specific operation: Using SQL queries against the database, candidates are extracted based on criteria such as "wine," "fruity," "sweet," and "under 3,000 yen."

[2184] Step 5:

[2185] Server: The candidate list obtained as a search result is passed to the generative AI model, which evaluates it.

[2186] Input: A list of beverage candidates extracted from the database

[2187] Output: Evaluation by the generative AI model and selection of the optimal beverage

[2188] Specific operation: Using an evaluation algorithm, each candidate is scored, taking into account word-of-mouth data and evaluation data.

[2189] Step 6:

[2190] Server: Sends information about the beverage that has been determined to be optimal to the user's terminal device.

[2191] Input: Selection results by generative AI model

[2192] Output: Detailed drink information sent to user device

[2193] Specific operation: Information including the name, rating, price, and detailed description of the selected drink is encoded in JSON format and sent to the user's device as an HTTP response.

[2194] Step 7:

[2195] Terminal: Displays the received information to the user, who can then check the details of the recommended beverage.

[2196] Input: detailed drink information sent from the server

[2197] Output: Detailed drink information displayed to the user

[2198] Specific behavior: Display detailed information about the drink on a webpage or app screen for the user to review.

[2199] Step 8:

[2200] User: Tastes the purchased beverage and enters a rating and comments in the feedback form.

[2201] Input: User ratings and comments

[2202] Output: Feedback data

[2203] Specific actions: Enter your ratings and comments in the feedback form and press the submit button.

[2204] Step 9:

[2205] Terminal: Transmits the input feedback data to the information processing device.

[2206] Input: User feedback data

[2207] Output: Feedback data sent to the server

[2208] Specific operation: The feedback data is encoded in JSON format and sent to the server as an HTTP request.

[2209] Step 10:

[2210] Server: The received feedback is stored in a storage device and used to improve the accuracy of recommendations next time.

[2211] Input: Feedback data sent from the device

[2212] Output: Feedback data stored in memory device

[2213] Specific operation: The feedback data is inserted into a database and used as training data for future algorithms.

[2214] (Application example 1)

[2215] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2216] Conventional beverage recommendation systems struggle to easily and effectively find beverages that match users' preferences. They also lack a visual and interactive experience, making it difficult to improve user satisfaction. Furthermore, they lack a means to utilize feedback to improve recommendation accuracy.

[2217] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[2218] In this invention, the server includes a means for visually inputting information from the user using a VR head-mounted display, a means for searching for beverage candidates that meet the criteria from a database based on the formatted data, and a means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling the user to easily and effectively find a beverage that suits their preferences through a visual and interactive experience.

[2219] "User" means a consumer who uses the system to receive beverage recommendations.

[2220] "Type of beverage" refers to the category of beverage consumed, such as wine, beer, or sake.

[2221] "Preferred taste" refers to the flavor or characteristics of a beverage desired by a user, including taste attributes such as fruity, sweet, dry, etc.

[2222] "Budget" indicates the maximum amount the user is willing to spend, and the price of the selected drink must not exceed this amount.

[2223] "Meal content" refers to the type of meal the user consumes with the beverage, including specific dish names such as pasta or sushi.

[2224] A "VR head-mounted display" is a display device that allows users to visually experience a virtual reality environment by wearing it.

[2225] The "server" is a computer system that receives user input, formats the data, and uses database search and generative AI to recommend beverages.

[2226] "Formatting" is the process of converting received user input data into an appropriate format, making it suitable for searching and evaluation.

[2227] A "database" is a collection of information that stores information about beverages and word-of-mouth data, and is a target that can be searched based on user criteria.

[2228] "Generative AI" is a type of artificial intelligence that has the ability to evaluate and select the most suitable beverage based on input data.

[2229] "Feedback" refers to the ratings and comments provided by users after their experience, which are used to improve the system's recommendation accuracy.

[2230] "VR space" is a virtual reality world that users experience through a VR head-mounted display.

[2231] System Overview

[2232] In this embodiment, a system is provided in which a user wears a VR head-mounted display and inputs the type of beverage they want to drink, their preferred flavor, their budget, and their meal plan in a virtual space, and the system recommends the most suitable beverage. The system sends the user's input to a server, which performs a database search and evaluation using a generation AI, selects the most suitable beverage, and displays the result in the VR space. The user then provides feedback, which the system stores in a database.

