System

Abstract

To provide a system capable of proposing a proper diet based on the feeling state and past selection history of a user to a user.SOLUTION: A system for assisting users in making meal choices includes: means for conducting machine learning on a learning model using the food preferences and food menu selection history of a user; means for generating a meal menu to be proposed to the user using the learning model; means for recognizing the feeling state of the user by analyzing the facial expressions and voice of the user by a feeling engine; means for generating prompt text instructing the proposal of a meal that the current user is likely to prefer on the basis of the current feeling state of the recognized user and the meal menu generated by the learning model; and means for proposing a meal to the user using a generative AI model.SELECTED DRAWING: Figure 1

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 a description and related instruction sentence regarding 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] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] When users choose a meal, they face the challenge of having to do the research and select customizations, which can be tedious and time-consuming. [Means for solving the problem]

[0005] To address this issue, we provide an AI system that learns from a user's preferences and past selection history and makes meal suggestions based on the results. Furthermore, we enable customization of the suggested meals and automatically customize them based on the user's past selection history and feedback, thereby efficiently supporting the user's meal selection. [Brief description of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 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. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 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. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 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. 2 is a sequence diagram showing a process flow of the data processing system in the first embodiment of the first form example. [Figure 12] FIG. 11 is a sequence diagram showing a process flow of the data processing system in the application example 1 of the first embodiment. [Figure 13] FIG. 11 is a sequence diagram showing a process flow of a data processing system in the second embodiment of the second embodiment. [Figure 14] FIG. 13 is a sequence diagram showing a process flow of the data processing system in application example 2 of embodiment 2. [Figure 15] FIG. 13 is a sequence diagram showing a process flow of a data processing system in a third embodiment of the present invention. [Figure 16] FIG. 13 is a sequence diagram showing a process flow of the data processing system in the third application example of the third embodiment. [Figure 17] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when combined with an emotion engine. [Figure 19] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system according to the embodiment 2 of the second embodiment in the case where an emotion engine is combined. [Figure 20] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when combined with an emotion engine. [Figure 21] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in the embodiment 3 of the form example 3 in the case where an emotion engine is combined. [Figure 22] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when combined with an emotion engine. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0011] 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.

[0012] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0013] 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. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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 wide area network (WAN) and / or a local area network (LAN).

[0018] 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.

[0019] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. 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.

[0020] 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 (e.g., voice and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs voice according to instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Fig. 2, 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. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process 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 process program 56 executed on the RAM 30.

[0024] 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.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores a reception output program 60. The reception output program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads out the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output processing 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] As one embodiment of the present invention, an AI system that supports a user in selecting meals is provided. This system has a function to learn the user's preferences and past selection history. Specifically, the system accumulates data on the meals the user has selected in the past and the user's tendency toward preferred meals, and the AI ​​learns from this data.

[0029] "Example 2"

[0030] In addition, the AI ​​also has the ability to suggest meals based on the results of its learning. Specifically, it suggests meals that the user is likely to choose next based on the user's preferences and past selection history that it has learned. For example, if the user has often chosen Japanese food in the past, the AI ​​will suggest Japanese food that the user is likely to like.

[0031] "Example 3"

[0032] It also has a function that allows users to customize the suggested meals. Specifically, users can customize the suggested meals by adding or removing ingredients, changing the cooking method, etc. For example, users can add toppings or increase the amount of cheese to the pizza suggested by the AI.

[0033] The process flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: When a user begins using the system, the system begins recording the user's food choices, including the types of food the user chooses, the time of meal, and frequency of meals.

[0036] Step 2: The system accumulates this data and the AI ​​learns about the user's preferences and choices. This learning process evolves every time the user uses the system, providing more accurate results.

[0037] "Example 2"

[0038] Step 1: After the AI ​​learns the user's preferences and selection trends, it will suggest the next meal the user is likely to choose. These suggestions will be displayed when the user logs into the system, when it is time to eat, etc.

[0039] Step 2: The user can choose to either select the suggested meal or choose a different meal. If the user selects the suggested meal, that information is again fed back into the system to help train the AI.

[0040] "Example 3"

[0041] Step 1: After the user selects a suggested meal, they can then customize it by adding or removing ingredients, changing the cooking method, etc.

[0042] Step 2: The user's customization information is also fed back to the system and used to train the AI, allowing the system to better understand the user's preferences and provide more personalized meal suggestions.

[0043] Example 1

[0044] Next, a description will be given of Example 1 of the embodiment 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".

[0045] Conventional meal selection support systems had the problem that they did not make suggestions that fully considered the user's preferences or past selection history, resulting in low user satisfaction. In addition, the suggested meals were customized manually, which was a hassle for the user. Furthermore, the system lacked the functionality to learn the user's selection trends and recommend appropriate meals, making it difficult to suggest meals that matched the user's preferences.

[0046] 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.

[0047] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for proposing meals based on the results of the learning, a means for enabling customization of the proposed meals, a means for collecting the user's meal data, a means for accumulating the collected data, a means for learning the user's selection tendency based on the accumulated data, a means for generating recommended meals based on the learned model, and a means for notifying the user of the generated recommended meals. This enables meal suggestions that take into account the user's preferences and past selection history, thereby improving user satisfaction. In addition, the proposed meals are automatically customized, reducing the user's effort. Furthermore, by learning the user's selection tendency and recommending appropriate meals, meal suggestions that match the user's preferences can be realized.

[0048] The "means for learning the user's preferences and past selection history" refers to a device or program that has the function of collecting data on the meals selected by the user in the past and analyzing the user's preferences and selection trends based on that data.

[0049] A "means for making meal suggestions based on the results of learning" is a device or program that has the function of recommending appropriate meals based on the user's preferences and past selection history.

[0050] A "means for enabling customization of meal suggestions" is a device or program that has the functionality to allow a user to modify the suggested meals to suit their personal preferences.

[0051] A "means for collecting user meal data" is a device or program that has the function of inputting information about meals selected by the user and collecting it as data.

[0052] The "means for storing collected data" refers to a device or program that has the function of storing collected user meal data in a storage device such as a database.

[0053] The "means for learning the user's selection tendencies based on accumulated data" refers to a device or program that has the function of analyzing accumulated meal data and modeling the user's selection tendencies.

[0054] A "means for generating recommended meals based on a learned model" is a device or program that has the function of generating appropriate meals based on the learned selection tendencies of a user.

[0055] The "means for notifying the user of the generated recommended meals" refers to a device or program having a function of transmitting the generated recommended meals to the user's terminal and notifying the user of the recommended meals.

[0056] This invention is an AI system for assisting users in choosing meals, learning from the user's preferences and past selection history and proposing appropriate meals. A specific embodiment of this system will be described below.

[0057] First, the user inputs meal information using the terminal. The terminal then transmits the input meal information (e.g., meal name, date and time, rating) to the server. The server receives the meal information transmitted from the terminal and stores it in a database. General database software such as MySQL (registered trademark) is used as the database management system.

[0058] Next, the server learns the user's selection trends based on the accumulated data. The machine learning library used is TENSORFLOW (registered trademark). The server uses TensorFlow to model the user's food selection patterns and analyze the user's preferences and selection trends.

[0059] Once the learning is complete, the server generates meal recommendations based on the learned model. Using the generative AI model, it suggests three meals that match the user's preferences. For example, if the user previously selected "curry rice," "ramen," and "sushi," the server will recommend "tempura," "udon," and "sushi."

[0060] Finally, the server sends the generated recommended meals to the user's device. The notification service used is Firebase Cloud Messaging. The device notifies the user of the received recommended meals and displays them on the screen.

[0061] As a specific example, a user selects "curry rice" for lunch and enters "curry rice" into an app on the device. The device sends the entered meal information "curry rice" to the server. The server stores the received "curry rice" data in a MySQL database. Next, the server uses TensorFlow to learn that the user tends to prefer "Japanese food." Based on the learned model, the server recommends "tempura," "udon," and "sushi" and notifies the user's device of this recommendation.

[0062] An example of a prompt is, "Generate the next meal recommendation based on the user's past meal selection data. Suggest three specific meal names taking into account the user's preferences and selection trends."

[0063] This is how the AI ​​system that assists users in making meal choices works.

[0064] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0065] Step 1:

[0066] Collecting user dietary data

[0067] The user selects a meal and enters the information into the terminal.

[0068] Input: Meal information such as meal name, date and time, and rating.

[0069] The terminal transmits the input meal information to the server.

[0070] Output: Meal information sent to the server.

[0071] Specific actions: The user selects "curry rice" for lunch and enters "curry rice" into the device app.

[0072] Step 2:

[0073] Accumulation of data

[0074] The server receives the meal information transmitted from the terminal.

[0075] Input: Meal information sent from the device.

[0076] The server stores the received data in a database.

[0077] Output: Meal information stored in a database.

[0078] Specific operation: The server stores the received "curry rice" data in a MySQL database.

[0079] Step 3:

[0080] Training the Data

[0081] The server learns the user's preferences and selection trends based on the meal information accumulated in the database.

[0082] Input: Food information stored in a database.

[0083] The server uses machine learning libraries (e.g., TensorFlow) to model users' food choice patterns.

[0084] Output: The trained model.

[0085] Specific behavior: The server uses TensorFlow to learn that users tend to like "Japanese food."

[0086] Step 4:

[0087] Generation of recommended meals

[0088] The server generates meal recommendations for the user based on the learned model.

[0089] Input: The trained model.

[0090] The server uses a generative AI model to suggest three meals that match the user's preferences.

[0091] Output: A recommended diet list.

[0092] Specific operation: The server recommends "tempura," "udon," and "sushi" based on the learned model.

[0093] Step 5:

[0094] Dietary recommendations

[0095] The server transmits the generated recommended meals to the user's terminal.

[0096] Input: A recommended meal list.

[0097] The device will notify the user of the recommended meals it has received.

[0098] Output: The meal recommendation communicated to the user.

[0099] Specific operation: The server sends the generated recommended meals, "tempura," "udon," and "sushi," to the user's terminal, and the terminal notifies the user of them.

[0100] (Application example 1)

[0101] Next, a description will be given of Application Example 1 of Form 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."

[0102] Conventional food delivery systems do not adequately suggest meals based on the user's preferences and past selection history. In addition, they lack the ability to suggest meals according to the user's current mood and health status, and provide nutritional information and calorie calculations for meals, which results in low user satisfaction and poor health management. This makes it difficult for users to select appropriate meals.

[0103] 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.

[0104] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for customizing the meal suggestions based on the user's current mood and health condition, and a means for providing nutritional information and calorie calculations for the meals. This allows optimal meal suggestions based on the user's preferences and health condition, improving the user's satisfaction and health management.

[0105] "User preferences" refers to the meals selected by the user in the past and the user's particular preferences for meals.

[0106] "Past selection history" refers to the record of meals the user has selected in the past.

[0107] "Means of learning" refers to how the AI ​​analyzes data and identifies patterns and trends based on user preferences and past choices.

[0108] "Means for providing meal suggestions" refers to a method for recommending appropriate meals to a user based on the results of learning.

[0109] "Means enabling customization of meal suggestions" refers to a method that allows a user to modify the suggested meals to suit their personal preferences.

[0110] "Means for customizing meal suggestions based on current mood and health status" refers to a method for tailoring meal suggestions to a user's current mood and health status.

[0111] "Means for providing nutritional information and calorie calculations for a meal" means a method for calculating and providing to a user the nutrients and calories contained in a suggested meal.

[0112] As an embodiment of the present invention, an AI system that assists users in choosing meals will be specifically described. This system learns the user's preferences and past selection history, and suggests optimal meals. The configuration and operation of the system will be described in detail below.

[0113] System Configuration

[0114] The system consists of the following main components:

[0115] 1. User device: A device operated by a user, such as a smartphone or tablet.

[0116] 2. Server: The central system that processes and stores data.

[0117] 3. AI model: A machine learning model that learns from user preferences and past selection history to provide meal suggestions.

[0118] Hardware and Software Used

[0119] Hardware: Smartphones, tablets, servers

[0120] Software: Python, Pandas, Scikit-learn

[0121] Data processing and calculation

[0122] 1. Data collection: Collect data from the user's device about the user's past order history and preferences, including the user's past meal selections and their ratings.

[0123] 2. Data storage: Collected data is stored on the server. A database is used to efficiently manage data.

[0124] 3. Model learning: The AI ​​model learns from the collected data on the server. Specifically, the data is preprocessed using Pandas, and the NearestNeighbors model of Scikit-learn is used to learn user preferences.

[0125] 4. Meal Suggestion: Using the learned model, the system suggests optimal meals for the user. The suggestions are displayed on the user's device.

[0126] 5. Customization: Users can customize the suggested meals to suit their tastes. The results of the customization are also saved on the server and reflected in the next suggestions.

[0127] 6. Providing Nutritional Information: Providing nutritional information and calorie calculations for suggested meals to users.

[0128] Examples

[0129] For example, if User A has ordered "sushi" five times, "ramen" four times, and "pizza" three times in the past, the AI ​​will learn from this and suggest the meals User A is likely to order next. If the user inputs "I want to eat healthy today," the system will suggest low-calorie meals accordingly.

[0130] Examples of prompt statements

[0131] "Based on a user's past ordering history, suggest the next meal they're likely to order."

[0132] In this way, by applying an AI system that learns from users' preferences and past selection history to food delivery apps, more personalized meal suggestions can be made, improving user satisfaction and health management.

[0133] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0134] Step 1:

[0135] Data collection

[0136] The user terminal collects the user's past order history and preference data. Specifically, it receives the user's past meal selections and their ratings as input and sends this to the server. The input data includes the type of meal, the rating, the date and time of the order, etc. The server receives this and stores it in a database.

[0137] Step 2:

[0138] Data Storage

[0139] The server stores the received user order history data in a database. The stored data includes the user ID, meal type, rating, order date and time, etc. This allows the past selection history of each user to be accumulated.

[0140] Step 3:

[0141] Model Training

[0142] The server trains an AI model based on the stored data. Specifically, it preprocesses the data using Pandas and learns user preferences using Scikit-learn's NearestNeighbors model. The input data are user IDs and ratings, and the output is a model that shows each user's food preferences.

[0143] Step 4:

[0144] Meal suggestions

[0145] The server uses the trained model to suggest optimal meals to the user. Specifically, it receives a request from the user device and provides input data (such as the user ID and current mood) to the model. The model generates meal suggestions based on this and sends the suggestion results to the user device. The output data is a list of suggested meals.

[0146] Step 5:

[0147] Customization

[0148] The user terminal allows the user to customize the suggested meals. The user modifies the suggested meals to suit his / her preferences and inputs the results into the terminal. The input data is the customized meal content, which is then sent to the server. The server receives it and stores it in a database.

[0149] Step 6:

[0150] Nutritional information provided

[0151] The server calculates the nutritional information and calorie counts of the suggested meals and provides them to the user. Specifically, it receives a list of suggested meals as input, calculates the nutrients and calories of each meal, and sends the calculation results to the user's terminal and displays them to the user. The output data is the nutritional information and calorie details.

[0152] In this way, by applying an AI system that learns from users' preferences and past selection history to food delivery apps, more personalized meal suggestions can be made, improving user satisfaction and health management.

[0153] Example 2

[0154] Next, a second embodiment of the second embodiment will be described. 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."

[0155] Conventional meal selection support systems do not provide suggestions that fully consider the user's preferences or past selection history, making it difficult to increase user satisfaction. In addition, the suggested meals are often customized manually, which is time-consuming for users. Furthermore, there is a lack of means to effectively utilize the user's selection data, making it difficult to accurately predict the meal that the user is likely to choose next.

[0156] 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.

[0157] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for collecting the user's selection data and storing it in a database, a means for analyzing the user's selection history using a generative AI model, a means for predicting the meal that is likely to be selected next based on the analysis results, and a means for presenting the prediction result to the user. This allows meals to be suggested taking into account the user's preferences and past selection history, thereby increasing user satisfaction. In addition, the suggested meals are automatically customized, reducing the user's effort. Furthermore, by effectively utilizing the user's selection data, it is possible to accurately predict the meal that is likely to be selected next.

[0158] "User" refers to an individual who uses the System to make meal selections.

[0159] "Artificial intelligence system" refers to a computer system that learns, analyzes user data, and makes appropriate suggestions.

[0160] "Preferences" refers to the eating patterns and tendencies that a user has chosen in the past.

[0161] "Choice History" refers to data regarding past food choices made by a User.

[0162] "Learning means" refers to the process of analyzing data and finding patterns based on user preferences and selection history.

[0163] "Means for making suggestions" refers to the process of suggesting appropriate meals to the user based on the results of learning.

[0164] "Customization features" refers to the ability to modify suggested meals to suit a user's preferences.

[0165] "Choice Data" refers to information about meals selected by a User through the System.

[0166] "Database" means a system for storing and managing User Selection Data.

[0167] "Generative AI model" refers to the artificial intelligence algorithms used to analyze user data and make predictions.

[0168] "Means of analyzing" refers to the process of using a generative AI model to analyze a user's selection history.

[0169] "Prediction means" refers to the process of estimating the next meal a user is likely to choose based on the analysis results.

[0170] "Presentation means" refers to the process of displaying the prediction results to the user.

[0171] This invention is an artificial intelligence system for assisting users in choosing meals, learning from the user's preferences and past selection history, and making meal suggestions based on the results. The system collects user selection data, stores it in a database, analyzes it using a generative AI model, predicts the next meal that the user is likely to choose, and presents it to the user.

[0172] Hardware and Software Used

[0173] Hardware: Servers, terminals (smartphones and PCs used by users)

[0174] Software: Generative AI models (e.g., GPT-4®), database management systems (e.g., MySQL), web application frameworks (e.g., Django)

[0175] Details of data processing and calculation

[0176] 1. Data collection:

[0177] The user uses the device to make meal selections.

[0178] The terminal transmits the user's selection data to the server.

[0179] The server stores the received data in a database.

[0180] 2. Data Analysis:

[0181] The server retrieves the user's selection history from the database.

[0182] The server uses the generative AI model to analyze the user's selection history.

[0183] A generative AI model learns your preferences based on your selection history.

[0184] 3. Meal suggestions:

[0185] The server uses a generative AI model to predict what meal the user is likely to choose next.

[0186] The server transmits the prediction results to the terminal.

[0187] The device will show meal suggestions to the user.

[0188] Examples

[0189] User selection history: The user has frequently chosen Japanese food in the past.

[0190] Recommendation content: The AI ​​will suggest Japanese dishes that the user is likely to like, such as "sushi" or "tempura."

[0191] Examples of prompt statements

[0192] "Please suggest meals that the user is likely to choose next based on the user's past meal choices. The user has often chosen Japanese food in the past."

[0193] The system supports users in choosing meals by learning their preferences and suggesting the meal they are likely to choose next. Specifically, it uses a generative AI model to analyze user data and suggest appropriate meals, thereby increasing user satisfaction. In addition, the suggested meals are automatically customized, reducing the effort required by the user. Furthermore, by effectively utilizing the user's selection data, it becomes possible to accurately predict the meal they are likely to choose next.

[0194] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0195] Step 1:

[0196] The user selects a meal.

[0197] The user selects a meal using a device (smartphone or PC). For example, the user opens the app and selects "Japanese food."

[0198] Input: User's meal selection (e.g. Japanese food)

[0199] Output: Selection data (e.g. Japanese food)

[0200] Step 2:

[0201] The terminal transmits the selection data to the server.

[0202] The terminal transmits the user's selection data to the server.

[0203] Input: User selection data (e.g. Japanese food)

[0204] Output: The selection data sent to the server.

[0205] Step 3:

[0206] The server stores the selection in a database.

[0207] The server stores the received selection data in a database, including the user ID, the type of meal selected, and the date and time.

[0208] Input: Selection data sent to the server (e.g. User ID: 12345, Selection: Japanese food, Date and time: 2023-10-01 12:00)

[0209] Output: Selection data stored in a database

[0210] Step 4:

[0211] The server retrieves the selection history from the database.

[0212] The server periodically retrieves the user's selection history from the database.

[0213] Input: Selection data stored in the database

[0214] Output: User selection history (e.g. User ID: 12345's selection history)

[0215] Step 5:

[0216] The server analyzes the selection history using a generative AI model.

[0217] The server uses a generative AI model to analyze the user's selection history. The generative AI model learns the user's preferences based on the user's selection history.

[0218] Input: User selection history (e.g. User ID: 12345's past selection history)

[0219] Output: User preference pattern (e.g. User ID: 12345 likes Japanese food)

[0220] Step 6:

[0221] Predict the meal your server is likely to choose next.

[0222] The server uses the generative AI model to predict what meal the user is likely to choose next.

[0223] Input: User preference pattern (e.g. User ID: 12345 likes Japanese food)

[0224] Output: The next most likely meal (e.g. sushi)

[0225] Step 7:

[0226] The server transmits the prediction results to the terminal.

[0227] The server transmits the prediction results to the terminal.

[0228] Input: The next most likely meal (e.g. sushi)

[0229] Output: Prediction results sent to the device

[0230] Step 8:

[0231] The terminal displays meal suggestions to the user.

[0232] The device displays meal suggestions to the user, for example, "How about sushi for your next meal?"

[0233] Input: Prediction result sent to the device (e.g. sushi)

[0234] Output: Dining suggestions displayed to the user (e.g., How about sushi for your next meal?)

[0235] In this way, the system collects and analyzes the user's selection data, and supports the user in choosing meals by suggesting the meal they are likely to choose next.

[0236] (Application example 2)

[0237] Next, a description will be given of Application Example 2 of Form 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."

[0238] Conventional food delivery services lack the ability to suggest meals based on the user's preferences and past selection history, which means that users have to take the trouble of choosing the menu themselves every time. In addition, the suggested meals are not automatically customized based on the user's past selection history or feedback, making it difficult to increase user satisfaction.

[0239] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for learning the user's preferences and past selection history, a means for proposing meals based on the learning results, a means for enabling customization of the proposed meals, and a means for vectorizing the user's past order history and calculating cosine similarity to predict the next meal to be proposed. This makes it possible to propose optimal meals based on the user's preferences and past selection history, and further to automatically customize the proposals.

[0240] "User preferences" are the characteristics of meals that a user is likely to like, inferred based on the user's past meal selections and preferences.

[0241] "Past selection history" is a record of meals previously selected or ordered by a user.

[0242] "Means of learning" refers to the methods and algorithms that allow the AI ​​to analyze data and understand the user's preferences based on the user's preferences and past selection history.

[0243] A "means for providing meal suggestions" is a method or system for presenting to a user the next meal that the user is likely to select based on the results of learning.

[0244] A "means for enabling customization of meal suggestions" is a method or system that allows a user to modify the suggested meals to suit their personal preferences.

[0245] "Vectorization" is the process of representing a user's past order history as numerical data and converting it into a computable form.

[0246] "Cosine similarity" is an index for calculating the similarity between vectorized data, and is a method for measuring the similarity based on the angle between two vectors.

[0247] A "means for predicting the next meal suggestion" is a method or system for predicting the next meal a user is likely to select based on the user's past ordering history and preferences.

[0248] The system for implementing this invention is an AI system that learns the user's preferences and past selection history and suggests meals based on them. A specific embodiment of this system will be described below.

[0249] First, the server collects the user's past order history and vectorizes it using the Python library CountVectorizer, which converts the user's order history into a numerical data format that can be calculated.

[0250] Next, the server calculates the cosine similarity based on the vectorized data. Cosine similarity is a method of measuring similarity based on the angle between two vectors, and is used to calculate the similarity between the user's past order history and the suggested meals.

[0251] The server predicts the next meal the user is likely to choose based on the calculated cosine similarity, and the prediction result is sent to the user's device and suggested to the user.

[0252] Users can customize the suggested meals, taking into account the user's past selections and feedback, resulting in meals that are optimized to the user's preferences.

[0253] For example, if a user has frequently ordered "sushi" and "tempura" in the past, the next menu item similar to "sushi" or "tempura" will be suggested. Users can take these suggestions and further customize them to suit their preferences.

[0254] Example of a prompt to be input to a generative AI model:

[0255] "Predict the next meal suggestion based on the user's past ordering history. The user's ordering history is: "Sushi", "Tempura", and "Ramen". What meal should be suggested next?"

[0256] This system can be implemented on a regular PC or smartphone. The software requires Python and its libraries (scikit-learn, numpy, json). This makes it possible to suggest optimal meals based on the user's preferences and past selection history, and further customize the suggestions automatically.

[0257] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0258] Step 1:

[0259] The server collects the user's past order history. Specifically, it retrieves data on meals the user has ordered in the past from the database. This data includes the order date and time, menu name, quantity, etc. The input is the user ID, and the output is the user's order history data.

[0260] Step 2:

[0261] The server vectorizes the collected order history data. Specifically, it uses Python's CountVectorizer to represent the order history as numerical data. This allows each menu to be represented as a numerical vector. The input is the order history data, and the output is the vectorized data.

[0262] Step 3:

[0263] The server calculates the cosine similarity based on the vectorized data. Specifically, it uses the cosine_similarity function to calculate the similarity between the user's past order history and the suggested meals. The input is the vectorized data, and the output is the similarity score.

[0264] Step 4:

[0265] The server predicts the next meal the user is likely to choose based on the calculated cosine similarity. Specifically, it selects the menu with the highest similarity score. The input is the similarity score, and the output is the predicted menu.

[0266] Step 5:

[0267] The server sends the predicted menu to the user's device. Specifically, it sends the prediction result to the user's smartphone app via API. The input is the predicted menu, and the output is the suggested menu displayed on the user's device.

[0268] Step 6:

[0269] The user customizes the proposed menu by viewing the proposed menu on a smartphone app and modifying it to suit their preferences. The input is the proposed menu, and the output is the customized menu.

[0270] Step 7:

[0271] The server receives the user's customization results and stores them in a database. Specifically, it records the customizations made by the user and reflects them in the next proposal. The input is the customized menu, and the output is the updated user selection history.

[0272] Example 3

[0273] Next, a third embodiment of the third embodiment will be described. 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."

[0274] Conventional meal recommendation systems make suggestions based on the user's preferences and past selection history, but the problem is that the suggested meals are not easy to customize. In addition, when a user tries to customize the suggested meals to suit their preferences, the system may not be able to properly reflect the request. Another issue is that the customization process is complicated, resulting in a poor user experience.

[0275] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0276] In this invention, the server includes a means for receiving a user request, a means for sending a prompt sentence to the generative AI model to generate meal suggestions, a means for sending the generated meal suggestions to a user terminal and displaying them, a means for receiving a user customization request, a means for sending a prompt sentence to the generative AI model again based on the customization request to generate updated meal suggestions, and a means for sending the updated meal suggestions to the user terminal and displaying them, thereby enabling the user to easily customize the suggested meals to suit their preferences.

[0277] The "means for receiving a user request" is a function that enables the user to send a request for meal suggestions from the terminal to the server.

[0278] "Means for sending prompt sentences to the generative AI model to generate meal suggestions" refers to a function by which the server sends appropriate prompt sentences to the generative AI model based on a user request and generates meal suggestions.

[0279] "Means for transmitting and displaying the generated meal suggestions to the user terminal" refers to a function by which the server transmits meal suggestions obtained from the generative AI model to the user terminal and displays them on the terminal.

[0280] The "means for receiving a user's customization request" is a function that enables a user to request customization of a proposed meal, such as adding or deleting ingredients or changing the cooking method.

[0281] "Means for sending a prompt sentence to the generative AI model again based on the customization request to generate updated meal suggestions" refers to a function by which the server receives a user's customization request, sends a prompt sentence to the generative AI model again based on the customization request, and generates updated meal suggestions.

[0282] "Means for transmitting and displaying updated meal suggestions to the user terminal" refers to a function by which the server transmits updated meal suggestions obtained from the generative AI model to the user terminal and displays them on the terminal.

[0283] This invention is an AI system for assisting users in choosing meals, learning the user's preferences and past selection history, making meal suggestions based on the user's preferences and past selection history, and enabling the suggested meals to be customized. A specific embodiment of this system will be described below.

[0284] Hardware and Software Used

[0285] Hardware: Servers, user devices (PCs, smartphones, tablets, etc.)

[0286] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), web application frameworks (e.g. Django)

[0287] System Overview

[0288] 1. Receiving a User Request

[0289] A user uses the device to request meal suggestions.

[0290] Example: User types, "Dinner suggestions please."

[0291] 2. Meal suggestions using generative AI models

[0292] The server sends prompts to a generative AI model (e.g., GPT-4) to generate meal suggestions.

[0293] Example: Prompt: "Please provide dinner suggestions to the user. Offer healthy and balanced menu options."

[0294] 3. Display of suggested meals

[0295] The server transmits the generated meal suggestions to the user terminal and displays them to the user.

[0296] Example: "Dinner suggestion: grilled chicken and vegetable salad."

[0297] 4. Customize your meals

[0298] Users can customize the suggested meals by adding or removing ingredients, changing the cooking method, etc.

[0299] Example: User types, "Add tomatoes to my salad please."

[0300] 5. Reflecting customizations

[0301] The server receives the user's customizations and again sends prompts to the generative AI model to generate updated meal suggestions.

[0302] Example: Prompt: "The user would like to add tomatoes to the suggested salad. Please provide an updated menu."

[0303] 6. View updated meal suggestions

[0304] The server sends the updated meal suggestions to the user's terminal and displays them to the user.

[0305] Example: "Updated dinner suggestion: Grilled chicken and vegetable salad with tomatoes."

[0306] Examples

[0307] Examples of prompt statements

[0308] Initial suggestion: "Give users dinner suggestions. Offer healthy, balanced meals."

[0309] After customization: "The user would like to add tomatoes to the suggested salad. Please provide an updated menu."

[0310] User operation example

[0311] 1. User types, "Can I have some dinner suggestions?"

[0312] 2. The server sends a prompt to the generative AI model, suggesting "grilled chicken and vegetable salad."

[0313] 3. User types, "Add tomatoes to my salad."

[0314] 4. The server sends the updated prompt to the generative AI model, suggesting "grilled chicken and vegetable salad with tomatoes."

[0315] In this way, the user can customize the suggested meals to suit his / her preferences. This system allows the user to receive more satisfying meal suggestions, and improves the efficiency of meal selection. The flow of the identification process in the third embodiment will be described with reference to FIG. 15.

[0316] Step 1:

[0317] Receiving a user request

[0318] A user uses a terminal to request meal suggestions. The user types "Give me dinner suggestions" and presses the send button. The terminal sends this request to the server. The input is the user's request, and the output is the transmission of the request data to the server.

[0319] Step 2:

[0320] Generating meal suggestions with generative AI models

[0321] The server receives the user's request and sends a prompt to the generative AI model. The prompt is "Please give the user dinner suggestions. Please provide healthy and balanced menus." The generative AI model generates meal suggestions based on this prompt. The input is the prompt and the output is the generated meal suggestions.

[0322] Step 3:

[0323] Show suggested meals

[0324] The server receives the generated meal suggestion and sends it to the user terminal. The terminal displays the suggestion to the user. Specifically, the server sends the suggestion "grilled chicken and vegetable salad" to the user terminal, which displays it on the screen. The input is the generated meal suggestion, and the output is the display on the user terminal.

[0325] Step 4:

[0326] Receiving customised meal requests

[0327] A user requests customization of a proposed meal. The user types "add tomatoes to the salad" and presses the submit button. The device sends this customization request to the server. The input is the user's customization request, and the output is the transmission of the request data to the server.

[0328] Step 5:

[0329] Reflecting customizations

[0330] The server receives the user's customization request and sends a prompt sentence to the generative AI model again. The prompt sentence is "The user wants to add tomatoes to the proposed salad. Please provide an updated menu." The generative AI model generates updated meal suggestions based on this prompt sentence. The input is the prompt sentence based on the customization request, and the output is the updated meal suggestions.

[0331] Step 6:

[0332] View Updated Proposals

[0333] The server receives the updated meal suggestions and sends them to the user's device. The device displays the updated suggestions to the user. Specifically, the server sends the updated suggestion "Grilled chicken and vegetable salad with tomatoes" to the user's device, which displays it on its screen. The input is the updated meal suggestions and the output is the display on the user's device.

[0334] (Application example 3)

[0335] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0336] Conventional food delivery services have the problem that it is difficult for users to freely customize the suggested menu, and they are unable to fully meet the user's preferences and needs. In addition, the customization process requires users to manually enter information, which makes ordering less efficient. Furthermore, there was a need for a system that can make more personalized suggestions by utilizing the user's past selection history and feedback.

[0337] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0338] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for the user to input customization content, a means for sending the customization content as a prompt sentence to the generative AI model, a means for receiving a customized menu from the generative AI model, and a means for displaying the customized menu to the user. This allows the user to freely customize the suggested menu and order efficiently. In addition, the server can make more personalized suggestions by utilizing the user's past selection history and feedback.

[0339] "User preferences" refers to information about the user's preferences for meals, ingredients, cooking methods, etc. that the user has previously selected.

[0340] "Past selection history" is a record of the user's previous meal and customization selections.

[0341] A "means of learning" is an algorithm or system that analyzes user preferences and past choices to extract patterns and trends.

[0342] The "means for making meal suggestions" is a function for presenting suitable meal menus to the user based on the results of learning.

[0343] "Means to customize suggested meals" refers to functionality that allows users to add or remove ingredients from suggested menus, change cooking methods, etc.

[0344] The "means for inputting customization details" refers to an interface for inputting the customization details desired by the user into the system.

[0345] A "generative AI model" is an artificial intelligence model that generates customized menus based on user input.

[0346] A "prompt sentence" is a text-based input sentence that instructs the generative AI model on customization details.

[0347] The "means for receiving a customized menu" is a function for receiving a customized menu generated from a generative AI model.

[0348] The "means for displaying the customized menu to the user" refers to an interface for visually presenting the generated customized menu to the user.

[0349] The system for implementing the present invention learns the preferences and past selection history of a user, makes meal suggestions based on the results of the learning, and enables customization of the suggested meals. A specific embodiment of this system is described below.

[0350] System Configuration

[0351] The system includes a user terminal, a server, and a generative AI model. The user terminal is a device such as a smartphone or tablet, and the server is located on the cloud. The generative AI model uses an advanced natural language processing model such as OpenAI's GPT-3.

[0352] Program Processing

[0353] User Device

[0354] The user device provides an interface for inputting the user's preferences and past selection history. Once the user receives meal suggestions, a screen is displayed for inputting customization details. Once the user has input the customization details, they are sent to the server.

[0355] server

[0356] The server has a database for learning user preferences and past selection history, receives the customization content sent by the user and sends it as a prompt to the generative AI model, receives the customized menu from the generative AI model and sends it to the user's device.

[0357] Generative AI Models

[0358] The generative AI model receives the prompt text sent by the server and generates a customized menu, which is returned to the server and finally displayed on the user's device.

[0359] Examples

[0360] For example, if a user is prompted for "pizza," the user can input "add toppings, more cheese." This customization is sent to the generative AI model as a prompt like this:

[0361] Examples of prompt statements

[0362] Menu suggestions: Pizza

[0363] Customizations: Add toppings, increase cheese

[0364] Generate your customized menu.

[0365] The generative AI model receives this prompt and generates a customized menu, for example, "Pizza with extra toppings and extra cheese." This menu is then displayed on the user's device via the server.

[0366] Hardware and Software Used

[0367] User devices: Smartphones, tablets

[0368] Server: Cloud server (e.g. AWS (registered trademark), Google (registered trademark) Cloud)

[0369] Generative AI model: OpenAI GPT-3

[0370] Database: A database to store user preferences and past selections (e.g. MySQL, PostgreSQL)

[0371] In this way, users can customize the menu suggestions to their liking and order efficiently, while also leveraging the user's past selections and feedback to provide more personalized suggestions.

[0372] The flow of the specific process in the application example 3 will be described with reference to FIG.

[0373] Step 1:

[0374] The user uses the device to receive meal suggestions.

[0375] Input: User preferences and past selections

[0376] Data processing: The server analyzes this data and generates a meal menu suitable for the user.

[0377] Output: The suggested meal menu is displayed on the user's device.

[0378] Step 2:

[0379] The user inputs customization details for the proposed menu.

[0380] Input: Suggested meal, user customizations (e.g., add toppings, increase cheese)

[0381] Data processing: The user device receives the customization content and generates data to send to the server.

[0382] Output: The customizations are sent to the server.

[0383] Step 3:

[0384] The server sends the customization content to the generative AI model as a prompt.

[0385] Input: User customizations

[0386] Data processing: The server converts the customization content into a prompt text and sends it to the generative AI model.

[0387] Output: The prompt is sent to the generative AI model.

[0388] Step 4:

[0389] A generative AI model takes the prompt and generates a customized menu.

[0390] Input: prompt statement

[0391] Data processing: The generative AI model analyzes the prompt text and generates a customized menu.

[0392] Output: The customized menu is returned to the server.

[0393] Step 5:

[0394] The server transmits the generated customized menu to the user terminal.

[0395] Input: Customized menu

[0396] Data processing: The server converts the generated menu into data to be sent to the user's terminal.

[0397] Output: The customized menu is sent to the user's device.

[0398] Step 6:

[0399] The user device displays the customized menu to the user.

[0400] Input: Customized menu

[0401] Data processing: The user device converts the received menu into data for visual display.

[0402] Output: The customized menu is displayed to the user.

[0403] Furthermore, an emotion engine that estimates the emotion of the user may be combined. That is, the identification processing unit 290 may estimate the emotion of the user using the emotion identification model 59, and perform identification processing using the emotion of the user.

[0404] "Example 1"

[0405] As one embodiment of the present invention, an AI system incorporating an emotion engine is provided. The system recognizes the user's emotions and makes meal suggestions based on the emotions. Specifically, if the user is expressing emotions of joy or excitement, the system suggests dishes that the user would like, such as sweets or celebratory meals. On the other hand, if the user is expressing emotions of depression or stress, the system suggests dishes that the user would find comforting, such as hot soup or familiar home-cooked meals.

[0406] "Example 2"

[0407] In another embodiment of the present invention, a system is provided in which the emotion engine adjusts the suggested meal customizations. Specifically, if the user is expressing emotions of happiness or excitement, the system may suggest customizations that the user would like, such as adding chocolate sauce or upgrading the dessert. On the other hand, if the user is expressing emotions of depression or stress, the system may suggest customizations that the user would like, such as adding ingredients or adjusting the thickness of the soup.

[0408] "Example 3"

[0409] Further, as another embodiment of the present invention, a system is provided in which an emotion engine simultaneously adjusts meal suggestions and customization based on the user's emotions. Specifically, when a user shows emotions of joy or excitement, the system simultaneously suggests a sweet dish that the user would like, such as adding chocolate sauce, while when a user shows emotions of depression or stress, the system simultaneously suggests a hot soup that the user would find reassuring, such as adding ingredients.

[0410] The process flow of each embodiment will be described below.

[0411] "Example 1"

[0412] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[0413] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[0414] Step 3: Suggested meals are presented to the user and the user selects from them.

[0415] "Example 2"

[0416] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[0417] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[0418] Step 3: Further tailor the suggested meals based on emotions.

[0419] Step 4: Tailored meal suggestions are presented to the user for them to choose from.

[0420] "Example 3"

[0421] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[0422] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[0423] Step 3: Further tailor the suggested meals based on emotions.

[0424] Step 4: Tailored meal suggestions and customization suggestions are presented simultaneously to the user, who can then choose from them.

[0425] Example 1

[0426] Next, a description will be given of Example 1 of the embodiment 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".

[0427] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they cannot suggest meals that take into account the user's emotional state, which means that the system is unable to sufficiently increase user satisfaction. In addition, the suggested meals are often customized manually, which is time-consuming for users.