[2233] 1. Accepting user input

[2234] Users put on a VR head-mounted display and log in to the virtual space, where they can use the interface to input the type of drink, their preferred flavor, budget, and meal details.

[2235] 2. Data transmission and formatting

[2236] The device converts the user's input data into JSON format and sends it to the server, which then formats the received data and converts it into the appropriate format.

[2237] 3. Database Search

[2238] The server then searches the database for beverage candidates that match the criteria based on the formatted data, using SQL queries.

[2239] 4. Candidate evaluation and selection

[2240] The server passes the search results to the AI ​​generator, which evaluates and selects the drink that best suits the user's criteria. The AI ​​generator also references word-of-mouth data and past evaluation data.

[2241] 5. Providing information about beverages

[2242] The server sends information about the best drink to the user's device, which then displays the information visually in the VR space.

[2243] 6. User Feedback

[2244] After trying the beverage, users enter their ratings and comments in a dedicated form in the virtual space, and the device sends the feedback to the server.

[2245] 7. Storage and Use of Feedback

[2246] The server stores the received feedback in a database and uses it to improve the accuracy of recommendations next time.

[2247] Hardware and software used

[2248] This system is implemented using the following hardware and software:

[2249] VR head-mounted display: Oculus Quest 2, etc.

[2250] Frontend: VR application

[2251] Server: Flask (Python web framework), SQLite (database)

[2252] Generation AI: OpenAI GPT-3.5 API

[2253] Specific examples

[2254] If a user wants to pair a "fruity and sweet wine" with pasta for dinner and under 3,000 yen, the system works as follows: The user puts on the Oculus Quest 2, selects a wine in the VR space, and enters criteria such as "fruity and sweet," "under 3,000 yen," and "pasta." The device converts the input data into JSON format and sends it to the server. The server formats the received data and performs any necessary preprocessing. The server searches its database for wines that meet the criteria. The server passes the search results to the generation AI, which evaluates and selects the most suitable wine. The server then sends information about the selected wine to the user's device, where the user can view it in the VR space. The user purchases the recommended wine and, after tasting it, provides feedback with a rating and comments. The device then sends the provided feedback to the server. The server stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[2255] Prompt Sentence Examples

[2256] In a beverage recommendation system, please recommend the best wine for the user based on the following criteria:

[2257] Type of drink: Wine

[2258] Favorite flavor: Fruity and sweet

[2259] Budget: Under 3,000 yen

[2260] Meal: Pasta

[2261] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2262] Step 1:

[2263] The user wears a VR head-mounted display and accesses the virtual space interface. The user inputs the type of drink, preferred flavor, budget, and meal contents. This input information is received by the terminal via a form in the VR space.

[2264] Step 2:

[2265] The device converts the user's input data into JSON format and sends the input information to the server. The input information includes "type of drink," "preferred flavor," "budget," and "meal details." The input data is formatted in JSON format and sent to the server as an HTTP request.

[2266] Step 3:

[2267] The server then formats the received JSON data, converting it into a format suitable for database queries. This formatting process includes format checks and string normalization for the input data. The formatted data is then in a format that can be applied to database queries.

[2268] Step 4:

[2269] The server uses the formatted data to search the database for drink candidates that match the criteria, using SQL queries to extract drinks that fit the drink type, flavor requirements, and budget. This search result may contain multiple candidates.

[2270] Step 5:

[2271] The server passes the search results to a generative AI model, which evaluates and selects the beverage that best suits the user's criteria. The generative AI model selects the optimal beverage based on the evaluation and word-of-mouth data for each beverage included in the search results.

[2272] Step 6:

[2273] The server converts the information about the optimal drink selected by the generative AI model into JSON format and sends it to the device. The transmitted data includes the drink name, rating, price, and description. The device receives this information and interprets the JSON data.

[2274] Step 7:

[2275] The device visually displays the received beverage information in the VR space. Detailed information about the selected beverage is visually presented to the user in the virtual space interface. The user can review this information and make a purchase decision.