[0428] 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.

[0429] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for recognizing the user's emotions and suggesting meals based on the emotions, and a means for enabling customization of the suggested meals. This allows meals to be suggested taking into account the user's emotional state, thereby increasing the user's satisfaction. In addition, the suggested meals are automatically customized, reducing the user's effort.

[0430] "User preferences" refers to the user's particular food preferences based on the user's past meal selections and / or meal ratings.

[0431] "Past selection history" refers to data such as the type of meal a user has selected, the date and time, and their rating.

[0432] "Means of learning" refers to machine learning algorithms and databases that analyze users' preferences and past selection history to understand their eating trends.

[0433] "Means for making meal suggestions" refers to algorithms or systems that suggest appropriate meals to users based on the results of their learning.

[0434] "Means of recognizing emotions" refers to emotion analysis engines or software used to analyze a user's facial expressions and voice to identify the user's emotional state.

[0435] "Means enabling customization of meal suggestions" means an interface or system that allows a user to modify the suggested meals to suit their personal preferences.

[0436] "Feedback" refers to ratings and comments that users make on suggested meals.

[0437] This invention is an artificial intelligence system for assisting a user in selecting meals, learning the user's preferences and past selection history, and recognizing the user's emotions to suggest meals. A specific embodiment of this system will be described below.

[0438] System Configuration

[0439] Hardware and Software

[0440] This system consists of a server, a terminal, and a user. The server is equipped with a database and a machine learning algorithm, and the terminal is equipped with an emotion analysis engine. Specific software used is Microsoft (registered trademark) Azure (registered trademark) Emotion API and Google Cloud's Natural Language API.

[0441] Data collection and learning

[0442] User data collection

[0443] A user inputs their daily meal selections into the terminal. For example, if the user chooses pasta for lunch, the user inputs that information into the terminal. The input data includes the type of meal, date and time, and rating (e.g., "delicious").

[0444] Accumulating and learning data

[0445] The server receives the user's meal data sent from the device and stores it in a database. The server uses the stored data to train a machine learning model. This model learns the user's eating habits and uses them to make future suggestions. For example, if the user tends to frequently choose pasta, the server learns that information.

[0446] Emotion recognition and meal suggestions

[0447] Emotion recognition

[0448] The device is equipped with an emotion engine to recognize the user's emotions. When the user speaks to the device, the device analyzes the user's facial expressions and voice to identify the user's emotional state. For example, if the user is smiling while speaking, the device recognizes that the user is expressing the emotion of happiness.

[0449] Meal suggestions

[0450] The server generates appropriate meal suggestions based on the emotion data sent from the device and the user's past selection history. For example, if the user expresses joy, the server will suggest sweets and celebratory meals that the user likes based on past data. Specifically, chocolate cakes and fruit tarts that the user has given high ratings in the past are suggested.

[0451] Examples

[0452] Consider a case where a user has been successful at work and is expressing emotions of joy. The user speaks to the device, saying, "I'm so happy today!" The device analyzes the user's facial expressions and voice to recognize the emotion of joy. The recognition results are sent to the server. The server then suggests meals that the user might like based on the received emotion data and past selection history. For example, it might suggest chocolate cake or fruit tart that the user has given a high rating in the past. The suggestions are sent to the device and displayed to the user.

[0453] Examples of prompt statements

[0454] Examples of prompts to be input to a generative AI model include:

[0455] "What are some examples of meals you would suggest when the user is expressing the emotion of pleasure? The user's past selections include chocolate cake and fruit tart."

[0456] Using this prompt, the generative AI model can generate appropriate meal suggestions based on the user's emotions and past selection history.

[0457] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0458] Step 1: Collect user data

[0459] A user inputs their daily meal selections into the terminal. For example, if the user chooses pasta for lunch, the user inputs that information into the terminal. The input data includes the type of meal, date and time, and rating (e.g., "delicious").

[0460] Input: Meal type, date, time, rating

[0461] Output: Meal data stored on the device

[0462] Step 2: Sending and storing data

[0463] The terminal transmits the collected meal data of the user to the server.

[0464] The server stores the received data in a database.

[0465] Input: Meal data sent from the device

[0466] Output: Meal data stored in a database

[0467] Step 3: Train the data

[0468] The server uses the user's food data stored in the database to train a machine learning model, which learns the user's eating habits and uses them to make future suggestions.

[0469] Input: Meal data stored in a database

[0470] Output: A trained machine learning model

[0471] Step 4: Recognize emotions

[0472] When a user speaks to the device, the device analyzes the user's facial expressions and voice to identify the user's emotional state. For example, if the user is smiling while speaking, the device recognizes that the user is expressing the emotion of joy. The emotion engine uses, for example, the Emotion API and Natural Language API.

[0473] Input: User's facial expression, voice

[0474] Output: Recognized emotion data

[0475] Step 5: Sending emotion data

[0476] The terminal transmits the recognized emotion data to the server.

[0477] Input: Recognized emotion data

[0478] Output: Emotion data sent to the server

[0479] Step 6: Meal Suggestions

[0480] The server generates appropriate meal suggestions based on the emotion data sent from the device and the user's past selection history. For example, if the user expresses joy, the server will suggest sweets and celebratory meals that the user likes based on past data.

[0481] Input: Emotion data, past selection history

[0482] Output: Suggested meals

[0483] Step 7: Viewing suggestions

[0484] The terminal displays the suggestions sent by the server to the user, who can then review the suggested meals and customize them if necessary.

[0485] Input: Suggested meals

[0486] Output: The suggestions displayed to the user

[0487] Examples

[0488] Consider a case where a user has been successful at work and is expressing emotions of joy. The user speaks to the device, saying, "I'm so happy today!" The device analyzes the user's facial expressions and voice to recognize the emotion of joy. The recognition results are sent to the server. The server then suggests meals that the user might like based on the received emotion data and past selection history. For example, it might suggest chocolate cake or fruit tart that the user has given a high rating in the past. The suggestions are sent to the device and displayed to the user.

[0489] (Application example 1)

[0490] Next, a description will be given of Application Example 1 of Form 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."

[0491] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they have the problem that it is difficult to provide meals that match the user's current mood because they cannot make suggestions that take into account the user's emotional state. In addition, there is a problem that it is not possible to improve user satisfaction because there is a lack of means to recognize the user's emotions in real time and suggest appropriate meals based on them.

[0492] 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.

[0493] In this invention, the server includes means for learning the user's preferences and past selection history, means for suggesting meals based on the results of the learning, means for enabling customization of the suggested meals, means for recognizing the user's emotions and suggesting meals based on the emotions, and means for recognizing the user's emotions in real time using the smartphone camera, thereby enabling meal suggestions according to the user's current emotional state and improving user satisfaction.

[0494] "User preferences" refers to the meals selected by the user in the past and the user's particular preferences for meals.

[0495] "Past selection history" refers to a record of meals previously selected by a user.

[0496] "Means of learning" refers to the function of accumulating data on user preferences and past selection history, and analyzing this to find patterns.

[0497] "Means for making meal suggestions" refers to a function that suggests suitable meals to the user based on the results of learning.

[0498] "Means enabling customization of meal suggestions" refers to functionality that allows a user to modify the suggested meals to suit their personal preferences.

[0499] "Means of recognizing the user's emotions" refers to the function of analyzing the user's emotions from their facial expressions, voice, etc., and determining their state.

[0500] "Means for making meal suggestions based on emotions" refers to a function that suggests appropriate meals based on the recognized emotional state of the user.

[0501] "Means of recognizing a user's emotions in real time using a smartphone camera" refers to a function that uses a smartphone camera to analyze a user's facial expressions and determine their emotions in real time.

[0502] As an embodiment of the present invention, we will specifically explain an AI system for assisting users in choosing meals. This system learns the user's preferences and past selection history, and incorporates an emotion engine to make meal suggestions based on the user's emotions.

[0503] System Configuration

[0504] The system consists of the following main components:

[0505] 1. User device: A mobile information terminal such as a smartphone equipped with a camera and microphone.

[0506] 2. Server: Stores user data and provides computing resources to run AI models.

[0507] 3. AI model: A generative AI model that learns user preferences and past selection history and recognizes emotions.

[0508] Hardware and Software Used

[0509] Hardware: Smartphone camera, microphone, server

[0510] Software: OpenCV (camera capture and face detection), Keras (loading and predicting emotion recognition models), JSON (saving and loading the user's past selection history)

[0511] Data processing and calculation

[0512] 1. User device:

[0513] It uses a smartphone camera to capture the user's face in real time.

[0514] Faces are detected using OpenCV and pre-processed for input into the emotion recognition model.

[0515] We use Keras to run an emotion recognition model and predict user emotions in real time.

[0516] 2. Server:

[0517] User preferences and past selection history are stored in JSON format and loaded as needed.

[0518] It runs an algorithm to suggest suitable meals based on the user's emotional state.

[0519] Examples

[0520] For example, when a user is using the app through the smartphone camera, the app detects the user's face and recognizes emotions. If the user is recognized as "Happy," the app will suggest "favorite_sweets" from the user's past selection history.

[0521] Examples of prompt statements

[0522] Develop an AI system that recognizes the user's emotions and suggests meals based on those emotions. Create a smartphone app that learns the user's past selection history, incorporates an emotion engine, and has the ability to recognize emotions in real time and suggest appropriate meals.

[0523] In this way, meal suggestions can be made according to the user's current emotional state, improving user satisfaction.

[0524] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0525] Step 1:

[0526] The user device captures the user's face in real time using the smartphone camera.

[0527] Input: Smartphone camera image

[0528] Output: A captured image of the user's face.

[0529] Specific operation: The smartphone camera is activated and captures an image of the user's face.

[0530] Step 2:

[0531] The user device uses OpenCV to detect faces from captured facial images and performs preprocessing for input into the emotion recognition model.

[0532] Input: A captured image of the user's face

[0533] Output: Preprocessed face image

[0534] Specific operation: Face detection is performed using OpenCV, the detected face part is converted to grayscale, and resized to a size suitable for the emotion recognition model.

[0535] Step 3:

[0536] The user device runs an emotion recognition model using Keras to predict the user's emotions in real time.

[0537] Input: Preprocessed face image

[0538] Output: Predicted user emotion label

[0539] Specific operation: The preprocessed facial image is input into the emotion recognition model, and the model predicts the emotion and outputs the emotion label.

[0540] Step 4:

[0541] The server stores user preferences and past selection history in JSON format and loads them as needed.

[0542] Input: User's past selection history data

[0543] Output: Saved JSON data

[0544] Specific behavior: Save the user's past selection history in a JSON format file and load the file when needed.

[0545] Step 5:

[0546] The server runs an algorithm to suggest appropriate meals based on the user's emotional state.

[0547] Input: Predicted user emotion label, user's past choice history data

[0548] Output: Suggested meals

[0549] How it works: Based on emotion labels and past selection history, the algorithm selects and suggests meals suitable for the user.

[0550] Step 6:

[0551] The user device displays the suggested meals to the user.

[0552] Input: Suggested meals

[0553] Output: Meal suggestions displayed to the user

[0554] Specific behavior: Suggested meals are displayed on the smartphone screen for the user to review.

[0555] Example 2

[0556] Next, a second embodiment of the second embodiment will be described. 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."

[0557] Conventional meal recommendation systems can suggest meals based on the user's past selection history, but they have a problem in that they cannot customize the suggestions based on the user's emotional state. This makes it difficult to improve user satisfaction because the system cannot suggest optimal meals according to the user's current emotions.

[0558] 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.

[0559] In this invention, the server includes means for learning the user's preferences and past selection history, means for making meal suggestions based on the learning results, and means for analyzing the user's emotional state and adjusting the customization of the suggested meals, thereby enabling automatic meal customization based on the user's past selection history and emotional data.

[0560] An "artificial intelligence system" is a computer system that learns from user data and makes predictions and suggestions.

[0561] "User preferences" refers to information about the meals and tastes selected by the user in the past.

[0562] "Past selection history" is a record of meals the user has previously chosen.

[0563] "Learning tools" refers to the algorithms and models that analyze user data and find patterns and trends.

[0564] "Means for making meal suggestions" is a function that presents suitable meals to the user based on the results of learning.

[0565] "Means for analyzing emotional state" refers to algorithms or models used to analyze a user's emotional data and determine their state.

[0566] "Means for adjusting customization" refers to the ability to change the content and presentation of a meal depending on the user's emotional state.

[0567] "Emotional data" is information about emotions expressed by a user, such as happiness, excitement, depression, stress, etc.

[0568] "Feedback" means any opinion or evaluation information provided by a User.

[0569] This invention is an artificial intelligence system for assisting a user in selecting meals by learning the user's preferences and past selection history, making meal suggestions based on the results of that learning, and further adjusting the customization of the suggested meals by analyzing the user's emotional state.

[0570] Hardware and Software Used

[0571] Hardware: Servers, terminals (smartphones, tablets, PCs)

[0572] Software: Generative AI models, emotion engine, database management system (DBMS)

[0573] Specific operation of the system

[0574] Data collection

[0575] The user uses the terminal to select a meal. The terminal sends the user's selected meal data to the server. The server stores the received data in a database. Specifically, it stores information such as the user ID, the type of meal selected, and the date and time of selection. If the user inputs emotion data when selecting a meal, the terminal also sends that emotion data to the server. The emotion data is expressed in categories such as "happiness," "excitement," "depression," and "stress."

[0576] Data Learning

[0577] The server periodically retrieves the user's selection history and emotion data from the database. The server uses the retrieved data to train a generative AI model. Specifically, the server uses the user's past selection history and emotion data as input to build a model that predicts the meal that is likely to be selected next. The server saves the trained generative AI model and uses it for the next suggestion.

[0578] Meal suggestions

[0579] When the user uses the device again to select a meal, the device sends a request to the server. The server receives the request and uses the generative AI model to predict which meal would suit the user's preferences. The server sends the prediction result to the device. Specifically, if the user has often selected Japanese food in the past, the server will suggest options such as "sushi" and "tempura" from among the Japanese dishes. The device will display the suggested meals to the user.

[0580] Emotion-Based Customization

[0581] When a user inputs emotion data when selecting a meal, the terminal transmits the emotion data to the server. The server uses an emotion engine to analyze the user's emotion state. The server generates customization suggestions based on the analysis result. For example, if the user indicates the emotion of "joy", the server suggests adding chocolate sauce or upgrading the dessert. The server transmits the customization suggestions to the terminal. The terminal displays the suggested customization to the user.

[0582] Examples

[0583] User's selection history: The user has previously selected Japanese food 10 times and Western food 5 times.

[0584] Emotional data: The user indicates the emotion of “happiness” when choosing a meal.

[0585] Examples of prompt statements

[0586] The user's past selection history is Japanese food 10 times and Western food 5 times. Currently, the user is showing the emotion of "delight". What meal and customization would you suggest for this user next?

[0587] In this way, the system suggests optimal meals and customizations based on the user's selection history and emotional data.

[0588] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0589] Step 1: Data collection

[0590] The user uses the terminal to select a meal. The terminal sends the user's selected meal data (input) to the server. The server stores the received data in a database (output). Specifically, it stores information such as the user ID, the type of meal selected, and the date and time of selection. If the user inputs emotion data when selecting a meal, the terminal also sends that emotion data to the server. The emotion data is expressed in categories such as "happiness," "excitement," "depression," and "stress."

[0591] Step 2: Data training

[0592] The server periodically retrieves the user's selection history and emotion data from the database (input). The server uses the retrieved data to train a generative AI model (data processing). Specifically, the server uses the user's past selection history and emotion data as input to build a model that predicts the meal that will likely be selected next (output). The server saves the trained generative AI model and uses it for the next suggestion.

[0593] Step 3: Meal Suggestions

[0594] When the user uses the device again to select a meal, the device sends a request to the server (input). The server receives the request and uses the generative AI model to predict which meal would suit the user's preferences (data calculation). The server sends the prediction result to the device (output). Specifically, if the user has often selected Japanese food in the past, the server will suggest options such as "sushi" and "tempura" from among the Japanese dishes. The device will display the suggested meals to the user.

[0595] Step 4: Emotional customization

[0596] When a user inputs emotion data when selecting a meal, the device sends the emotion data to the server (input). The server uses an emotion engine to analyze the user's emotional state (data calculation). The server generates customization suggestions based on the analysis results (output). For example, if the user indicates the emotion of "joy", it suggests adding chocolate sauce or upgrading the dessert. The server sends the customization suggestions to the device. The device displays the suggested customizations to the user.

[0597] In this way, the system suggests optimal meals and customizations based on the user's selection history and emotional data.

[0598] (Application example 2)

[0599] Next, a description will be given of Application Example 2 of Form 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."

[0600] Conventional food delivery services can suggest meals based on a user's past selection history and preferences, but they have the problem of not being able to provide customized suggestions that take into account the user's emotional state. This has left them lacking a means to further improve user satisfaction.

[0601] 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.

[0602] In this invention, the server includes means for learning the user's preferences and past selection history, means for suggesting meals based on the learning results, means for enabling customization of the suggested meals, means for detecting the user's emotional state, and means for suggesting meal customization based on the detected emotional state, thereby enabling meal suggestions and customization according to the user's preferences and emotional state.

[0603] "User Preferences" refers to a User's preferences for certain meals and ingredients based on the User's past meal selections and preferences.

[0604] "Past selection history" refers to a record of a user's previous meal selections and orders.

[0605] "Means of learning" refers to algorithms and systems that analyze user preferences and past choices to extract patterns and trends.

[0606] "Means for providing meal suggestions" refers to an algorithm or system that uses the results of learning to recommend suitable meals to users.

[0607] "Means enabling customization of suggested meals" refers to functionality that allows a user to add to or change suggested meals.

[0608] "Means for detecting a user's emotional state" refers to technologies or systems that analyze a user's facial expressions and behavior to identify their current emotional state (happiness, excitement, depression, stress, etc.).

[0609] "Means for suggesting meal customization based on detected emotional state" refers to an algorithm or system for suggesting meal content and customization options based on the user's emotional state.

[0610] The system for implementing the present invention learns the user's preferences and past selection history to provide meal suggestions, and also detects the user's emotional state to provide customized suggestions. A specific embodiment of this system will be described below.

[0611] Hardware and Software Used

[0612] Hardware: Smartphone, camera (for emotion detection)

[0613] Software: TensorFlow (AI model), OpenCV (emotion detection), Flask (backend), React Native (frontend)

[0614] Data processing and calculation

[0615] Learn user preferences and history

[0616] The server uses TensorFlow to learn from the user's past order history data, which allows it to analyze the user's preferences and selection trends and predict the meal they are likely to choose next.

[0617] Emotion Detection

[0618] The device (smartphone) uses a camera to analyze the user's face in real time, and uses OpenCV to detect the user's emotional state (happiness, excitement, depression, stress, etc.) from their facial expressions.

[0619] Customization suggestions

[0620] The server then uses the results of the emotion engine to make customized meal suggestions, such as upgrading the dessert if the user is happy, or adjusting the soup strength if the user is depressed.

[0621] Examples

[0622] User A: Has often ordered Japanese food in the past. Currently, the emotion detected by the camera is "Happiness."

[0623] Recommendation: Suggest sushi as part of the Japanese menu and also suggest matcha ice cream as an additional dessert.

[0624] User B: Often ordered Western food in the past. Currently, the emotion detected by the camera is "stress."

[0625] Suggestion: Suggest pasta from the Western menu, with an additional suggestion to adjust the soup thickness.

[0626] Examples of prompt statements

[0627] Suggest the next meal that the user is likely to choose based on their past order history. Detect the user's emotional state with a camera and provide customized suggestions based on their emotions.

[0628] In this way, a system can be realized that allows for meal suggestions and customization based on the user's preferences and emotional state.

[0629] The flow of the identification process in the application example 2 will be described with reference to FIG.

[0630] Step 1:

[0631] The server receives the user's past order history data. The input includes the user ID and the past order history data. The server uses TensorFlow to learn from this data and analyze the user's preferences and selection trends. As an output, a list of meal suggestions based on the user's preferences is generated.

[0632] Step 2:

[0633] The device (smartphone) uses a camera to capture a picture of the user's face in real time. The input includes the user's facial image data. The device uses OpenCV to analyze this image data and detect the user's emotional state. As an output, the user's emotional state (e.g., happy, excited, depressed, stressed, etc.) is identified.

[0634] Step 3:

[0635] The server receives the meal suggestion list generated in step 1 and the user's emotional state detected in step 2. As input, it includes the meal suggestion list and the emotional state data. The server uses an emotion engine to provide customized meal suggestions based on the emotional state. As output, a customized meal suggestion list is generated.

[0636] Step 4:

[0637] The terminal displays the customized meal suggestion list received from the server to the user. As input, it contains the customized meal suggestion list. The terminal displays the meal suggestions on a user interface using React Native. As output, the user can see the suggested meals.

[0638] Step 5:

[0639] The user selects a desired meal from a customized meal suggestion list displayed on the terminal. As input, the customized meal suggestion list is included. The user confirms the selected meal and places the order. As output, order data for the selected meal is generated.

[0640] Step 6:

[0641] The server receives order data from the user and sends the order to the food delivery service. As input, the server includes the user's order data. The server sends the order data to the food delivery service's system and confirms the order. As output, order confirmation data is generated.

[0642] In this way, meal suggestions and customizations are made based on the user's preferences and emotional state, allowing the user to select and order meals that satisfy them.

[0643] Example 3

[0644] Next, a third embodiment of the third embodiment will be described. 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."

[0645] Conventional meal recommendation systems make suggestions based on the user's preferences and past selection history, but they have the problem of not being able to customize or make suggestions that take into account the user's emotional state. This makes it difficult to make meal suggestions that suit the user's current emotional state, making it difficult to increase user satisfaction.

[0646] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0647] In this invention, the server includes means for learning the user's preferences and past selection history, means for making meal suggestions based on the learning results, means for enabling customization of the suggested meals, means for analyzing the user's emotional state, and means for adjusting the meal suggestions and customization based on the analyzed emotional state, thereby enabling meal suggestions and customization according to the user's emotional state.

[0648] "Means of learning about the user's preferences and past selection history" refers to a function for understanding the user's tastes by collecting and analyzing data on the meals and preferences selected by the user in the past.

[0649] "Means of suggesting meals based on learned results" is a function for suggesting appropriate meals based on collected user preferences and past selection history.

[0650] "Means for customizing suggested meals" refers to functionality that allows a user to add or remove ingredients from suggested meals, change cooking methods, etc.

[0651] "Means for analyzing a user's emotional state" refers to a function for analyzing and understanding a user's current emotional state from their input and actions.

[0652] "Means for tailoring meal suggestions and customization based on analyzed emotional state" refers to functionality for tailoring meal suggestions and customization options based on the analyzed emotional state of the user.

[0653] The present invention relates to an artificial intelligence system for assisting a user in selecting meals. A specific embodiment of the system will be described below.

[0654] The server uses a database to learn the user's preferences and past selection history. The database stores the user's past meal selections and preferences. The server analyzes this data and runs algorithms to understand the user's tastes. The algorithms use machine learning techniques.

[0655] Next, the server uses the generative AI model to make meal suggestions based on the learned results. For example, a model using natural language processing technology (e.g., GPT-4) is used as the generative AI model. The server receives the user's input data and generates a prompt sentence for the generative AI model. An example of a prompt sentence is "The user says he wants to eat pizza. Please recommend a pizza." The generative AI model generates appropriate meal suggestions based on this prompt sentence.

[0656] In addition, the server provides a user interface to enable customization of the suggested meals. The user can use the terminal to add or remove ingredients from the suggested meals, change the cooking method, etc. For example, if the user inputs "I want to add toppings to my pizza," the server receives this customization information and again uses the generative AI model to generate a customized meal suggestion.

[0657] The server also uses an emotion engine to analyze the user's emotional state. The emotion engine executes algorithms to analyze and understand the user's current emotional state from the user's inputs and actions. For example, if the user inputs "I'm tired today, please suggest a meal that will help me relax," the emotion engine analyzes the emotional state as "tired."

[0658] The server adjusts meal suggestions and customization based on the analyzed emotional state. For example, if the user is tired, the server will suggest "hot soup" and simultaneously suggest adding ingredients (e.g., chicken and vegetables). This allows for meal suggestions and customization according to the user's emotional state.

[0659] As a specific example, if a user inputs, "I'm tired today, please suggest a meal that will help me relax," the following actions will be taken:

[0660] 1. The user types into the device, "I'm tired today, so please suggest some meals that will help me relax."

[0661] 2. The terminal sends this input data to the server.

[0662] 3. The server analyzes the input data and generates a prompt saying, "The user is tired. Please suggest a meal that will help him relax."

[0663] 4. The server uses the generative AI model to suggest "hot soup."

[0664] 5. The server uses the emotion engine to analyze the user’s emotional state as “tired.”

[0665] 6. The server offers to add extra ingredients (e.g. chicken or vegetables) in addition to the "hot soup."

[0666] 7. The server sends this proposal to the device.

[0667] 8. The device displays to the user, "Hot soup and add toppings (chicken and / or vegetables)."

[0668] 9. The user enters a customization: "I would like to add chicken to the soup."

[0669] 10. The terminal sends this customization information to the server.

[0670] 11. The server generates a customized "hot chicken soup" and sends it to the terminal.

[0671] 12. The device displays the final “Hot Chicken Soup” to the user.

[0672] In this way, a system is realized in which the server, the terminal, and the user work together to suggest and customize meals. The flow of the specification process in the third embodiment will be described with reference to FIG.

[0673] Step 1:

[0674] A user enters a meal request.

[0675] A user uses the terminal interface to input a meal request, for example, "I'd like pizza." The input data includes the user's request.

[0676] Step 2:

[0677] The terminal transmits the user's input data to the server.

[0678] The terminal sends the user's input data to the server. The input data includes the user's request. The output is the user's request data sent to the server.

[0679] Step 3:

[0680] The server analyzes the user's input data and generates a prompt for the generative AI model.

[0681] The server analyzes the user's input data and generates a prompt sentence suitable for the generative AI model. For example, it generates a prompt sentence such as "The user says he wants to eat pizza. Please recommend some pizzas." The input is the user's request data, and the output is the generated prompt sentence.

[0682] Step 4:

[0683] The server uses the generative AI model to generate meal suggestions.

[0684] The server uses a generative AI model (e.g., GPT-4) to generate meal suggestions based on the prompt sentence, e.g., suggest "Margherita Pizza." The input is the generated prompt sentence, and the output is the generated meal suggestions.

[0685] Step 5:

[0686] The server uses an emotion engine to analyze the user's emotional state.

[0687] The server uses the emotion engine to analyze the user's emotional state. For example, if the user inputs "I'm tired," the emotion engine analyzes the emotional state "tired." The input is data about the user's emotions, and the output is the analyzed emotional state.

[0688] Step 6:

[0689] Servers tailor meal suggestions and customizations based on emotional state.

[0690] The server adjusts the meal suggestions and customization based on the emotional state. For example, if the user is tired, the server may suggest "hot soup" and simultaneously suggest adding ingredients (e.g., chicken and vegetables). The inputs are the parsed emotional state and the generated meal suggestions, and the outputs are the adjusted meal suggestions and customization options.

[0691] Step 7:

[0692] The server sends the final proposal to the device.

[0693] The server sends the final meal suggestions to the terminal. The inputs are the adjusted meal suggestions and the customization options, and the output is the final suggestions sent to the terminal.

[0694] Step 8:

[0695] The terminal displays the suggestions to the user.

[0696] The terminal displays the meal suggestions received from the server to the user. The input is the final suggestions sent by the server and the output is the suggestions displayed to the user.

[0697] Step 9:

[0698] The user customizes the suggestions.

[0699] The user customizes the proposed meal by adding or deleting ingredients, changing the cooking method, etc. For example, the user inputs "I would like to add toppings to the pizza." The input is the user's customization request, and the output is the customization information.

[0700] Step 10:

[0701] The terminal transmits the customization information to the server.

[0702] The terminal transmits the user's customization information to the server, where the input is the user's customization information and the output is the customization information transmitted to the server.

[0703] Step 11:

[0704] The server generates customized meal suggestions and transmits them to the terminal.

[0705] The server generates customized meal suggestions based on the user's customization information and sends them to the terminal, where the input is the user's customization information and the output is the customized meal suggestions.

[0706] Step 12:

[0707] The terminal displays the final customization proposal to the user.

[0708] The terminal displays the final customized proposal received from the server to the user. The input is the customized meal proposal sent from the server, and the output is the final customized proposal displayed to the user.

[0709] (Application example 3)

[0710] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0711] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they have the problem of being unable to customize or suggest meals that take into account the user's emotional state. This means that meal suggestions that are appropriate for the user's current emotional state cannot be made, and user satisfaction cannot be sufficiently increased.

[0712] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0713] In this invention, the server includes means for learning the user's preferences and past selection history, means for making meal suggestions based on the learning results, means for enabling customization of the suggested meals, means for analyzing the user's emotions, and means for adjusting the meal suggestions and customization based on the analyzed emotions, thereby enabling personalized meal suggestions and customization according to the user's emotional state.

[0714] "User preferences" are dietary trends identified by a user based on past dietary choices and preferences.

[0715] "Past selection history" is a record of meals previously selected by the user.

[0716] A "means of learning" is an algorithm or system that analyzes user preferences and past choices to extract patterns.

[0717] The "means for making meal suggestions" is a function for presenting suitable meals to the user based on the results of learning.

[0718] "Means to customise" refers to functionality that allows users to add or remove ingredients from suggested meals, change cooking methods, etc.

[0719] "Means of analyzing emotions" refers to technologies or systems for reading emotions from a user's facial expressions, voice, etc.

[0720] "Means for tailoring meal suggestions and customization based on emotions" refers to a function that changes meal suggestions and customization options in response to the analyzed emotions.

[0721] The system for implementing the present invention learns the user's preferences and past selection history, analyzes the user's emotions, and proposes and customizes meals. A specific embodiment of this system will be described below.

[0722] System Configuration

[0723] The system consists of a user terminal, a server, and an emotion analysis engine. The user terminal is a device such as a smartphone or tablet, and the server is located on the cloud. The emotion analysis engine is software for analyzing the user's facial expressions and voice.

[0724] Program Processing

[0725] The server uses a database to learn the user's preferences and past selection history. Every time the user selects a meal, the information is stored in the database and analyzed using machine learning algorithms (e.g., Keras and TensorFlow), which allows the server to learn the user's preferences and selection patterns.

[0726] The user device uses a camera and microphone to capture the user's facial expressions and voice, and then sends the data to the sentiment analysis engine, which uses OpenCV and deep learning models to analyze the user's emotions. The analysis results are sent to the server, which then makes meal suggestions and customizations based on the results.

[0727] Examples

[0728] For example, consider a scenario where a user opens a smartphone app, takes a picture of their face with their camera, and uploads it. The app analyzes the user's facial expressions and detects, for example, an emotion of "happiness." As a result, the server suggests "chocolate cake" and "add chocolate sauce."

[0729] Example of a prompt to be input to a generative AI model:

[0730] The user opens the smartphone app, takes a picture of their face with their camera and uploads it. The app analyzes the user's facial expressions and detects the emotion of joy. As a result, the app suggests adding chocolate cake and chocolate sauce.

[0731] In this way, personalized meal suggestions and customization based on the user's emotional state is possible.

[0732] The flow of the specific process in the application example 3 will be described with reference to FIG.

[0733] Step 1:

[0734] Users open the smartphone app, take a photo of their face with their camera, and upload it.

[0735] Input: User's face image

[0736] Output: Face image data

[0737] What it does: A user takes a picture of their face using their smartphone camera and uploads the image to the app.

[0738] Step 2:

[0739] The device sends the uploaded facial image to an emotion analysis engine.

[0740] Input: Face image data

[0741] Output: Emotion analysis results

[0742] Specific operation: The device sends facial image data to the emotion analysis engine, which then uses OpenCV and deep learning models to analyze the user's emotions.

[0743] Step 3:

[0744] The emotion analysis engine analyzes the user's emotions and sends the results to the server.

[0745] Input: Face image data

[0746] Output: Emotion label (e.g. happy, excited, depressed, stressed)

[0747] How it works: The emotion analysis engine analyzes facial image data to identify the user's emotions. The analysis results are sent to the server as emotion labels.

[0748] Step 4:

[0749] The server receives the user's emotion labels and makes meal suggestions based on them.

[0750] Input: emotion label

[0751] Output: Meal suggestions (e.g. chocolate cake, pizza, hot soup, herbal tea)

[0752] Specific behavior: The server receives the emotion labels and suggests appropriate meals based on pre-defined rules and databases.

[0753] Step 5:

[0754] The server references the user's past selection history to customize the suggested meals.

[0755] Input: Meal suggestions, past selection history

[0756] Output: Customized meal suggestions (e.g. chocolate cake + add chocolate sauce)

[0757] Specific behavior: The server retrieves the user's past selection history from the database and makes appropriate customizations to the suggested meals.

[0758] Step 6:

[0759] The server transmits the customized meal suggestions to the user terminal.

[0760] Input: Customized Meal Suggestions

[0761] Output: Notification to user device

[0762] Specific operation: The server sends the customized meal suggestions to the user terminal and notifies the user.

[0763] Step 7:

[0764] The user reviews the suggested meals and further customizes them if necessary.

[0765] Input: Customized Meal Suggestions

[0766] Output: Final meal selection

[0767] Specific behavior: The user can review the suggested meals on the device, add or remove ingredients, change the cooking method, etc. The final meal selection is confirmed.

[0768] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice 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 voice data.

[0769] 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<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to 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.

[0770] Another example of generative AI is Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0771] In the above embodiment, an example was given in which the specific process was performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0772] [Second embodiment]

[0773] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0774] 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.

[0775] 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 wide area network (WAN) and / or a local area network (LAN).

[0776] 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.

[0777] 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 the voice according to instructions from the processor 46.

[0778] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0779] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0780] Fig. 4 shows an example of 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.

[0781] 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.

[0782] The storage 32 stores a data generation model 58 and an emotion identification model 59 .

[0783] The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0784] 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.

[0785] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0786] "Example 1"

[0787] As one embodiment of the present invention, an AI system that supports a user in selecting meals is provided. This system has a function to learn the user's preferences and past selection history. Specifically, the system accumulates data on the meals the user has selected in the past and the user's tendency toward preferred meals, and the AI ​​learns from this data.

[0788] "Example 2"

[0789] In addition, the AI ​​also has the ability to suggest meals based on the results of its learning. Specifically, it suggests meals that the user is likely to choose next based on the user's preferences and past selection history that it has learned. For example, if the user has often chosen Japanese food in the past, the AI ​​will suggest Japanese food that the user is likely to like.

[0790] "Example 3"

[0791] It also has a function that allows users to customize the suggested meals. Specifically, users can customize the suggested meals by adding or removing ingredients, changing the cooking method, etc. For example, users can add toppings or increase the amount of cheese to the pizza suggested by the AI.

[0792] The process flow of each embodiment will be described below.

[0793] "Example 1"

[0794] Step 1: When a user begins using the system, the system begins recording the user's food choices, including the types of food the user chooses, the time of meal, and frequency of meals.

[0795] Step 2: The system accumulates this data and the AI ​​learns about the user's preferences and choices. This learning process evolves every time the user uses the system, providing more accurate results.

[0796] "Example 2"

[0797] Step 1: After the AI ​​learns the user's preferences and selection trends, it will suggest the next meal the user is likely to choose. These suggestions will be displayed when the user logs into the system, when it is time to eat, etc.

[0798] Step 2: The user can choose to either select the suggested meal or choose a different meal. If the user selects the suggested meal, that information is again fed back into the system to help train the AI.

[0799] "Example 3"

[0800] Step 1: After the user selects a suggested meal, they can then customize it by adding or removing ingredients, changing the cooking method, etc.

[0801] Step 2: The user's customization information is also fed back to the system and used to train the AI, allowing the system to better understand the user's preferences and provide more personalized meal suggestions.

[0802] Example 1

[0803] Next, a description will be given of Example 1 of Form 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".

[0804] Conventional meal selection support systems had the problem that they did not make suggestions that fully considered the user's preferences or past selection history, resulting in low user satisfaction. In addition, the suggested meals were customized manually, which was a hassle for the user. Furthermore, the system lacked the functionality to learn the user's selection trends and recommend appropriate meals, making it difficult to suggest meals that matched the user's preferences.

[0805] 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.

[0806] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for proposing meals based on the results of the learning, a means for enabling customization of the proposed meals, a means for collecting the user's meal data, a means for accumulating the collected data, a means for learning the user's selection tendency based on the accumulated data, a means for generating recommended meals based on the learned model, and a means for notifying the user of the generated recommended meals. This enables meal suggestions that take into account the user's preferences and past selection history, thereby improving user satisfaction. In addition, the proposed meals are automatically customized, reducing the user's effort. Furthermore, by learning the user's selection tendency and recommending appropriate meals, meal suggestions that match the user's preferences can be realized.

[0807] The "means for learning the user's preferences and past selection history" refers to a device or program that has the function of collecting data on the meals selected by the user in the past and analyzing the user's preferences and selection trends based on that data.

[0808] A "means for making meal suggestions based on the results of learning" is a device or program that has the function of recommending appropriate meals based on the user's preferences and past selection history.

[0809] A "means for enabling customization of meal suggestions" is a device or program that has the functionality to allow a user to modify the suggested meals to suit their personal preferences.

[0810] A "means for collecting user meal data" is a device or program that has the function of inputting information about meals selected by the user and collecting it as data.

[0811] The "means for storing collected data" refers to a device or program that has the function of storing the collected user's dietary data in a storage device such as a database.

[0812] The "means for learning the user's selection tendencies based on accumulated data" refers to a device or program that has the function of analyzing accumulated meal data and modeling the user's selection tendencies.

[0813] A "means for generating recommended meals based on a learned model" is a device or program that has the function of generating appropriate meals based on the learned selection tendencies of a user.

[0814] The "means for notifying the user of the generated recommended meals" refers to a device or program having a function of transmitting the generated recommended meals to the user's terminal and notifying the user of the recommended meals.

[0815] This invention is an AI system for assisting users in choosing meals, learning from the user's preferences and past selection history and proposing appropriate meals. A specific embodiment of this system will be described below.

[0816] First, the user inputs meal information using the terminal. The terminal then sends the input meal information (e.g., meal name, date and time, rating) to the server. The server receives the meal information sent from the terminal and stores it in a database. General database software such as MySQL is used as the database management system.

[0817] Next, the server learns the user's selection trends based on the accumulated data. TensorFlow is used as the machine learning library. The server uses TensorFlow to model the user's food selection patterns and analyze the user's preferences and selection trends.

[0818] Once the learning is complete, the server generates meal recommendations based on the learned model. Using the generative AI model, it suggests three meals that match the user's preferences. For example, if the user previously selected "curry rice," "ramen," and "sushi," the server will recommend "tempura," "udon," and "sushi."

[0819] Finally, the server sends the generated recommended meals to the user's device. The notification service used is Firebase Cloud Messaging. The device notifies the user of the received recommended meals and displays them on the screen.

[0820] As a specific example, a user selects "curry rice" for lunch and enters "curry rice" into an app on the device. The device sends the entered meal information "curry rice" to the server. The server stores the received "curry rice" data in a MySQL database. Next, the server uses TensorFlow to learn that the user tends to prefer "Japanese food." Based on the learned model, the server recommends "tempura," "udon," and "sushi" and notifies the user's device of this recommendation.

[0821] An example of a prompt is, "Generate the next meal recommendation based on the user's past meal selection data. Suggest three specific meal names taking into account the user's preferences and selection trends."

[0822] This is how the AI ​​system that assists users in making meal choices works.