[2276] Step 8:

[2277] After purchasing and experiencing the beverage, users can provide their rating and feedback through a virtual interface, which is entered into a dedicated form.

[2278] Step 9:

[2279] The device converts the feedback provided by the user into JSON format and sends it to the server. The feedback data includes rating values ​​and comments. The server interprets the received feedback data and stores it in a database.

[2280] Step 10:

[2281] The server stores the feedback information in a database and refers to it the next time the server recommends a beverage. The stored feedback is used to improve the accuracy of future recommendations.

[2282] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[2283] System Overview

[2284] This system recommends the best drink based on the type of drink a user wants, their preferred flavor, their budget, and their meal. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system can select drinks that take into account the user's emotional information. This allows ordinary employees and consumers to make selections comparable to those of top sommeliers, providing more personalized and precise recommendations.

[2285] DETAILED DESCRIPTION OF THE EMBODIMENTS

[2286] 1. Accepting user input

[2287] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[2288] Example user input:

[2289] Type of drink: Wine

[2290] Favorite flavor: Fruity and sweet

[2291] Budget: Under 3,000 yen

[2292] Meal: Pasta

[2293] Emotional data: facial photos, audio recordings, text comments

[2294] 2. Data transmission and formatting

[2295] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[2296] Example of transmitted data:

[2297] {

[2298] "type": "wine",

[2299] "preferences": "Fruity and sweet",

[2300] "budget": 3000,

[2301] "meal": "pasta",

[2302] "emotion_data": {

[2303] "photo": "image data",

[2304] "voice": "voice data",

[2305] "text": "I'm tired today and would like a relaxing drink"

[2306] }

[2307] }

[2308] 3. Data Formatting

[2309] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[2310] Formatting example:

[2311] Check the format

[2312] String normalization

[2313] Category assignment

[2314] 4. Emotion Data Analysis

[2315] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[2316] Emotion analysis example:

[2317] Emotion: Tired

[2318] 5. Database Search

[2319] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[2320] SQL query example:

[2321] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[2322] 6. Evaluation and selection of candidates

[2323] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[2324] Example of evaluation process:

[2325] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[2326] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[2327] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[2328] 7. Transmission of Information

[2329] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[2330] Examples of information to send:

[2331] {

[2332] "name": "Japanese fruity wine",

[2333] "rating": 4.7,

[2334] "price": 2800,

[2335] "description": "A very fruity, sweet and relaxing wine."

[2336] }

[2337] 8. Display of Information

[2338] Terminal: Displays the received information to the user, who can then check the details of the selected beverage based on the displayed information.

[2339] Display example:

[2340] Drink Name: Japanese Fruity Wine

[2341] Rating: 4.7

[2342] Price: 2800 yen

[2343] Description: A very fruity, sweet and relaxing wine.

[2344] 9. Providing Feedback

[2345] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[2346] Feedback input example:

[2347] Rating: 4 (out of 5)

[2348] Comment: "Very satisfied. It was relaxing just as described."

[2349] 10. Submitting Feedback

[2350] Terminal: Sends the feedback information entered by the user to the server.

[2351] Example of submitted feedback:

[2352] {

[2353] "rating": 4,

[2354] "comment": "Very satisfied. It was as relaxing as described."

[2355] }

[2356] 11. Saving Feedback

[2357] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[2358] Save example:

[2359] Feedback data stored in a database

[2360] Use it for your next recommendation

[2361] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[2362] The processing flow will be explained below.

[2363] Step 1: Accepting User Input

[2364] User: Accesses the system and inputs the type of drink they want, their preferred flavor, budget, and dietary needs. They also provide this data so that the system can analyze their emotions from their facial expressions, voice, and text. This can be done using a device such as a smartphone, tablet, or PC.

[2365] example:

[2366] Type of drink: Wine

[2367] Favorite flavor: Fruity and sweet

[2368] Budget: Under 3,000 yen

[2369] Meal: Pasta

[2370] Emotional data: facial photos, audio recordings, text comments

[2371] Step 2: Send data

[2372] Terminal: Converts the user's input data and emotion data into an appropriate format (e.g., JSON) and sends it to the server. When sending the data, it checks the input content for errors.