[0823] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0824] Step 1:

[0825] Collecting user dietary data

[0826] The user selects a meal and enters the information into the terminal.

[0827] Input: Meal information such as meal name, date and time, and rating.

[0828] The terminal transmits the input meal information to the server.

[0829] Output: Meal information sent to the server.

[0830] Specific actions: The user selects "curry rice" for lunch and enters "curry rice" into the device app.

[0831] Step 2:

[0832] Accumulation of data

[0833] The server receives the meal information transmitted from the terminal.

[0834] Input: Meal information sent from the device.

[0835] The server stores the received data in a database.

[0836] Output: Meal information stored in a database.

[0837] Specific operation: The server stores the received "curry rice" data in a MySQL database.

[0838] Step 3:

[0839] Training the Data

[0840] The server learns the user's preferences and selection trends based on the meal information accumulated in the database.

[0841] Input: Food information stored in a database.

[0842] The server uses machine learning libraries (e.g., TensorFlow) to model users' food choice patterns.

[0843] Output: The trained model.

[0844] Specific behavior: The server uses TensorFlow to learn that users tend to like "Japanese food."

[0845] Step 4:

[0846] Generation of recommended meals

[0847] The server generates meal recommendations for the user based on the learned model.

[0848] Input: The trained model.

[0849] The server uses a generative AI model to suggest three meals that match the user's preferences.

[0850] Output: A recommended diet list.

[0851] Specific operation: The server recommends "tempura," "udon," and "sushi" based on the learned model.

[0852] Step 5:

[0853] Dietary recommendations

[0854] The server transmits the generated recommended meals to the user's terminal.

[0855] Input: A recommended meal list.

[0856] The device will notify the user of the recommended meals it has received.

[0857] Output: The meal recommendation communicated to the user.

[0858] Specific operation: The server sends the generated recommended meals, "tempura," "udon," and "sushi," to the user's terminal, and the terminal notifies the user of them.

[0859] (Application example 1)

[0860] Next, a description will be given of Application Example 1 of Form 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".

[0861] Conventional food delivery systems do not adequately suggest meals based on the user's preferences and past selection history. In addition, they lack the ability to suggest meals according to the user's current mood and health status, and provide nutritional information and calorie calculations for meals, which results in low user satisfaction and poor health management. This makes it difficult for users to select appropriate meals.

[0862] 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.

[0863] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for customizing the meal suggestions based on the user's current mood and health condition, and a means for providing nutritional information and calorie calculations for the meals. This allows optimal meal suggestions based on the user's preferences and health condition, improving the user's satisfaction and health management.

[0864] "User preferences" refers to the meals selected by the user in the past and the user's particular preferences for meals.

[0865] "Past selection history" refers to the record of meals the user has selected in the past.

[0866] "Means of learning" refers to how the AI ​​analyzes data and identifies patterns and trends based on user preferences and past choices.

[0867] "Means for providing meal suggestions" refers to a method for recommending appropriate meals to a user based on the results of learning.

[0868] "Means enabling customization of meal suggestions" refers to a method that allows a user to modify the suggested meals to suit their personal preferences.

[0869] "Means for customizing meal suggestions based on current mood and health status" refers to a method for tailoring meal suggestions to a user's current mood and health status.

[0870] "Means for providing nutritional information and calorie calculations for a meal" means a method for calculating and providing to a user the nutrients and calories contained in a suggested meal.

[0871] As an embodiment of the present invention, an AI system that assists users in choosing meals will be specifically described. This system learns the user's preferences and past selection history, and suggests optimal meals. The configuration and operation of the system will be described in detail below.

[0872] System Configuration

[0873] The system consists of the following main components:

[0874] 1. User device: A device operated by a user, such as a smartphone or tablet.

[0875] 2. Server: The central system that processes and stores data.

[0876] 3. AI model: A machine learning model that learns from user preferences and past selection history to provide meal suggestions.

[0877] Hardware and Software Used

[0878] Hardware: Smartphones, tablets, servers

[0879] Software: Python, Pandas, Scikit-learn

[0880] Data processing and calculation

[0881] 1. Data collection: Collect data from the user's device about the user's past order history and preferences, including the user's past meal selections and their ratings.

[0882] 2. Data storage: Collected data is stored on the server. A database is used to efficiently manage data.

[0883] 3. Model learning: The AI ​​model learns from the collected data on the server. Specifically, the data is preprocessed using Pandas, and the NearestNeighbors model of Scikit-learn is used to learn user preferences.

[0884] 4. Meal Suggestion: Using the learned model, the system suggests optimal meals for the user. The suggestions are displayed on the user's device.

[0885] 5. Customization: Users can customize the suggested meals to suit their tastes. The results of the customization are also saved on the server and reflected in the next suggestions.

[0886] 6. Providing Nutritional Information: Providing nutritional information and calorie calculations for suggested meals to users.

[0887] Examples

[0888] For example, if User A has ordered "sushi" five times, "ramen" four times, and "pizza" three times in the past, the AI ​​will learn from this and suggest the meals User A is likely to order next. If the user inputs "I want to eat healthy today," the system will suggest low-calorie meals accordingly.

[0889] Examples of prompt statements

[0890] "Based on a user's past ordering history, suggest the next meal they're likely to order."

[0891] In this way, by applying an AI system that learns from users' preferences and past selection history to food delivery apps, more personalized meal suggestions can be made, improving user satisfaction and health management.

[0892] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0893] Step 1:

[0894] Data collection

[0895] The user terminal collects the user's past order history and preference data. Specifically, it receives the user's past meal selections and their ratings as input and sends this to the server. The input data includes the type of meal, the rating, the date and time of the order, etc. The server receives this and stores it in a database.

[0896] Step 2:

[0897] Data Storage

[0898] The server stores the received user order history data in a database. The stored data includes the user ID, meal type, rating, order date and time, etc. This allows the past selection history of each user to be accumulated.

[0899] Step 3:

[0900] Model Training

[0901] The server trains an AI model based on the stored data. Specifically, it preprocesses the data using Pandas and learns user preferences using Scikit-learn's NearestNeighbors model. The input data are user IDs and ratings, and the output is a model that shows each user's food preferences.

[0902] Step 4:

[0903] Meal suggestions

[0904] The server uses the trained model to suggest optimal meals to the user. Specifically, it receives a request from the user device and provides input data (such as the user ID and current mood) to the model. The model generates meal suggestions based on this and sends the suggestion results to the user device. The output data is a list of suggested meals.

[0905] Step 5:

[0906] Customization

[0907] The user terminal allows the user to customize the suggested meals. The user modifies the suggested meals to suit his / her preferences and inputs the results into the terminal. The input data is the customized meal content, which is then sent to the server. The server receives it and stores it in a database.

[0908] Step 6:

[0909] Nutritional information provided

[0910] The server calculates the nutritional information and calorie counts of the suggested meals and provides them to the user. Specifically, it receives a list of suggested meals as input, calculates the nutrients and calories of each meal, and sends the calculation results to the user's terminal and displays them to the user. The output data is the nutritional information and calorie details.

[0911] In this way, by applying an AI system that learns from users' preferences and past selection history to food delivery apps, more personalized meal suggestions can be made, improving user satisfaction and health management.

[0912] Example 2

[0913] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[0914] Conventional meal selection support systems do not provide suggestions that fully consider the user's preferences or past selection history, making it difficult to increase user satisfaction. In addition, the suggested meals are often customized manually, which is time-consuming for users. Furthermore, there is a lack of means to effectively utilize the user's selection data, making it difficult to accurately predict the meal that the user is likely to choose next.

[0915] 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.

[0916] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for collecting the user's selection data and storing it in a database, a means for analyzing the user's selection history using a generative AI model, a means for predicting the meal that is likely to be selected next based on the analysis results, and a means for presenting the prediction result to the user. This allows meals to be suggested taking into account the user's preferences and past selection history, thereby increasing user satisfaction. In addition, the suggested meals are automatically customized, reducing the user's effort. Furthermore, by effectively utilizing the user's selection data, it is possible to accurately predict the meal that is likely to be selected next.

[0917] "User" refers to an individual who uses the System to make meal selections.

[0918] "Artificial intelligence system" refers to a computer system that learns, analyzes user data, and makes appropriate suggestions.

[0919] "Preferences" refers to the eating patterns and tendencies that a user has chosen in the past.

[0920] "Choice History" refers to data regarding past food choices made by a User.

[0921] "Learning means" refers to the process of analyzing data and finding patterns based on user preferences and selection history.

[0922] "Means for making suggestions" refers to the process of suggesting appropriate meals to the user based on the results of learning.

[0923] "Customization features" refers to the ability to modify suggested meals to suit a user's preferences.

[0924] "Choice Data" refers to information about meals selected by a User through the System.

[0925] "Database" means a system for storing and managing User Selection Data.

[0926] "Generative AI model" refers to the artificial intelligence algorithms used to analyze user data and make predictions.

[0927] "Means of analyzing" refers to the process of using a generative AI model to analyze a user's selection history.

[0928] "Prediction means" refers to the process of estimating the next meal a user is likely to choose based on the analysis results.

[0929] "Presentation means" refers to the process of displaying the prediction results to the user.

[0930] This invention is an artificial intelligence system for assisting users in choosing meals, learning from the user's preferences and past selection history, and making meal suggestions based on the results. The system collects user selection data, stores it in a database, analyzes it using a generative AI model, predicts the next meal that the user is likely to choose, and presents it to the user.

[0931] Hardware and Software Used

[0932] Hardware: Servers, terminals (smartphones and PCs used by users)

[0933] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), web application frameworks (e.g. Django)

[0934] Details of data processing and calculation

[0935] 1. Data collection:

[0936] The user uses the device to make meal selections.

[0937] The terminal transmits the user's selection data to the server.

[0938] The server stores the received data in a database.

[0939] 2. Data Analysis:

[0940] The server retrieves the user's selection history from the database.

[0941] The server uses the generative AI model to analyze the user's selection history.

[0942] A generative AI model learns your preferences based on your selection history.

[0943] 3. Meal suggestions:

[0944] The server uses a generative AI model to predict what meal the user is likely to choose next.

[0945] The server transmits the prediction results to the terminal.

[0946] The device will show meal suggestions to the user.

[0947] Examples

[0948] User selection history: The user has frequently chosen Japanese food in the past.

[0949] Recommendation content: The AI ​​will suggest Japanese dishes that the user is likely to like, such as "sushi" or "tempura."

[0950] Examples of prompt statements

[0951] "Please suggest meals that the user is likely to choose next based on the user's past meal choices. The user has often chosen Japanese food in the past."

[0952] The system supports users in choosing meals by learning their preferences and suggesting the meal they are likely to choose next. Specifically, it uses a generative AI model to analyze user data and suggest appropriate meals, thereby increasing user satisfaction. In addition, the suggested meals are automatically customized, reducing the effort required by the user. Furthermore, by effectively utilizing the user's selection data, it becomes possible to accurately predict the meal they are likely to choose next.

[0953] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0954] Step 1:

[0955] The user selects a meal.

[0956] The user selects a meal using a device (smartphone or PC). For example, the user opens the app and selects "Japanese food."

[0957] Input: User's meal selection (e.g. Japanese food)

[0958] Output: Selection data (e.g. Japanese food)

[0959] Step 2:

[0960] The terminal transmits the selection data to the server.

[0961] The terminal transmits the user's selection data to the server.

[0962] Input: User selection data (e.g. Japanese food)

[0963] Output: Selection data sent to the server

[0964] Step 3:

[0965] The server stores the selection in a database.

[0966] The server stores the received selection data in a database, including the user ID, the type of meal selected, and the date and time.

[0967] Input: Selection data sent to the server (e.g. User ID: 12345, Selection: Japanese food, Date and time: 2023-10-01 12:00)

[0968] Output: Selection data stored in a database

[0969] Step 4:

[0970] The server retrieves the selection history from the database.

[0971] The server periodically retrieves the user's selection history from the database.

[0972] Input: Selection data stored in the database

[0973] Output: User selection history (e.g. User ID: 12345's selection history)

[0974] Step 5:

[0975] The server analyzes the selection history using a generative AI model.

[0976] The server uses a generative AI model to analyze the user's selection history. The generative AI model learns the user's preferences based on the user's selection history.

[0977] Input: User selection history (e.g. User ID: 12345's past selection history)

[0978] Output: User preference pattern (e.g. User ID: 12345 likes Japanese food)

[0979] Step 6:

[0980] Predict the meal your server is likely to choose next.

[0981] The server uses the generative AI model to predict what meal the user is likely to choose next.

[0982] Input: User preference pattern (e.g. User ID: 12345 likes Japanese food)

[0983] Output: The next most likely meal (e.g. sushi)

[0984] Step 7:

[0985] The server transmits the prediction results to the terminal.

[0986] The server transmits the prediction results to the terminal.

[0987] Input: The next most likely meal (e.g. sushi)

[0988] Output: Prediction results sent to the device

[0989] Step 8:

[0990] The terminal displays meal suggestions to the user.

[0991] The device displays meal suggestions to the user, for example, "How about sushi for your next meal?"

[0992] Input: Prediction result sent to the device (e.g. sushi)

[0993] Output: Dining suggestions displayed to the user (e.g., How about sushi for your next meal?)

[0994] In this way, the system collects and analyzes the user's selection data, and supports the user in choosing meals by suggesting the meal they are likely to choose next.

[0995] (Application example 2)

[0996] Next, a description will be given of Application Example 2 of Form 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".

[0997] Conventional food delivery services lack the ability to suggest meals based on the user's preferences and past selection history, which means that users have to take the trouble of choosing the menu themselves every time. In addition, the suggested meals are not automatically customized based on the user's past selection history or feedback, making it difficult to increase user satisfaction.

[0998] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for learning the user's preferences and past selection history, a means for proposing meals based on the learning results, a means for enabling customization of the proposed meals, and a means for vectorizing the user's past order history and calculating cosine similarity to predict the next meal to be proposed. This makes it possible to propose optimal meals based on the user's preferences and past selection history, and further to automatically customize the proposals.

[0999] "User preferences" are the characteristics of meals that a user is likely to like, inferred based on the user's past meal selections and preferences.

[1000] "Past selection history" is a record of meals previously selected or ordered by a user.

[1001] "Means of learning" refers to the methods and algorithms that allow the AI ​​to analyze data and understand the user's preferences based on the user's preferences and past selection history.

[1002] A "means for providing meal suggestions" is a method or system for presenting to a user the next meal that the user is likely to select based on the results of learning.

[1003] A "means for enabling customization of meal suggestions" is a method or system that allows a user to modify the suggested meals to suit their personal preferences.

[1004] "Vectorization" is the process of representing a user's past order history as numerical data and converting it into a computable form.

[1005] "Cosine similarity" is an index for calculating the similarity between vectorized data, and is a method for measuring the similarity based on the angle between two vectors.

[1006] A "means for predicting the next meal suggestion" is a method or system for predicting the next meal a user is likely to select based on the user's past ordering history and preferences.

[1007] The system for implementing this invention is an AI system that learns the user's preferences and past selection history and suggests meals based on them. A specific embodiment of this system will be described below.

[1008] First, the server collects the user's past order history and vectorizes it using the Python library CountVectorizer, which converts the user's order history into a numerical data format that can be calculated.

[1009] Next, the server calculates the cosine similarity based on the vectorized data. Cosine similarity is a method of measuring similarity based on the angle between two vectors, and is used to calculate the similarity between the user's past order history and the suggested meals.

[1010] The server predicts the next meal the user is likely to choose based on the calculated cosine similarity, and the prediction result is sent to the user's device and suggested to the user.

[1011] Users can customize the suggested meals, taking into account the user's past selections and feedback, resulting in meals that are optimized to the user's preferences.

[1012] For example, if a user has frequently ordered "sushi" and "tempura" in the past, the next menu item similar to "sushi" or "tempura" will be suggested. Users can take these suggestions and further customize them to suit their preferences.

[1013] Example of a prompt to be input to a generative AI model:

[1014] "Predict the next meal suggestion based on the user's past ordering history. The user's ordering history is: "Sushi", "Tempura", and "Ramen". What meal should be suggested next?"

[1015] This system can be implemented on a regular PC or smartphone. The software requires Python and its libraries (scikit-learn, numpy, json). This makes it possible to suggest optimal meals based on the user's preferences and past selection history, and further customize the suggestions automatically.

[1016] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1017] Step 1:

[1018] The server collects the user's past order history. Specifically, it retrieves data on meals the user has ordered in the past from the database. This data includes the order date and time, menu name, quantity, etc. The input is the user ID, and the output is the user's order history data.

[1019] Step 2:

[1020] The server vectorizes the collected order history data. Specifically, it uses Python's CountVectorizer to represent the order history as numerical data. This allows each menu to be represented as a numerical vector. The input is the order history data, and the output is the vectorized data.

[1021] Step 3:

[1022] The server calculates the cosine similarity based on the vectorized data. Specifically, it uses the cosine_similarity function to calculate the similarity between the user's past order history and the suggested meals. The input is the vectorized data, and the output is the similarity score.

[1023] Step 4:

[1024] The server predicts the next meal the user is likely to choose based on the calculated cosine similarity. Specifically, it selects the menu with the highest similarity score. The input is the similarity score, and the output is the predicted menu.

[1025] Step 5:

[1026] The server sends the predicted menu to the user's device. Specifically, it sends the prediction result to the user's smartphone app via API. The input is the predicted menu, and the output is the suggested menu displayed on the user's device.

[1027] Step 6:

[1028] The user customizes the proposed menu by viewing the proposed menu on a smartphone app and modifying it to suit their preferences. The input is the proposed menu, and the output is the customized menu.

[1029] Step 7:

[1030] The server receives the user's customization results and stores them in a database. Specifically, it records the customizations made by the user and reflects them in the next proposal. The input is the customized menu, and the output is the updated user selection history.

[1031] Example 3

[1032] Next, a third embodiment of the third embodiment will be described. 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".

[1033] Conventional meal recommendation systems make suggestions based on the user's preferences and past selection history, but the problem is that the suggested meals are not easy to customize. In addition, when a user tries to customize the suggested meals to suit their preferences, the system may not be able to properly reflect the request. Another issue is that the customization process is complicated, resulting in a poor user experience.

[1034] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1035] In this invention, the server includes a means for receiving a user request, a means for sending a prompt sentence to the generative AI model to generate meal suggestions, a means for sending the generated meal suggestions to a user terminal and displaying them, a means for receiving a user customization request, a means for sending a prompt sentence to the generative AI model again based on the customization request to generate updated meal suggestions, and a means for sending the updated meal suggestions to the user terminal and displaying them, thereby enabling the user to easily customize the suggested meals to suit their preferences.

[1036] The "means for receiving a user request" is a function that enables the user to send a request for meal suggestions from the terminal to the server.

[1037] "Means for sending prompt sentences to the generative AI model to generate meal suggestions" refers to a function by which the server sends appropriate prompt sentences to the generative AI model based on a user request and generates meal suggestions.

[1038] "Means for transmitting and displaying the generated meal suggestions to the user terminal" refers to a function by which the server transmits meal suggestions obtained from the generative AI model to the user terminal and displays them on the terminal.

[1039] The "means for receiving a user's customization request" is a function that enables a user to request customization of a proposed meal, such as adding or deleting ingredients or changing the cooking method.

[1040] "Means for sending a prompt sentence to the regenerative AI model based on the customization request to generate updated meal suggestions" refers to a function by which the server receives a user's customization request, sends a prompt sentence to the regenerative AI model based on the customization request, and generates updated meal suggestions.

[1041] "Means for transmitting and displaying updated meal suggestions to the user terminal" refers to a function by which the server transmits updated meal suggestions obtained from the generative AI model to the user terminal and displays them on the terminal.

[1042] This invention is an AI system for assisting users in choosing meals, learning the user's preferences and past selection history, making meal suggestions based on the user's preferences and past selection history, and enabling the suggested meals to be customized. A specific embodiment of this system will be described below.