[2373] Example of transmitted data:

[2374] {

[2375] "type": "wine",

[2376] "preferences": "Fruity and sweet",

[2377] "budget": 3000,

[2378] "meal": "pasta",

[2379] "emotion_data": {

[2380] "photo": "image data",

[2381] "voice": "voice data",

[2382] "text": "I'm tired today and would like a relaxing drink"

[2383] }

[2384] }

[2385] Step 3: Data Shaping

[2386] Server: Formats the data received from the device, including format checks and any necessary preprocessing (e.g., string normalization, category assignment), and passes the data to the emotion engine for emotion analysis.

[2387] Formatting example:

[2388] Check the format

[2389] String normalization

[2390] Category assignment

[2391] Step 4: Analyze the sentiment data

[2392] Server: The emotion engine analyzes the user's facial expressions, voice, and text to determine their emotions. The analyzed emotion data is then extracted as information based on the user's current emotions (e.g., fatigue, stress, joy, etc.).

[2393] Emotion analysis example:

[2394] Emotion: Tired

[2395] Step 5: Database Search

[2396] Server: Queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that match the criteria. SQL is used for query processing.

[2397] SQL query example:

[2398] SELECT FROM beverage_data WHERE type='wine' AND flavor LIKE '%fruity%' AND flavor LIKE '%sweet%' AND price <= 3000 AND suitable_for='relaxing';

[2399] Step 6: Evaluate and select candidates

[2400] Server: Passes beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data.

[2401] Example of evaluation process:

[2402] 1. Candidate Wine A - Rating 4.5 (Review: "Very fruity and sweet, perfect with pasta.")

[2403] 2. Wine B - Rating: 4.2 (Review: "Well-balanced, slightly sweet.")

[2404] 3. Wine C - Rating 4.7 (Review: "Relaxing taste")

[2405] Step 7: Submit your information

[2406] Server: Sends information about the selected best drink to the user's device, including the name, rating, price, and detailed description.

[2407] Examples of information to send:

[2408] {

[2409] "name": "Japanese fruity wine",

[2410] "rating": 4.7,

[2411] "price": 2800,

[2412] "description": "A very fruity, sweet and relaxing wine."

[2413] }

[2414] Step 8: Viewing Information

[2415] Terminal: The received information is displayed to the user, who can then check the details of the selected beverage based on the displayed information.

[2416] Display example:

[2417] Drink Name: Japanese Fruity Wine

[2418] Rating: 4.7

[2419] Price: 2800 yen

[2420] Description: A very fruity, sweet and relaxing wine.

[2421] Step 9: Provide feedback

[2422] User: After trying the beverage, provide feedback such as a rating and comments. Enter the rating details into the dedicated form and press the submit button.

[2423] Feedback input example:

[2424] Rating: 4 (out of 5)

[2425] Comment: "Very satisfied. It was relaxing just as described."

[2426] Step 10: Submit your feedback

[2427] Terminal: Sends the feedback information entered by the user to the server.

[2428] Example of submitted feedback:

[2429] {

[2430] "rating": 4,

[2431] "comment": "Very satisfied. It was as relaxing as described."

[2432] }

[2433] Step 11: Save your feedback

[2434] Server: Receives feedback and stores it in a database. This feedback is used to improve the accuracy of the recommendation system in the future.

[2435] Save example:

[2436] Feedback data stored in a database

[2437] Use it for your next recommendation

[2438] Through these steps, users can easily select the perfect drink based on their preferences and emotions, and then rate and provide feedback on their experience. Through this process, we can provide a more personalized and refined drink selection.

[2439] Example 2

[2440] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2441] Conventional beverage recommendation systems recommend drinks based on a user's preferred taste, budget, and dietary preferences, but because they are unable to consider the user's emotional information, it is difficult to recommend the optimal drink for the user's psychological state. Furthermore, there is a lack of technology to effectively utilize feedback to improve recommendation accuracy.

[2442] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[2443] In this invention, the server includes means for extracting emotion data using an emotion engine that analyzes user emotions, means for searching a database for beverage candidates that meet the criteria based on the formatted data and analyzed emotion data, and means for evaluating the candidates using a generative AI and selecting the optimal beverage, thereby enabling sophisticated beverage recommendations based on the user's preferences and emotion information.