[1043] Hardware and Software Used

[1044] Hardware: Servers, user devices (PCs, smartphones, tablets, etc.)

[1045] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), web application frameworks (e.g. Django)

[1046] System Overview

[1047] 1. Receiving a User Request

[1048] A user uses the device to request meal suggestions.

[1049] Example: User types, "Dinner suggestions please."

[1050] 2. Meal suggestions using generative AI models

[1051] The server sends prompts to a generative AI model (e.g., GPT-4) to generate meal suggestions.

[1052] Example: Prompt: "Please provide dinner suggestions to the user. Offer healthy and balanced menu options."

[1053] 3. Display of suggested meals

[1054] The server transmits the generated meal suggestions to the user terminal and displays them to the user.

[1055] Example: "Dinner suggestion: grilled chicken and vegetable salad."

[1056] 4. Customize your meals

[1057] Users can customize the suggested meals by adding or removing ingredients, changing the cooking method, etc.

[1058] Example: User types, "Add tomatoes to my salad please."

[1059] 5. Reflecting customizations

[1060] The server receives the user's customizations and again sends prompts to the generative AI model to generate updated meal suggestions.

[1061] Example: Prompt: "The user would like to add tomatoes to the suggested salad. Please provide an updated menu."

[1062] 6. View updated meal suggestions

[1063] The server sends the updated meal suggestions to the user's terminal and displays them to the user.

[1064] Example: "Updated dinner suggestion: Grilled chicken and vegetable salad with tomatoes."

[1065] Examples

[1066] Examples of prompt statements

[1067] Initial suggestion: "Give users dinner suggestions. Offer healthy, balanced meals."

[1068] After customization: "The user would like to add tomatoes to the suggested salad. Please provide an updated menu."

[1069] User operation example

[1070] 1. User types, "Can I have some dinner suggestions?"

[1071] 2. The server sends a prompt to the generative AI model, suggesting "grilled chicken and vegetable salad."

[1072] 3. User types, "Add tomatoes to my salad."

[1073] 4. The server sends the updated prompt to the generative AI model, suggesting "grilled chicken and vegetable salad with tomatoes."

[1074] In this way, the user can customize the suggested meals to suit his / her preferences. This system allows the user to receive more satisfying meal suggestions, and improves the efficiency of meal selection. The flow of the identification process in the third embodiment will be described with reference to FIG. 15.

[1075] Step 1:

[1076] Receiving a user request

[1077] A user uses a terminal to request meal suggestions. The user types "Give me dinner suggestions" and presses the send button. The terminal sends this request to the server. The input is the user's request, and the output is the transmission of the request data to the server.

[1078] Step 2:

[1079] Generating meal suggestions with generative AI models

[1080] The server receives the user's request and sends a prompt to the generative AI model. The prompt is "Please give the user dinner suggestions. Please provide healthy and balanced menus." The generative AI model generates meal suggestions based on this prompt. The input is the prompt and the output is the generated meal suggestions.

[1081] Step 3:

[1082] Show suggested meals

[1083] The server receives the generated meal suggestion and sends it to the user terminal. The terminal displays the suggestion to the user. Specifically, the server sends the suggestion "grilled chicken and vegetable salad" to the user terminal, which displays it on the screen. The input is the generated meal suggestion, and the output is the display on the user terminal.

[1084] Step 4:

[1085] Receiving customised meal requests

[1086] A user requests customization of a proposed meal. The user types "add tomatoes to the salad" and presses the submit button. The device sends this customization request to the server. The input is the user's customization request, and the output is the transmission of the request data to the server.

[1087] Step 5:

[1088] Reflecting customizations

[1089] The server receives the user's customization request and sends a prompt sentence to the generative AI model again. The prompt sentence is "The user wants to add tomatoes to the proposed salad. Please provide an updated menu." The generative AI model generates updated meal suggestions based on this prompt sentence. The input is the prompt sentence based on the customization request, and the output is the updated meal suggestions.

[1090] Step 6:

[1091] View Updated Proposals

[1092] The server receives the updated meal suggestions and sends them to the user's device. The device displays the updated suggestions to the user. Specifically, the server sends the updated suggestion "Grilled chicken and vegetable salad with tomatoes" to the user's device, which displays it on its screen. The input is the updated meal suggestions and the output is the display on the user's device.

[1093] (Application example 3)

[1094] Next, a description will be given of Application Example 3 of Form Example 3. 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".

[1095] Conventional food delivery services have the problem that it is difficult for users to freely customize the suggested menu, and they are unable to fully meet the user's preferences and needs. In addition, the customization process requires users to manually enter information, which makes ordering less efficient. Furthermore, there was a need for a system that can make more personalized suggestions by utilizing the user's past selection history and feedback.

[1096] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1097] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for the user to input customization content, a means for sending the customization content as a prompt sentence to the generative AI model, a means for receiving a customized menu from the generative AI model, and a means for displaying the customized menu to the user. This allows the user to freely customize the suggested menu and order efficiently. In addition, the server can make more personalized suggestions by utilizing the user's past selection history and feedback.

[1098] "User preferences" refers to information about the user's preferences for meals, ingredients, cooking methods, etc. that the user has previously selected.

[1099] "Past selection history" is a record of the user's previous meal and customization selections.

[1100] A "means of learning" is an algorithm or system that analyzes user preferences and past choices to extract patterns and trends.

[1101] The "means for making meal suggestions" is a function for presenting suitable meal menus to the user based on the results of learning.

[1102] "Means to customize suggested meals" refers to functionality that allows users to add or remove ingredients from suggested menus, change cooking methods, etc.

[1103] The "means for inputting customization details" refers to an interface for inputting the customization details desired by the user into the system.

[1104] A "generative AI model" is an artificial intelligence model that generates customized menus based on user input.

[1105] A "prompt sentence" is a text-based input sentence that instructs the generative AI model on customization details.

[1106] The "means for receiving a customized menu" is a function for receiving a customized menu generated from a generative AI model.

[1107] The "means for displaying the customized menu to the user" refers to an interface for visually presenting the generated customized menu to the user.

[1108] The system for implementing the present invention learns the user's preferences and past selection history, makes meal suggestions based on the results of the learning, and enables customization of the suggested meals. A specific embodiment of this system is described below.

[1109] System Configuration

[1110] The system includes a user terminal, a server, and a generative AI model. The user terminal is a device such as a smartphone or tablet, and the server is located on the cloud. The generative AI model uses an advanced natural language processing model such as OpenAI's GPT-3.

[1111] Program Processing

[1112] User Device

[1113] The user device provides an interface for inputting the user's preferences and past selection history. Once the user receives meal suggestions, a screen is displayed for inputting customization details. Once the user has input the customization details, they are sent to the server.

[1114] server

[1115] The server has a database for learning user preferences and past selection history, receives the customization content sent by the user and sends it as a prompt to the generative AI model, receives the customized menu from the generative AI model and sends it to the user's device.

[1116] Generative AI Models

[1117] The generative AI model receives the prompt text sent by the server and generates a customized menu, which is returned to the server and finally displayed on the user's device.

[1118] Examples

[1119] For example, if a user is prompted for "pizza," the user can input "add toppings, more cheese." This customization is sent to the generative AI model as a prompt like this:

[1120] Examples of prompt statements

[1121] Menu suggestions: Pizza

[1122] Customizations: Add toppings, increase cheese

[1123] Generate your customized menu.

[1124] The generative AI model receives this prompt and generates a customized menu, for example, "Pizza with extra toppings and extra cheese." This menu is then displayed on the user's device via the server.

[1125] Hardware and Software Used

[1126] User devices: Smartphones, tablets

[1127] Server: Cloud server (e.g. AWS, Google Cloud)

[1128] Generative AI model: OpenAI GPT-3

[1129] Database: A database to store user preferences and past selections (e.g. MySQL, PostgreSQL)

[1130] In this way, users can customize the menu suggestions to their liking and order efficiently, while also leveraging the user's past selections and feedback to provide more personalized suggestions.

[1131] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1132] Step 1:

[1133] The user uses the device to receive meal suggestions.

[1134] Input: User preferences and past selections

[1135] Data processing: The server analyzes this data and generates a meal menu suitable for the user.

[1136] Output: The suggested meal menu is displayed on the user's device.

[1137] Step 2:

[1138] The user inputs customization details for the proposed menu.

[1139] Input: Suggested meal, user customizations (e.g., add toppings, increase cheese)

[1140] Data processing: The user device receives the customization content and generates data to send to the server.

[1141] Output: The customizations are sent to the server.

[1142] Step 3:

[1143] The server sends the customization content to the generative AI model as a prompt.

[1144] Input: User customizations

[1145] Data processing: The server converts the customization content into a prompt text and sends it to the generative AI model.

[1146] Output: The prompt is sent to the generative AI model.

[1147] Step 4:

[1148] A generative AI model takes the prompt and generates a customized menu.

[1149] Input: prompt statement

[1150] Data processing: The generative AI model analyzes the prompt text and generates a customized menu.

[1151] Output: The customized menu is returned to the server.

[1152] Step 5:

[1153] The server transmits the generated customized menu to the user terminal.

[1154] Input: Customized menu

[1155] Data processing: The server converts the generated menu into data to be sent to the user's terminal.

[1156] Output: The customized menu is sent to the user's device.

[1157] Step 6:

[1158] The user device displays the customized menu to the user.

[1159] Input: Customized menu

[1160] Data processing: The user device converts the received menu into data for visual display.

[1161] Output: The customized menu is displayed to the user.

[1162] In addition, 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.

[1163] "Example 1"

[1164] As one embodiment of the present invention, an AI system incorporating an emotion engine is provided. The system recognizes the user's emotions and makes meal suggestions based on the emotions. Specifically, if the user is expressing emotions of joy or excitement, the system suggests dishes that the user would like, such as sweets or celebratory meals. On the other hand, if the user is expressing emotions of depression or stress, the system suggests dishes that the user would find comforting, such as hot soup or familiar home-cooked meals.

[1165] "Example 2"

[1166] In another embodiment of the present invention, a system is provided in which the emotion engine adjusts the suggested meal customizations. Specifically, if the user is expressing emotions of happiness or excitement, the system may suggest customizations that the user would like, such as adding chocolate sauce or upgrading the dessert. On the other hand, if the user is expressing emotions of depression or stress, the system may suggest customizations that the user would like, such as adding ingredients or adjusting the thickness of the soup.

[1167] "Example 3"

[1168] Further, as another embodiment of the present invention, a system is provided in which an emotion engine simultaneously adjusts meal suggestions and customization based on the user's emotions. Specifically, when a user shows emotions of joy or excitement, the system simultaneously suggests a sweet dish that the user would like, such as adding chocolate sauce, while when a user shows emotions of depression or stress, the system simultaneously suggests a hot soup that the user would find reassuring, such as adding ingredients.

[1169] The process flow of each embodiment will be described below.

[1170] "Example 1"

[1171] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[1172] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[1173] Step 3: Suggested meals are presented to the user and the user selects from them.

[1174] "Example 2"

[1175] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[1176] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[1177] Step 3: Further tailor the suggested meals based on emotions.

[1178] Step 4: Tailored meal suggestions are presented to the user for them to choose from.

[1179] "Example 3"

[1180] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[1181] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[1182] Step 3: Further tailor the suggested meals based on emotions.

[1183] Step 4: Tailored meal suggestions and customization suggestions are presented simultaneously to the user, who can then choose from them.

[1184] Example 1

[1185] Next, a description will be given of Example 1 of Form 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".

[1186] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they cannot suggest meals that take into account the user's emotional state, which means that the system is unable to sufficiently increase user satisfaction. In addition, the suggested meals are often customized manually, which is time-consuming for users.

[1187] 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.

[1188] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for recognizing the user's emotions and suggesting meals based on the emotions, and a means for enabling customization of the suggested meals. This allows meals to be suggested taking into account the user's emotional state, thereby increasing the user's satisfaction. In addition, the suggested meals are automatically customized, reducing the user's effort.

[1189] "User preferences" refers to the user's particular food preferences based on the user's past meal selections and / or meal ratings.

[1190] "Past selection history" refers to data such as the type of meal a user has selected, the date and time, and their rating.

[1191] "Means of learning" refers to machine learning algorithms and databases that analyze users' preferences and past selection history to understand their eating trends.

[1192] "Means for making meal suggestions" refers to algorithms or systems that suggest appropriate meals to users based on the results of their learning.

[1193] "Means of recognizing emotions" refers to emotion analysis engines or software used to analyze a user's facial expressions and voice to identify the user's emotional state.

[1194] "Means enabling customization of meal suggestions" means an interface or system that allows a user to modify the suggested meals to suit their personal preferences.

[1195] "Feedback" refers to ratings and comments that users make on suggested meals.

[1196] This invention is an artificial intelligence system for assisting a user in selecting meals, learning the user's preferences and past selection history, and recognizing the user's emotions to suggest meals. A specific embodiment of this system will be described below.

[1197] System Configuration

[1198] Hardware and Software

[1199] The system consists of a server, a terminal, and a user. The server is equipped with a database and a machine learning algorithm, and the terminal is equipped with an emotion analysis engine. Specific software used is Microsoft Azure's Emotion API and Google Cloud's Natural Language API.

[1200] Data collection and learning

[1201] User data collection

[1202] A user inputs their daily meal selections into the terminal. For example, if the user chooses pasta for lunch, the user inputs that information into the terminal. The input data includes the type of meal, date and time, and rating (e.g., "delicious").

[1203] Accumulating and learning data

[1204] The server receives the user's meal data sent from the device and stores it in a database. The server uses the stored data to train a machine learning model. This model learns the user's eating habits and uses them to make future suggestions. For example, if the user tends to frequently choose pasta, the server learns that information.

[1205] Emotion recognition and meal suggestions

[1206] Emotion recognition

[1207] The device is equipped with an emotion engine to recognize the user's emotions. When the user speaks to the device, the device analyzes the user's facial expressions and voice to identify the user's emotional state. For example, if the user is smiling while speaking, the device recognizes that the user is expressing the emotion of happiness.

[1208] Meal suggestions

[1209] The server generates appropriate meal suggestions based on the emotion data sent from the device and the user's past selection history. For example, if the user expresses joy, the server will suggest sweets and celebratory meals that the user likes based on past data. Specifically, chocolate cakes and fruit tarts that the user has given high ratings in the past are suggested.

[1210] Examples

[1211] Consider a case where a user has been successful at work and is expressing emotions of joy. The user speaks to the device, saying, "I'm so happy today!" The device analyzes the user's facial expressions and voice to recognize the emotion of joy. The recognition results are sent to the server. The server then suggests meals that the user might like based on the received emotion data and past selection history. For example, it might suggest chocolate cake or fruit tart that the user has given a high rating in the past. The suggestions are sent to the device and displayed to the user.

[1212] Examples of prompt statements

[1213] Examples of prompts to be input to a generative AI model include:

[1214] What are some examples of meals you would suggest if the user is expressing the emotion of pleasure? The user's past selection history includes chocolate cake and fruit tart.

[1215] Using this prompt, the generative AI model can generate appropriate meal suggestions based on the user's emotions and past selection history.

[1216] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1217] Step 1: Collect user data

[1218] A user inputs their daily meal selections into the terminal. For example, if the user chooses pasta for lunch, the user inputs that information into the terminal. The input data includes the type of meal, date and time, and rating (e.g., "delicious").

[1219] Input: Meal type, date, time, rating

[1220] Output: Meal data stored on the device

[1221] Step 2: Sending and storing data

[1222] The terminal transmits the collected meal data of the user to the server.

[1223] The server stores the received data in a database.

[1224] Input: Meal data sent from the device

[1225] Output: Meal data stored in a database

[1226] Step 3: Train the data

[1227] The server uses the user's food data stored in the database to train a machine learning model, which learns the user's eating habits and uses them to make future suggestions.

[1228] Input: Meal data stored in a database

[1229] Output: A trained machine learning model

[1230] Step 4: Recognize emotions

[1231] When a user speaks to the device, the device analyzes the user's facial expressions and voice to identify the user's emotional state. For example, if the user is smiling while speaking, the device recognizes that the user is expressing the emotion of joy. The emotion engine uses, for example, the Emotion API and Natural Language API.

[1232] Input: User's facial expression, voice

[1233] Output: Recognized emotion data

[1234] Step 5: Sending emotion data

[1235] The terminal transmits the recognized emotion data to the server.

[1236] Input: Recognized emotion data

[1237] Output: Emotion data sent to the server

[1238] Step 6: Meal Suggestions

[1239] The server generates appropriate meal suggestions based on the emotion data sent from the device and the user's past selection history. For example, if the user expresses joy, the server will suggest sweets and celebratory meals that the user likes based on past data.

[1240] Input: Emotion data, past selection history

[1241] Output: Suggested meals

[1242] Step 7: Viewing Proposals

[1243] The terminal displays the suggestions sent by the server to the user, who can then review the suggested meals and customize them if necessary.

[1244] Input: Suggested meals

[1245] Output: The suggestions displayed to the user

[1246] Examples

[1247] Consider a case where a user has been successful at work and is expressing emotions of joy. The user speaks to the device, saying, "I'm so happy today!" The device analyzes the user's facial expressions and voice to recognize the emotion of joy. The recognition results are sent to the server. The server then suggests meals that the user might like based on the received emotion data and past selection history. For example, it might suggest chocolate cake or fruit tart that the user has given a high rating in the past. The suggestions are sent to the device and displayed to the user.

[1248] (Application example 1)

[1249] Next, a description will be given of Application Example 1 of Form 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".

[1250] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they have the problem that it is difficult to provide meals that match the user's current mood because they cannot make suggestions that take into account the user's emotional state. In addition, there is a problem that it is not possible to improve user satisfaction because there is a lack of means to recognize the user's emotions in real time and suggest appropriate meals based on them.

[1251] 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.

[1252] In this invention, the server includes means for learning the user's preferences and past selection history, means for suggesting meals based on the results of the learning, means for enabling customization of the suggested meals, means for recognizing the user's emotions and suggesting meals based on the emotions, and means for recognizing the user's emotions in real time using the smartphone camera, thereby enabling meal suggestions according to the user's current emotional state and improving user satisfaction.

[1253] "User preferences" refers to the meals selected by the user in the past and the user's particular preferences for meals.

[1254] "Past selection history" refers to a record of meals previously selected by a user.

[1255] "Means of learning" refers to the function of accumulating data on user preferences and past selection history, and analyzing this to find patterns.

[1256] "Means for making meal suggestions" refers to a function that suggests suitable meals to the user based on the results of learning.

[1257] "Means enabling customization of meal suggestions" refers to functionality that allows a user to modify the suggested meals to suit their personal preferences.

[1258] "Means of recognizing the user's emotions" refers to the function of analyzing the user's emotions from their facial expressions, voice, etc., and determining their state.

[1259] "Means for making meal suggestions based on emotions" refers to a function that suggests appropriate meals based on the recognized emotional state of the user.

[1260] "Means of recognizing a user's emotions in real time using a smartphone camera" refers to a function that uses a smartphone camera to analyze a user's facial expressions and determine their emotions in real time.

[1261] As an embodiment of the present invention, we will specifically explain an AI system for assisting users in choosing meals. This system learns the user's preferences and past selection history, and incorporates an emotion engine to make meal suggestions based on the user's emotions.

[1262] System Configuration

[1263] The system consists of the following main components:

[1264] 1. User device: A mobile information terminal such as a smartphone equipped with a camera and microphone.

[1265] 2. Server: Stores user data and provides computing resources to run AI models.

[1266] 3. AI model: A generative AI model that learns user preferences and past selection history and recognizes emotions.

[1267] Hardware and Software Used

[1268] Hardware: Smartphone camera, microphone, server

[1269] Software: OpenCV (camera capture and face detection), Keras (loading and predicting emotion recognition models), JSON (saving and loading the user's past selection history)

[1270] Data processing and calculation

[1271] 1. User device:

[1272] It uses a smartphone camera to capture the user's face in real time.