[2444] "User" means an individual or entity that intends to use the System to select a beverage.

[2445] "Beverage type" refers to the category of beverage selected, such as wine, beer, juice, etc.

[2446] "Preferred taste" refers to the flavor or characteristics of the beverage that the user prefers, such as fruity, sweet, dry, etc.

[2447] "Budget" refers to the monetary limit a User is willing to pay for a beverage.

[2448] "Meal content" refers to the type and menu of food that the user consumes along with the beverage they drink.

[2449] "Emotional Data" refers to data that indicates a user's emotional state analyzed from facial expressions, voice, and text.

[2450] "Terminal" means a device used by a User to access the System, including a smartphone, tablet, or PC.

[2451] "Server" refers to the central processing unit that processes, stores, and analyzes data submitted by users.

[2452] "Formatting" refers to the process of processing data received from a user and converting it into a usable format.

[2453] "Emotion engine" refers to software or hardware for analyzing a user's facial expressions, voice, and text to extract emotional data.

[2454] "Database" refers to a data management system that stores information about beverages and makes it accessible by query.

[2455] "Generative AI" refers to an artificial intelligence model that evaluates candidates based on past data and reviews and selects the most suitable beverage.

[2456] "Feedback" refers to the ratings and comments that users make about the beverages they have experienced.

[2457] The present invention is a system that recommends optimal drinks based on the user's input of the type of drink they want to drink, their preferred taste, budget, dietary content, and emotional data. Detailed explanations and concrete examples necessary for implementing this system are provided below.

[2458] Accepting user input

[2459] Users access the system from devices such as smartphones, tablets, and PCs and enter the type of beverage they want to drink, their preferred flavor, budget, dietary preferences, and emotional data. For example, users can enter the following data:

[2460] Example prompt sentence:

[2461] Type of drink: Wine

[2462] Favorite flavor: Fruity and sweet

[2463] Budget: Under 3,000 yen

[2464] Meal: Pasta

[2465] Emotional data: "I'm tired today and would like a drink to help me relax."

[2466] Data transmission and formatting

[2467] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input content and checks for errors. For example, the device generates the following JSON data and sends it to the server.

[2468] {

[2469] "type": "wine",

[2470] "preferences": "Fruity and sweet",

[2471] "budget": 3000,

[2472] "meal": "pasta",

[2473] "emotion_data": {

[2474] "photo": "image data",

[2475] "voice": "voice data",

[2476] "text": "I'm tired today and would like a relaxing drink"

[2477] }

[2478] }

[2479] Data Shaping and Sentiment Analysis

[2480] The server formats the data received from the device. This process includes format checks, string normalization, and category assignment. It also uses an emotion engine to analyze emotion data, extracting emotion data from the user's facial expressions, voice, and text. For example, it can detect fatigue levels from the text "I'm tired today."

[2481] Database search

[2482] The server queries the database based on the formatted beverage data and analyzed emotion data to find beverage candidates that meet the criteria. SQL is used for query processing. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[2483] Candidate evaluation and selection

[2484] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the best wine from candidate wines A, B, and C.

[2485] Sending and Displaying Information

[2486] The server sends the information about the selected optimal drink to the user's terminal. The terminal displays the received information to the user. For example, the following information is displayed:

[2487] Display example:

[2488] Drink Name: Japanese Fruity Wine

[2489] Rating: 4.7

[2490] Price: 2800 yen

[2491] Description: A very fruity, sweet and relaxing wine.

[2492] Providing and storing feedback

[2493] After trying the beverage, the user provides feedback such as a rating and comments. The device sends the feedback information entered by the user to the server. The server receives the feedback and stores it in a database. This feedback will be used to improve the accuracy of the recommendation system in the future.

[2494] This allows users to easily select the perfect drink based on their preferences and emotions, and rate and provide feedback on their experience, allowing for a more personalized and refined selection of drinks.

[2495] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2496] Step 1:

[2497] Users access the system using devices such as smartphones, tablets, and PCs and input the type of beverage they want, their preferred flavor, budget, dietary preferences, and emotional data. This input includes a facial photo, audio recording, and text comments. An example of a prompt sentence is, "Please choose your favorite wine. I like fruity and sweet flavors, and I'd like something that goes well with pasta, within a budget of 3,000 yen or less. I'm tired after a long day today, so I'd like a drink that helps me relax."

[2498] Input: type of drink, preferred flavor, budget, meal contents, emotional data

[2499] Output: Information entered by the user

[2500] Step 2:

[2501] The device converts the data entered by the user into JSON format and sends it to the server. At this time, it also checks the input and performs error checks. For example, it checks whether all required fields are filled in. Data that passes the error check is composed as JSON and sent.

[2502] Input: User-entered data

[2503] Output: JSON format data

[2504] Step 3:

[2505] The server formats the JSON data received from the device. This includes formatting, string normalization, and category assignment. For example, it might verify that the type of drink entered is correct and correct any typos in the string. The formatted data is then converted into a standard format for internal processing.

[2506] Input: JSON format data

[2507] Output: Formatted user data

[2508] Step 4:

[2509] The server uses an emotion engine to analyze emotion data from the user's facial expressions, voice, and text. For example, the emotion engine detects the user's fatigue state from text such as "I'm tired." Based on the results of this analysis, the server quantifies or categorizes the user's emotional state.

[2510] Input: User's emotional data (face photo, voice, text)

[2511] Output: Analyzed emotion data (e.g., fatigue level)

[2512] Step 5:

[2513] The server queries the database based on the formatted beverage data and the analyzed emotion data to find beverage candidates that meet the criteria. This query uses SQL. For example, it searches for beverages that match the criteria "wine," "fruity," "sweet," "budget under 3,000 yen," and "relaxing."

[2514] Input: formatted beverage data, parsed emotion data

[2515] Output: List of beverage candidates

[2516] Step 6:

[2517] The server passes the beverage candidates retrieved from the database to the generation AI, which selects the candidate that best suits the user's conditions and emotions. The generation AI evaluates the candidates based on word-of-mouth data and past evaluation data. For example, it selects the candidate with the highest evaluation score.

[2518] Input: Beverage candidate list, review data, past evaluation data

[2519] Output: Best beverage candidates

[2520] Step 7:

[2521] The server then sends information about the selected optimal drink to the user's device. The information sent includes the name of the drink, its rating, price, and a detailed description. For example, "Japanese fruity wine, rating 4.7, price 2,800 yen, very fruity, sweet, and relaxing."

[2522] Input: Best beverage candidates

[2523] Output: detailed information about the beverage

[2524] Step 8:

[2525] The device then displays the received information to the user, allowing the user to check the details of the selected beverage based on the displayed information. For example, the name of the beverage, its rating, price, and detailed description are visually displayed.

[2526] Input: Beverage details

[2527] Output: Beverage information displayed to the user

[2528] Step 9:

[2529] After trying the beverage, users provide feedback by rating and commenting on it. They enter their rating in a special feedback form and press the "Submit" button. For example, a rating of 4 (out of 5) and a comment like "I was very satisfied. It was relaxing, just as described."

[2530] Input: Feedback after the experience

[2531] Output: Feedback information

[2532] Step 10:

[2533] The terminal sends the feedback information entered by the user to the server. The sending process includes converting the feedback data into JSON format and sending it.

[2534] Input: User feedback information

[2535] Output: Feedback data sent to the server

[2536] Step 11:

[2537] The server receives the feedback and stores it in a database, which can be used to improve the accuracy of the recommendation system in the future, for example, by improving the recommendation algorithm next time.

[2538] Input: Feedback data sent to the server

[2539] Output: Feedback stored in a database

[2540] (Application example 2)

[2541] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2542] In today's food and beverage selection process, it is difficult to make recommendations that accurately reflect a user's individual preferences and current emotional state. Food delivery services also require recommendation systems that can quickly respond to diverse user needs. While existing systems can recommend beverages based on a user's preferences and budget, they are unable to take into account the user's emotions. This makes it difficult for users to select a beverage that satisfies them. Furthermore, there has been a lack of systems that can improve the accuracy of recommendations based on feedback.