[1273] Faces are detected using OpenCV and pre-processed for input into the emotion recognition model.

[1274] We use Keras to run an emotion recognition model and predict user emotions in real time.

[1275] 2. Server:

[1276] User preferences and past selection history are stored in JSON format and loaded as needed.

[1277] It runs an algorithm to suggest suitable meals based on the user's emotional state.

[1278] Examples

[1279] For example, when a user is using the app through the smartphone camera, the app detects the user's face and recognizes emotions. If the user is recognized as "Happy," the app will suggest "favorite_sweets" from the user's past selection history.

[1280] Examples of prompt statements

[1281] Develop an AI system that recognizes the user's emotions and suggests meals based on those emotions. Create a smartphone app that learns the user's past selection history, incorporates an emotion engine, and has the ability to recognize emotions in real time and suggest appropriate meals.

[1282] In this way, meal suggestions can be made according to the user's current emotional state, improving user satisfaction.

[1283] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1284] Step 1:

[1285] The user device captures the user's face in real time using the smartphone camera.

[1286] Input: Smartphone camera image

[1287] Output: A captured image of the user's face.

[1288] Specific operation: The smartphone camera is activated and captures an image of the user's face.

[1289] Step 2:

[1290] The user device uses OpenCV to detect faces from captured facial images and performs preprocessing for input into the emotion recognition model.

[1291] Input: A captured image of the user's face

[1292] Output: Preprocessed face image

[1293] Specific operation: Face detection is performed using OpenCV, the detected face part is converted to grayscale, and resized to a size suitable for the emotion recognition model.

[1294] Step 3:

[1295] The user device runs an emotion recognition model using Keras to predict the user's emotions in real time.

[1296] Input: Preprocessed face image

[1297] Output: Predicted user emotion label

[1298] Specific operation: The preprocessed facial image is input into the emotion recognition model, and the model predicts the emotion and outputs the emotion label.

[1299] Step 4:

[1300] The server stores user preferences and past selection history in JSON format and loads them as needed.

[1301] Input: User's past selection history data

[1302] Output: Saved JSON data

[1303] Specific behavior: Save the user's past selection history in a JSON format file and load the file when needed.

[1304] Step 5:

[1305] The server runs an algorithm to suggest appropriate meals based on the user's emotional state.

[1306] Input: Predicted user emotion label, user's past choice history data

[1307] Output: Suggested meals

[1308] How it works: Based on emotion labels and past selection history, the algorithm selects and suggests meals suitable for the user.

[1309] Step 6:

[1310] The user device displays the suggested meals to the user.

[1311] Input: Suggested meals

[1312] Output: Meal suggestions displayed to the user

[1313] Specific behavior: Suggested meals are displayed on the smartphone screen for the user to review.

[1314] Example 2

[1315] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[1316] Conventional meal recommendation systems can suggest meals based on the user's past selection history, but they have a problem in that they cannot customize the suggestions based on the user's emotional state. This makes it difficult to improve user satisfaction because the system cannot suggest optimal meals according to the user's current emotions.

[1317] 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.

[1318] In this invention, the server includes means for learning the user's preferences and past selection history, means for making meal suggestions based on the learning results, and means for analyzing the user's emotional state and adjusting the customization of the suggested meals, thereby enabling automatic meal customization based on the user's past selection history and emotional data.

[1319] An "artificial intelligence system" is a computer system that learns from user data and makes predictions and suggestions.

[1320] "User preferences" refers to information about the meals and tastes selected by the user in the past.

[1321] "Past selection history" is a record of meals the user has previously chosen.

[1322] "Learning tools" refers to the algorithms and models that analyze user data and find patterns and trends.

[1323] "Means for making meal suggestions" is a function that presents suitable meals to the user based on the results of learning.

[1324] "Means for analyzing emotional state" refers to algorithms or models used to analyze a user's emotional data and determine their state.

[1325] "Means for adjusting customization" refers to the ability to change the content and presentation of a meal depending on the user's emotional state.

[1326] "Emotional data" is information about emotions expressed by a user, such as happiness, excitement, depression, stress, etc.

[1327] "Feedback" means any opinion or evaluation information provided by a User.

[1328] This invention is an artificial intelligence system for assisting a user in selecting meals by learning the user's preferences and past selection history, making meal suggestions based on the results of that learning, and further adjusting the customization of the suggested meals by analyzing the user's emotional state.

[1329] Hardware and Software Used

[1330] Hardware: Servers, terminals (smartphones, tablets, PCs)

[1331] Software: Generative AI models, emotion engine, database management system (DBMS)

[1332] Specific operation of the system

[1333] Data collection

[1334] The user uses the terminal to select a meal. The terminal sends the user's selected meal data to the server. The server stores the received data in a database. Specifically, it stores information such as the user ID, the type of meal selected, and the date and time of selection. If the user inputs emotion data when selecting a meal, the terminal also sends that emotion data to the server. The emotion data is expressed in categories such as "happiness," "excitement," "depression," and "stress."

[1335] Data Learning

[1336] The server periodically retrieves the user's selection history and emotion data from the database. The server uses the retrieved data to train a generative AI model. Specifically, the server uses the user's past selection history and emotion data as input to build a model that predicts the meal that is likely to be selected next. The server saves the trained generative AI model and uses it for the next suggestion.

[1337] Meal suggestions

[1338] When the user uses the device again to select a meal, the device sends a request to the server. The server receives the request and uses the generative AI model to predict which meal would suit the user's preferences. The server sends the prediction result to the device. Specifically, if the user has often selected Japanese food in the past, the server will suggest options such as "sushi" and "tempura" from among the Japanese dishes. The device will display the suggested meals to the user.

[1339] Emotion-Based Customization

[1340] When a user inputs emotion data when selecting a meal, the terminal transmits the emotion data to the server. The server uses an emotion engine to analyze the user's emotion state. The server generates customization suggestions based on the analysis result. For example, if the user indicates the emotion of "joy", the server suggests adding chocolate sauce or upgrading the dessert. The server transmits the customization suggestions to the terminal. The terminal displays the suggested customization to the user.

[1341] Examples

[1342] User's selection history: The user has previously selected Japanese food 10 times and Western food 5 times.

[1343] Emotional data: The user indicates the emotion of “happiness” when choosing a meal.

[1344] Examples of prompt statements

[1345] The user's past selection history is Japanese food 10 times and Western food 5 times. Currently, the user is showing the emotion of "delight". What meal and customization would you suggest for this user next?

[1346] In this way, the system suggests optimal meals and customizations based on the user's selection history and emotional data.

[1347] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1348] Step 1: Data collection

[1349] The user uses the terminal to select a meal. The terminal sends the user's selected meal data (input) to the server. The server stores the received data in a database (output). Specifically, it stores information such as the user ID, the type of meal selected, and the date and time of selection. If the user inputs emotion data when selecting a meal, the terminal also sends that emotion data to the server. The emotion data is expressed in categories such as "happiness," "excitement," "depression," and "stress."

[1350] Step 2: Data training

[1351] The server periodically retrieves the user's selection history and emotion data from the database (input). The server uses the retrieved data to train a generative AI model (data processing). Specifically, the server uses the user's past selection history and emotion data as input to build a model that predicts the meal that will likely be selected next (output). The server saves the trained generative AI model and uses it for the next suggestion.

[1352] Step 3: Meal Suggestions

[1353] When the user uses the device again to select a meal, the device sends a request to the server (input). The server receives the request and uses the generative AI model to predict which meal would suit the user's preferences (data calculation). The server sends the prediction result to the device (output). Specifically, if the user has often selected Japanese food in the past, the server will suggest options such as "sushi" and "tempura" from among the Japanese dishes. The device will display the suggested meals to the user.

[1354] Step 4: Emotional customization

[1355] When a user inputs emotion data when selecting a meal, the device sends the emotion data to the server (input). The server uses an emotion engine to analyze the user's emotional state (data calculation). The server generates customization suggestions based on the analysis results (output). For example, if the user indicates the emotion of "joy", it suggests adding chocolate sauce or upgrading the dessert. The server sends the customization suggestions to the device. The device displays the suggested customizations to the user.

[1356] In this way, the system suggests optimal meals and customizations based on the user's selection history and emotional data.

[1357] (Application example 2)

[1358] Next, a description will be given of Application Example 2 of Form 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".

[1359] Conventional food delivery services can suggest meals based on a user's past selection history and preferences, but they have the problem of not being able to provide customized suggestions that take into account the user's emotional state. This has left them lacking a means to further improve user satisfaction.

[1360] 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.

[1361] In this invention, the server includes means for learning the user's preferences and past selection history, means for suggesting meals based on the learning results, means for enabling customization of the suggested meals, means for detecting the user's emotional state, and means for suggesting meal customization based on the detected emotional state, thereby enabling meal suggestions and customization according to the user's preferences and emotional state.

[1362] "User Preferences" refers to a User's preferences for certain meals and ingredients based on the User's past meal selections and preferences.

[1363] "Past selection history" refers to a record of a user's previous meal selections and orders.

[1364] "Means of learning" refers to algorithms and systems that analyze user preferences and past choices to extract patterns and trends.

[1365] "Means for providing meal suggestions" refers to an algorithm or system that uses the results of learning to recommend suitable meals to users.

[1366] "Means enabling customization of suggested meals" refers to functionality that allows a user to add to or change suggested meals.

[1367] "Means for detecting a user's emotional state" refers to technologies or systems that analyze a user's facial expressions and behavior to identify their current emotional state (happiness, excitement, depression, stress, etc.).

[1368] "Means for suggesting meal customization based on detected emotional state" refers to an algorithm or system for suggesting meal content and customization options based on the user's emotional state.

[1369] The system for implementing the present invention learns the user's preferences and past selection history to provide meal suggestions, and also detects the user's emotional state to provide customized suggestions. A specific embodiment of this system will be described below.

[1370] Hardware and Software Used

[1371] Hardware: Smartphone, camera (for emotion detection)

[1372] Software: TensorFlow (AI model), OpenCV (emotion detection), Flask (backend), React Native (frontend)

[1373] Data processing and calculation

[1374] Learn user preferences and history

[1375] The server uses TensorFlow to learn from the user's past order history data, which allows it to analyze the user's preferences and selection trends and predict the meal they are likely to choose next.

[1376] Emotion Detection

[1377] The device (smartphone) uses a camera to analyze the user's face in real time, and uses OpenCV to detect the user's emotional state (happiness, excitement, depression, stress, etc.) from their facial expressions.

[1378] Customization suggestions

[1379] The server then uses the results of the emotion engine to make customized meal suggestions, such as upgrading the dessert if the user is happy, or adjusting the soup strength if the user is depressed.

[1380] Examples

[1381] User A: Has often ordered Japanese food in the past. Currently, the emotion detected by the camera is "Happiness."

[1382] Recommendation: Suggest sushi as part of the Japanese menu and also suggest matcha ice cream as an additional dessert.

[1383] User B: Often ordered Western food in the past. Currently, the emotion detected by the camera is "stress."

[1384] Suggestion: Suggest pasta from the Western menu, with an additional suggestion to adjust the soup thickness.

[1385] Examples of prompt statements

[1386] Suggest the next meal that the user is likely to choose based on their past order history. Detect the user's emotional state with a camera and provide customized suggestions based on their emotions.

[1387] In this way, a system can be realized that allows for meal suggestions and customization based on the user's preferences and emotional state.

[1388] The flow of the identification process in the application example 2 will be described with reference to FIG.

[1389] Step 1:

[1390] The server receives the user's past order history data. The input includes the user ID and the past order history data. The server uses TensorFlow to learn from this data and analyze the user's preferences and selection trends. As an output, a list of meal suggestions based on the user's preferences is generated.

[1391] Step 2:

[1392] The device (smartphone) uses a camera to capture a picture of the user's face in real time. The input includes the user's facial image data. The device uses OpenCV to analyze this image data and detect the user's emotional state. As an output, the user's emotional state (e.g., happy, excited, depressed, stressed, etc.) is identified.

[1393] Step 3:

[1394] The server receives the meal suggestion list generated in step 1 and the user's emotional state detected in step 2. As input, it includes the meal suggestion list and the emotional state data. The server uses an emotion engine to provide customized meal suggestions based on the emotional state. As output, a customized meal suggestion list is generated.

[1395] Step 4:

[1396] The terminal displays the customized meal suggestion list received from the server to the user. As input, it contains the customized meal suggestion list. The terminal displays the meal suggestions on a user interface using React Native. As output, the user can see the suggested meals.

[1397] Step 5:

[1398] The user selects a desired meal from a customized meal suggestion list displayed on the terminal. As input, the customized meal suggestion list is included. The user confirms the selected meal and places the order. As output, order data for the selected meal is generated.

[1399] Step 6:

[1400] The server receives order data from the user and sends the order to the food delivery service. As input, the server includes the user's order data. The server sends the order data to the food delivery service's system and confirms the order. As output, order confirmation data is generated.

[1401] In this way, meal suggestions and customizations are made based on the user's preferences and emotional state, allowing the user to select and order meals that satisfy them.

[1402] Example 3

[1403] Next, a third embodiment of the third embodiment will be described. 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".

[1404] Conventional meal recommendation systems make suggestions based on the user's preferences and past selection history, but they have the problem of not being able to customize or make suggestions that take into account the user's emotional state. This makes it difficult to make meal suggestions that are appropriate for the user's current emotional state, making it difficult to increase user satisfaction.

[1405] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1406] In this invention, the server includes means for learning the user's preferences and past selection history, means for making meal suggestions based on the learning results, means for enabling customization of the suggested meals, means for analyzing the user's emotional state, and means for adjusting the meal suggestions and customization based on the analyzed emotional state, thereby enabling meal suggestions and customization according to the user's emotional state.

[1407] "Means of learning about the user's preferences and past selection history" refers to a function for understanding the user's tastes by collecting and analyzing data on the meals and preferences selected by the user in the past.

[1408] "Means of suggesting meals based on learned results" is a function for suggesting appropriate meals based on collected user preferences and past selection history.

[1409] "Means for customizing suggested meals" refers to functionality that allows a user to add or remove ingredients from suggested meals, change cooking methods, etc.

[1410] "Means for analyzing a user's emotional state" refers to a function for analyzing and understanding a user's current emotional state from their input and actions.

[1411] "Means for tailoring meal suggestions and customization based on analyzed emotional state" refers to functionality for tailoring meal suggestions and customization options based on the analyzed emotional state of the user.

[1412] The present invention relates to an artificial intelligence system for assisting a user in selecting meals. A specific embodiment of the system will be described below.

[1413] The server uses a database to learn the user's preferences and past selection history. The database stores the user's past meal selections and preferences. The server analyzes this data and runs algorithms to understand the user's tastes. The algorithms use machine learning techniques.

[1414] Next, the server uses the generative AI model to make meal suggestions based on the learned results. For example, a model using natural language processing technology (e.g., GPT-4) is used as the generative AI model. The server receives the user's input data and generates a prompt sentence for the generative AI model. An example of a prompt sentence is "The user says he wants to eat pizza. Please recommend a pizza." The generative AI model generates appropriate meal suggestions based on this prompt sentence.

[1415] In addition, the server provides a user interface to enable customization of the suggested meals. The user can use the terminal to add or remove ingredients from the suggested meals, change the cooking method, etc. For example, if the user inputs "I want to add toppings to my pizza," the server receives this customization information and again uses the generative AI model to generate a customized meal suggestion.

[1416] The server also uses an emotion engine to analyze the user's emotional state. The emotion engine executes algorithms to analyze and understand the user's current emotional state from the user's inputs and actions. For example, if the user inputs "I'm tired today, please suggest a meal that will help me relax," the emotion engine analyzes the emotional state as "tired."

[1417] The server adjusts meal suggestions and customization based on the analyzed emotional state. For example, if the user is tired, the server will suggest "hot soup" and simultaneously suggest adding ingredients (e.g., chicken and vegetables). This allows for meal suggestions and customization according to the user's emotional state.

[1418] As a specific example, if a user inputs, "I'm tired today, please suggest a meal that will help me relax," the following actions will be taken:

[1419] 1. The user types into the device, "I'm tired today, so please suggest some meals that will help me relax."

[1420] 2. The terminal sends this input data to the server.

[1421] 3. The server analyzes the input data and generates a prompt saying, "The user is tired. Please suggest a meal that will help him relax."

[1422] 4. The server uses the generative AI model to suggest "hot soup."

[1423] 5. The server uses the emotion engine to analyze the user’s emotional state as “tired.”

[1424] 6. The server offers to add extra ingredients (e.g. chicken or vegetables) in addition to the "hot soup."

[1425] 7. The server sends this proposal to the device.

[1426] 8. The device displays to the user, "Hot soup and add toppings (chicken and / or vegetables)."

[1427] 9. The user enters a customization: "I would like to add chicken to the soup."

[1428] 10. The terminal sends this customization information to the server.

[1429] 11. The server generates a customized "hot chicken soup" and sends it to the terminal.

[1430] 12. The device displays the final “Hot Chicken Soup” to the user.

[1431] In this way, a system is realized in which the server, the terminal, and the user work together to suggest and customize meals. The flow of the specification process in the third embodiment will be described with reference to FIG.

[1432] Step 1:

[1433] A user enters a meal request.

[1434] A user uses the terminal interface to input a meal request, for example, "I'd like pizza." The input data includes the user's request.

[1435] Step 2:

[1436] The terminal transmits the user's input data to the server.

[1437] The terminal sends the user's input data to the server. The input data includes the user's request. The output is the user's request data sent to the server.

[1438] Step 3:

[1439] The server analyzes the user's input data and generates a prompt for the generative AI model.

[1440] The server analyzes the user's input data and generates a prompt sentence suitable for the generative AI model. For example, it generates a prompt sentence such as "The user says he wants to eat pizza. Please recommend some pizzas." The input is the user's request data, and the output is the generated prompt sentence.

[1441] Step 4:

[1442] The server uses the generative AI model to generate meal suggestions.

[1443] The server uses a generative AI model (e.g., GPT-4) to generate meal suggestions based on the prompt sentence, e.g., suggest "Margherita Pizza." The input is the generated prompt sentence, and the output is the generated meal suggestions.

[1444] Step 5:

[1445] The server uses an emotion engine to analyze the user's emotional state.

[1446] The server uses the emotion engine to analyze the user's emotional state. For example, if the user inputs "I'm tired," the emotion engine analyzes the emotional state "tired." The input is data about the user's emotions, and the output is the analyzed emotional state.

[1447] Step 6:

[1448] Servers tailor meal suggestions and customizations based on emotional state.

[1449] The server adjusts the meal suggestions and customization based on the emotional state. For example, if the user is tired, the server may suggest "hot soup" and simultaneously suggest adding ingredients (e.g., chicken and vegetables). The inputs are the parsed emotional state and the generated meal suggestions, and the outputs are the adjusted meal suggestions and customization options.

[1450] Step 7:

[1451] The server sends the final proposal to the device.

[1452] The server sends the final meal suggestions to the terminal. The inputs are the adjusted meal suggestions and the customization options, and the output is the final suggestions sent to the terminal.

[1453] Step 8:

[1454] The terminal displays the suggestions to the user.

[1455] The terminal displays the meal suggestions received from the server to the user. The input is the final suggestions sent by the server and the output is the suggestions displayed to the user.

[1456] Step 9:

[1457] The user customizes the suggestions.

[1458] The user customizes the proposed meal by adding or deleting ingredients, changing the cooking method, etc. For example, the user inputs "I would like to add toppings to the pizza." The input is the user's customization request, and the output is the customization information.

[1459] Step 10:

[1460] The terminal transmits the customization information to the server.

[1461] The terminal transmits the user's customization information to the server, where the input is the user's customization information and the output is the customization information transmitted to the server.

[1462] Step 11:

[1463] The server generates customized meal suggestions and transmits them to the terminal.

[1464] The server generates customized meal suggestions based on the user's customization information and sends them to the terminal, where the input is the user's customization information and the output is the customized meal suggestions.

[1465] Step 12:

[1466] The terminal displays the final customization proposal to the user.