[2543] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[2544] In this invention, the server includes means for formatting the type, preferences, budget, meal details, and emotion data entered by the user and performing the necessary preprocessing, means for analyzing emotions using an emotion engine based on the emotion data, and means for searching a database for beverage candidates that meet the conditions based on the analyzed emotion data, thereby making it possible to recommend optimal beverages that match the user's individual preferences and emotional state.

[2545] "User" refers to a person who uses the system to receive beverage recommendations.

[2546] "Beverage type" refers to the classification of different beverages, such as wine, beer, juice, etc.

[2547] "Preferred taste" refers to the taste characteristics of a beverage that a user prefers, such as fruity, sweet, or dry.

[2548] "Budget" refers to the maximum amount a user is willing to pay for a beverage.

[2549] "Meal Content" refers to the type and details of the meal the user plans to consume with the beverage.

[2550] "Emotional Data" refers to data about a user's current emotional state, including a photo of the user's face, audio recordings, text comments, etc.

[2551] "Formatting" refers to the process of converting user-entered data into a format suitable for analysis and search.

[2552] "Preprocessing" refers to preparing data to make it easier to analyze, such as normalizing the data or assigning categories.

[2553] An "emotion engine" refers to a system that analyzes a user's emotions from data such as facial photos, voice, and text.

[2554] A "database" refers to a collection of information that systematically stores information such as types of beverages, flavors, prices, and ratings.

[2555] "Candidates" refers to a list of beverages extracted based on the user's conditions and emotions.

[2556] "Generative AI" refers to an artificial intelligence algorithm that selects the optimal beverage based on given data.

[2557] "Terminal" refers to a device that a user uses to operate the system, such as a smartphone, tablet, or PC.

[2558] "Feedback" refers to ratings and comments provided by users based on their usage experience.

[2559] "Storing in a database" refers to storing feedback and other data in a database so that it can be reused for later retrieval and analysis.

[2560] The present invention relates to a system for recommending optimal beverages based on a user's individual preferences and emotional state, which is implemented using a set of hardware and software, such as:

[2561] System Overview

[2562] 1. Accepting user input

[2563] Users use devices such as smartphones, tablets, and computers to input the type of drink they want, their preferred flavor, budget, meal contents, and emotional data, which includes a photo of their face, audio recordings, and text comments.

[2564] 2. Data transmission and formatting

[2565] The device converts the user's input data and emotion data into JSON format and sends it to the server, which then performs formatting processes such as format checks and normalization on the received data.

[2566] 3. Emotion Data Analysis

[2567] The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format, and include emotional states such as "fatigue."

[2568] 4. Search for candidates

[2569] The server uses an SQL query to search the database for beverage candidates that meet the analyzed emotional data and the user's formatted conditions.

[2570] 5. Candidate evaluation and selection

[2571] The server uses generative AI (e.g., OpenAI's GPT model) to evaluate beverage candidates retrieved from the database and select the beverage that best suits the user's conditions and emotions.

[2572] 6. Transmission and Display of Information

[2573] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which then displays this information to the user.

[2574] 7. Providing and Saving Feedback

[2575] After tasting the beverage, the user sends their rating and comments to the server, which stores the feedback in a database and uses it to improve the accuracy of future recommendations.

[2576] Specific examples

[2577] Prompt Sentence Examples

[2578] Choose the best beverage from the following options: Emotional information: Fatigue, Options: [Wine A (Rating: 4.5, Review: "Very fruity and sweet, perfect with pasta"), Wine B (Rating: 4.2, Review: "Well-balanced, with a touch of sweetness"), Wine C (Rating: 4.7, Review: "Relaxing taste")]

[2579] By inputting this prompt into a generative AI model, the optimal beverage can be selected based on the user's emotions and review information. This invention enables more personalized and precise beverage recommendations based on the user's emotional state. This allows users to easily select the beverage that best suits their specific emotional state, improving the quality of their dining experience.

[2580] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2581] Step 1:

[2582] Users use devices such as smartphones, tablets, and PCs to input the type of drink they want, their preferred flavor, budget, dietary preferences, and emotional data, including facial photos, audio recordings, and text comments. This data is collected through an input form on the device.

[2583] Input: User's drink type, preferred flavor, budget, dietary details, emotional data

[2584] Output: User input data in JSON format

[2585] Step 2:

[2586] The device converts the user's input data and emotion data into JSON format and sends it to the server, checking the data format and checking for errors before sending.

[2587] Input: User-entered data

[2588] Output: HTTP request with data in JSON format

[2589] Step 3:

[2590] The server performs format checks and normalization on the received data, including normalizing strings and assigning categories, and then sends the resulting data to a sentiment analysis engine.

[2591] Input: User-supplied data in JSON format

[2592] Output: Formatted data and data for sentiment analysis

[2593] Step 4:

[2594] The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure Emotion API) to analyze the user's emotions from their facial expressions, voice, and text. The analysis results are returned in JSON format, and include emotional states such as "fatigue."

[2595] Input: Formatted data and sentiment data

[2596] Output: Parsed emotional state (e.g. "fatigue")

[2597] Step 5:

[2598] The server uses SQL queries to search for beverage candidates that match the analyzed emotion data and the user's preferences from a database containing information such as the type of beverage, flavor characteristics, price, and suitable meals.

[2599] Input: A search query for a beverage that matches your criteria (type, preference, budget, diet, emotional state)

[2600] Output: A list of beverage candidates as search results

[2601] Step 6:

[2602] The server evaluates and prioritizes potential drinks using a generative AI (e.g., OpenAI's GPT model). The model selects the best drink based on the user's sentiment and reviews based on the input prompt.

[2603] Input: Drink candidate list and emotional state

[2604] Output: Best drink information (name, rating, price, detailed description)

[2605] Step 7:

[2606] The server sends information about the selected drink (such as name, rating, price, and detailed description) to the user's device, which receives this information and displays it in a user-friendly format.

[2607] Input: Optimal beverage information

[2608] Output: Beverage information displayed on the terminal

[2609] Step 8:

[2610] After tasting the beverage, the user sends their rating and comments to the server from their device using a feedback form provided in the UI.

[2611] Input: User ratings and comments

[2612] Output: Feedback data in JSON format

[2613] Step 9:

[2614] The server verifies the received feedback and stores it in a database, which then serves as data to improve the accuracy of future recommendations.

[2615] Input: Feedback data in JSON format

[2616] Output: Ratings and comments stored in the database

[2617] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2618] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2619] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2620] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2621] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2622] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2623] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2624] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2625] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2626] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2627] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2628] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2629] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2630] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2631] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2632] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2633] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2634] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2635] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2636] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2637] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2638] The following is further disclosed regarding the above embodiment.

[2639] (Claim 1)

[2640] A way for users to input the type of beverage they want to drink, their preferred flavor, budget, and meal contents;

[2641] means for transmitting said user input to a server;

[2642] means for said server to format user input and perform any necessary pre-processing;

[2643] A means for the server to search a database for beverage candidates that meet the conditions based on the formatted data;

[2644] A means for the server to evaluate candidates using a generation AI and select an optimal beverage;

[2645] means for the server to transmit information about the selected beverage to a user's terminal;

[2646] means for the terminal to display information about the beverage to the user;

[2647] a means for said user to provide feedback;

[2648] means for transmitting said feedback to a server;

[2649] means for the server to store the feedback in a database;

[2650] A system including:

[2651] (Claim 2)

[2652] The system of claim 1, wherein the generation AI selects beverages based on word-of-mouth data and past evaluation data.

[2653] (Claim 3)

[2654] 10. The system of claim 1, wherein the server...

Claims

1. A way for users to input the type of beverage they want to drink, their preferred flavor, budget, and meal contents; means for transmitting said user input to a server; means for said server to format user input and perform any necessary pre-processing; A means for the server to search a database for beverage candidates that meet the conditions based on the formatted data; A means for the server to evaluate candidates using a generation AI and select an optimal beverage; means for the server to transmit information about the selected beverage to a user's terminal; means for the terminal to display information about the beverage to the user; a means for said user to provide feedback; means for transmitting said feedback to a server; means for the server to store the feedback in a database; A system including:

2. The system according to claim 1, wherein the generation AI selects beverages based on word-of-mouth data and past evaluation data.

3. 10. The system of claim 1, wherein the server considers price, taste preferences, and food type when rating and selecting beverages based on the user's requirements.

Images