[1467] The terminal displays the final customized proposal received from the server to the user. The input is the customized meal proposal sent from the server, and the output is the final customized proposal displayed to the user.

[1468] (Application example 3)

[1469] Next, a description will be given of Application Example 3 of Form Example 3. 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".

[1470] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they have the problem of being unable to customize or suggest meals that take into account the user's emotional state. This means that meal suggestions that are appropriate for the user's current emotional state cannot be made, and user satisfaction cannot be sufficiently increased.

[1471] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1472] In this invention, the server includes means for learning the user's preferences and past selection history, means for making meal suggestions based on the learning results, means for enabling customization of the suggested meals, means for analyzing the user's emotions, and means for adjusting the meal suggestions and customization based on the analyzed emotions, thereby enabling personalized meal suggestions and customization according to the user's emotional state.

[1473] "User preferences" are dietary trends identified by a user based on past dietary choices and preferences.

[1474] "Past selection history" is a record of meals previously selected by the user.

[1475] A "means of learning" is an algorithm or system that analyzes user preferences and past choices to extract patterns.

[1476] The "means for making meal suggestions" is a function for presenting suitable meals to the user based on the results of learning.

[1477] "Means to customise" refers to functionality that allows users to add or remove ingredients from suggested meals, change cooking methods, etc.

[1478] "Means of analyzing emotions" refers to technologies or systems for reading emotions from a user's facial expressions, voice, etc.

[1479] "Means for tailoring meal suggestions and customization based on emotions" refers to a function that changes meal suggestions and customization options in response to the analyzed emotions.

[1480] The system for implementing the present invention learns the user's preferences and past selection history, analyzes the user's emotions, and proposes and customizes meals. A specific embodiment of this system will be described below.

[1481] System Configuration

[1482] The system consists of a user terminal, a server, and an emotion analysis engine. The user terminal is a device such as a smartphone or tablet, and the server is located on the cloud. The emotion analysis engine is software for analyzing the user's facial expressions and voice.

[1483] Program Processing

[1484] The server uses a database to learn the user's preferences and past selection history. Every time the user selects a meal, the information is stored in the database and analyzed using machine learning algorithms (e.g., Keras and TensorFlow), which allows the server to learn the user's preferences and selection patterns.

[1485] The user device uses a camera and microphone to capture the user's facial expressions and voice, and then sends the data to the sentiment analysis engine, which uses OpenCV and deep learning models to analyze the user's emotions. The analysis results are sent to the server, which then makes meal suggestions and customizations based on the results.

[1486] Examples

[1487] For example, consider a scenario where a user opens a smartphone app, takes a picture of their face with their camera, and uploads it. The app analyzes the user's facial expressions and detects, for example, an emotion of "happiness." As a result, the server suggests "chocolate cake" and "add chocolate sauce."

[1488] Example of a prompt to be input to a generative AI model:

[1489] "A user opens the smartphone app, takes a picture of their face with their camera and uploads it. The app analyzes the user's facial expressions and detects the emotion of joy. As a result, the app suggests adding chocolate cake and chocolate sauce."

[1490] In this way, personalized meal suggestions and customization based on the user's emotional state is possible.

[1491] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1492] Step 1:

[1493] Users open the smartphone app, take a photo of their face with their camera, and upload it.

[1494] Input: User's face image

[1495] Output: Face image data

[1496] What it does: A user takes a picture of their face using their smartphone camera and uploads the image to the app.

[1497] Step 2:

[1498] The device sends the uploaded facial image to an emotion analysis engine.

[1499] Input: Face image data

[1500] Output: Emotion analysis results

[1501] Specific operation: The device sends facial image data to the emotion analysis engine, which then uses OpenCV and deep learning models to analyze the user's emotions.

[1502] Step 3:

[1503] The emotion analysis engine analyzes the user's emotions and sends the results to the server.

[1504] Input: Face image data

[1505] Output: Emotion label (e.g. happy, excited, depressed, stressed)

[1506] How it works: The emotion analysis engine analyzes facial image data to identify the user's emotions. The analysis results are sent to the server as emotion labels.

[1507] Step 4:

[1508] The server receives the user's emotion labels and makes meal suggestions based on them.

[1509] Input: emotion label

[1510] Output: Meal suggestions (e.g. chocolate cake, pizza, hot soup, herbal tea)

[1511] Specific behavior: The server receives the emotion labels and suggests appropriate meals based on pre-defined rules and databases.

[1512] Step 5:

[1513] The server references the user's past selection history to customize the suggested meals.

[1514] Input: Meal suggestions, past selection history

[1515] Output: Customized meal suggestions (e.g. chocolate cake + add chocolate sauce)

[1516] Specific behavior: The server retrieves the user's past selection history from the database and makes appropriate customizations to the suggested meals.

[1517] Step 6:

[1518] The server transmits the customized meal suggestions to the user terminal.

[1519] Input: Customized Meal Suggestions

[1520] Output: Notification to user device

[1521] Specific operation: The server sends the customized meal suggestions to the user terminal and notifies the user.

[1522] Step 7:

[1523] The user reviews the suggested meals and further customizes them if necessary.

[1524] Input: Customized Meal Suggestions

[1525] Output: Final meal selection

[1526] Specific behavior: The user can review the suggested meals on the device, add or remove ingredients, change the cooking method, etc. The final meal selection is confirmed.

[1527] 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 a voice indicating a user input on the result of the specific processing.

[1528] The unit 46A transmits voice data representing the user input captured by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 captures the voice data.

[1529] 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> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to 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.

[1530] Another example of generative AI is Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1531] 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 smart glasses 214.

[1532] [Third embodiment]

[1533] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1534] 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.

[1535] 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 wide area network (WAN) and / or a local area network (LAN).

[1536] 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.

[1537] 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 the voice according to instructions from the processor 46.

[1538] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[1539] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[1540] Fig. 6 shows an example of 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.

[1541] 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.

[1542] 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.

[1543] In the headset type terminal 314, 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.

[1544] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1545] "Example 1"

[1546] As one embodiment of the present invention, an AI system that supports a user in selecting meals is provided. This system has a function to learn the user's preferences and past selection history. Specifically, the system accumulates data on the meals the user has selected in the past and the user's tendency toward preferred meals, and the AI ​​learns from this data.

[1547] "Example 2"

[1548] In addition, the AI ​​also has the ability to suggest meals based on the results of its learning. Specifically, it suggests meals that the user is likely to choose next based on the user's preferences and past selection history that it has learned. For example, if the user has often chosen Japanese food in the past, the AI ​​will suggest Japanese food that the user is likely to like.

[1549] "Example 3"

[1550] It also has a function that allows users to customize the suggested meals. Specifically, users can customize the suggested meals by adding or removing ingredients, changing the cooking method, etc. For example, users can add toppings or increase the amount of cheese to the pizza suggested by the AI.

[1551] The process flow of each embodiment will be described below.

[1552] "Example 1"

[1553] Step 1: When a user begins using the system, the system begins recording the user's food choices, including the types of food the user chooses, the time of meal, and frequency of meals.

[1554] Step 2: The system accumulates this data and the AI ​​learns about the user's preferences and choices. This learning process evolves every time the user uses the system, providing more accurate results.

[1555] "Example 2"

[1556] Step 1: After the AI ​​learns the user's preferences and selection trends, it will suggest the next meal the user is likely to choose. These suggestions will be displayed when the user logs into the system, when it is time to eat, etc.

[1557] Step 2: The user can choose to either select the suggested meal or choose a different meal. If the user selects the suggested meal, that information is again fed back into the system to help train the AI.

[1558] "Example 3"

[1559] Step 1: After the user selects a suggested meal, they can then customize it by adding or removing ingredients, changing the cooking method, etc.

[1560] Step 2: The user's customization information is also fed back to the system and used to train the AI, allowing the system to better understand the user's preferences and provide more personalized meal suggestions.

[1561] Example 1

[1562] Next, a description will be given of Example 1 of the embodiment 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".

[1563] Conventional meal selection support systems had the problem that they did not make suggestions that fully considered the user's preferences or past selection history, resulting in low user satisfaction. In addition, the suggested meals were customized manually, which was a hassle for the user. Furthermore, the system lacked the functionality to learn the user's selection trends and recommend appropriate meals, making it difficult to suggest meals that matched the user's preferences.

[1564] 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.

[1565] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for proposing meals based on the results of the learning, a means for enabling customization of the proposed meals, a means for collecting the user's meal data, a means for accumulating the collected data, a means for learning the user's selection tendency based on the accumulated data, a means for generating recommended meals based on the learned model, and a means for notifying the user of the generated recommended meals. This enables meal suggestions that take into account the user's preferences and past selection history, thereby improving user satisfaction. In addition, the proposed meals are automatically customized, reducing the user's effort. Furthermore, by learning the user's selection tendency and recommending appropriate meals, meal suggestions that match the user's preferences can be realized.

[1566] The "means for learning the user's preferences and past selection history" refers to a device or program that has the function of collecting data on the meals selected by the user in the past and analyzing the user's preferences and selection trends based on that data.

[1567] A "means for making meal suggestions based on the results of learning" is a device or program that has the function of recommending appropriate meals based on the user's preferences and past selection history.

[1568] A "means for enabling customization of meal suggestions" is a device or program that has the functionality to allow a user to modify the suggested meals to suit their personal preferences.

[1569] A "means for collecting user meal data" is a device or program that has the function of inputting information about meals selected by the user and collecting it as data.

[1570] The "means for storing collected data" refers to a device or program that has the function of storing collected user meal data in a storage device such as a database.

[1571] The "means for learning the user's selection tendencies based on accumulated data" refers to a device or program that has the function of analyzing accumulated meal data and modeling the user's selection tendencies.

[1572] A "means for generating recommended meals based on a learned model" is a device or program that has the function of generating appropriate meals based on the learned selection tendencies of a user.

[1573] The "means for notifying the user of the generated recommended meals" refers to a device or program having a function of transmitting the generated recommended meals to the user's terminal and notifying the user of the recommended meals.

[1574] This invention is an AI system for assisting users in choosing meals, learning from the user's preferences and past selection history and proposing appropriate meals. A specific embodiment of this system will be described below.

[1575] First, the user inputs meal information using the terminal. The terminal then sends the input meal information (e.g., meal name, date and time, rating) to the server. The server receives the meal information sent from the terminal and stores it in a database. General database software such as MySQL is used as the database management system.

[1576] Next, the server learns the user's selection trends based on the accumulated data. TensorFlow is used as the machine learning library. The server uses TensorFlow to model the user's food selection patterns and analyze the user's preferences and selection trends.

[1577] Once the learning is complete, the server generates meal recommendations based on the learned model. Using the generative AI model, it suggests three meals that match the user's preferences. For example, if the user previously selected "curry rice," "ramen," and "sushi," the server will recommend "tempura," "udon," and "sushi."

[1578] Finally, the server sends the generated recommended meals to the user's device. The notification service used is Firebase Cloud Messaging. The device notifies the user of the received recommended meals and displays them on the screen.

[1579] As a specific example, a user selects "curry rice" for lunch and enters "curry rice" into an app on the device. The device sends the entered meal information "curry rice" to the server. The server stores the received "curry rice" data in a MySQL database. Next, the server uses TensorFlow to learn that the user tends to prefer "Japanese food." Based on the learned model, the server recommends "tempura," "udon," and "sushi" and notifies the user's device of this recommendation.

[1580] An example of a prompt is, "Generate the next meal recommendation based on the user's past meal selection data. Suggest three specific meal names taking into account the user's preferences and selection trends."

[1581] This is how the AI ​​system that assists users in making meal choices works.

[1582] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1583] Step 1:

[1584] Collecting user dietary data

[1585] The user selects a meal and enters the information into the terminal.

[1586] Input: Meal information such as meal name, date and time, and rating.

[1587] The terminal transmits the input meal information to the server.

[1588] Output: Meal information sent to the server.

[1589] Specific actions: The user selects "curry rice" for lunch and enters "curry rice" into the device app.

[1590] Step 2:

[1591] Accumulation of data

[1592] The server receives the meal information transmitted from the terminal.

[1593] Input: Meal information sent from the device.

[1594] The server stores the received data in a database.

[1595] Output: Meal information stored in a database.

[1596] Specific operation: The server stores the received "curry rice" data in a MySQL database.

[1597] Step 3:

[1598] Training the Data

[1599] The server learns the user's preferences and selection trends based on the meal information accumulated in the database.

[1600] Input: Food information stored in a database.

[1601] The server uses machine learning libraries (e.g., TensorFlow) to model users' food choice patterns.

[1602] Output: The trained model.

[1603] Specific behavior: The server uses TensorFlow to learn that users tend to like "Japanese food."

[1604] Step 4:

[1605] Generation of recommended meals

[1606] The server generates meal recommendations for the user based on the learned model.

[1607] Input: The trained model.

[1608] The server uses a generative AI model to suggest three meals that match the user's preferences.

[1609] Output: A recommended diet list.

[1610] Specific operation: The server recommends "tempura," "udon," and "sushi" based on the learned model.

[1611] Step 5:

[1612] Dietary recommendations

[1613] The server transmits the generated recommended meals to the user's terminal.

[1614] Input: A recommended meal list.

[1615] The device will notify the user of the recommended meals it has received.

[1616] Output: The meal recommendation communicated to the user.

[1617] Specific operation: The server sends the generated recommended meals, "tempura," "udon," and "sushi," to the user's terminal, and the terminal notifies the user of them.

[1618] (Application example 1)

[1619] Next, a description will be given of Application Example 1 of Form 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".

[1620] Conventional food delivery systems do not adequately suggest meals based on the user's preferences and past selection history. In addition, they lack the ability to suggest meals according to the user's current mood and health status, and provide nutritional information and calorie calculations for meals, which results in low user satisfaction and poor health management. This makes it difficult for users to select appropriate meals.

[1621] 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.

[1622] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for customizing the meal suggestions based on the user's current mood and health condition, and a means for providing nutritional information and calorie calculations for the meals. This allows optimal meal suggestions based on the user's preferences and health condition, improving the user's satisfaction and health management.

[1623] "User preferences" refers to the meals selected by the user in the past and the user's particular preferences for meals.

[1624] "Past selection history" refers to the record of meals the user has selected in the past.

[1625] "Means of learning" refers to how the AI ​​analyzes data and identifies patterns and trends based on user preferences and past choices.

[1626] "Means for providing meal suggestions" refers to a method for recommending appropriate meals to a user based on the results of learning.

[1627] "Means enabling customization of meal suggestions" refers to a method that allows a user to modify the suggested meals to suit their personal preferences.

[1628] "Means for customizing meal suggestions based on current mood and health status" refers to a method for tailoring meal suggestions to a user's current mood and health status.

[1629] "Means for providing nutritional information and calorie calculations for a meal" means a method for calculating and providing to a user the nutrients and calories contained in a suggested meal.

[1630] As an embodiment of the present invention, an AI system that assists users in choosing meals will be specifically described. This system learns the user's preferences and past selection history, and suggests optimal meals. The configuration and operation of the system will be described in detail below.

[1631] System Configuration

[1632] The system consists of the following main components:

[1633] 1. User device: A device operated by a user, such as a smartphone or tablet.

[1634] 2. Server: The central system that processes and stores data.

[1635] 3. AI model: A machine learning model that learns from user preferences and past selection history to provide meal suggestions.

[1636] Hardware and Software Used

[1637] Hardware: Smartphones, tablets, servers

[1638] Software: Python, Pandas, Scikit-learn

[1639] Data processing and calculation

[1640] 1. Data collection: Collect data from the user's device about the user's past order history and preferences, including the user's past meal selections and their ratings.

[1641] 2. Data storage: Collected data is stored on the server. A database is used to efficiently manage data.

[1642] 3. Model learning: The AI ​​model learns from the collected data on the server. Specifically, the data is preprocessed using Pandas, and the NearestNeighbors model of Scikit-learn is used to learn user preferences.

[1643] 4. Meal Suggestion: Using the learned model, the system suggests optimal meals for the user. The suggestions are displayed on the user's device.

[1644] 5. Customization: Users can customize the suggested meals to suit their tastes. The results of the customization are also saved on the server and reflected in the next suggestions.

[1645] 6. Providing Nutritional Information: Providing nutritional information and calorie calculations for suggested meals to users.

[1646] Examples

[1647] For example, if User A has ordered "sushi" five times, "ramen" four times, and "pizza" three times in the past, the AI ​​will learn from this and suggest the meals User A is likely to order next. If the user inputs "I want to eat healthy today," the system will suggest low-calorie meals accordingly.

[1648] Examples of prompt statements

[1649] "Based on a user's past ordering history, suggest the next meal they're likely to order."

[1650] In this way, by applying an AI system that learns from users' preferences and past selection history to food delivery apps, more personalized meal suggestions can be made, improving user satisfaction and health management.

[1651] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1652] Step 1:

[1653] Data collection

[1654] The user terminal collects the user's past order history and preference data. Specifically, it receives the user's past meal selections and their ratings as input and sends this to the server. The input data includes the type of meal, the rating, the date and time of the order, etc. The server receives this and stores it in a database.

[1655] Step 2:

[1656] Data Storage

[1657] The server stores the received user order history data in a database. The stored data includes the user ID, meal type, rating, order date and time, etc. This allows the past selection history of each user to be accumulated.

[1658] Step 3:

[1659] Model Training

[1660] The server trains an AI model based on the stored data. Specifically, it preprocesses the data using Pandas and learns user preferences using Scikit-learn's NearestNeighbors model. The input data are user IDs and ratings, and the output is a model that shows each user's food preferences.

[1661] Step 4:

[1662] Meal suggestions

[1663] The server uses the trained model to suggest optimal meals to the user. Specifically, it receives a request from the user device and provides input data (such as the user ID and current mood) to the model. The model generates meal suggestions based on this and sends the suggestion results to the user device. The output data is a list of suggested meals.

[1664] Step 5:

[1665] Customization

[1666] The user terminal allows the user to customize the suggested meals. The user modifies the suggested meals to suit his / her preferences and inputs the results into the terminal. The input data is the customized meal content, which is then sent to the server. The server receives it and stores it in a database.

[1667] Step 6:

[1668] Nutritional information provided

[1669] The server calculates the nutritional information and calorie counts of the suggested meals and provides them to the user. Specifically, it receives a list of suggested meals as input, calculates the nutrients and calories of each meal, and sends the calculation results to the user's terminal and displays them to the user. The output data is the nutritional information and calorie details.

[1670] In this way, by applying an AI system that learns from users' preferences and past selection history to food delivery apps, more personalized meal suggestions can be made, improving user satisfaction and health management.

[1671] Example 2

[1672] Next, a second embodiment of the second form will be described. 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".

[1673] Conventional meal selection support systems do not provide suggestions that fully consider the user's preferences or past selection history, making it difficult to increase user satisfaction. In addition, the suggested meals are often customized manually, which is time-consuming for users. Furthermore, there is a lack of means to effectively utilize the user's selection data, making it difficult to accurately predict the meal that the user is likely to choose next.

[1674] 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.

[1675] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for collecting the user's selection data and storing it in a database, a means for analyzing the user's selection history using a generative AI model, a means for predicting the meal that is likely to be selected next based on the analysis results, and a means for presenting the prediction result to the user. This allows meals to be suggested taking into account the user's preferences and past selection history, thereby increasing user satisfaction. In addition, the suggested meals are automatically customized, reducing the user's effort. Furthermore, by effectively utilizing the user's selection data, it is possible to accurately predict the meal that is likely to be selected next.

[1676] "User" refers to an individual who uses the System to make meal selections.

[1677] "Artificial intelligence system" refers to a computer system that learns, analyzes user data, and makes appropriate suggestions.

[1678] "Preferences" refers to the eating patterns and tendencies that a user has chosen in the past.

[1679] "Choice History" refers to data regarding past food choices made by a User.

[1680] "Learning means" refers to the process of analyzing data and finding patterns based on user preferences and selection history.

[1681] "Means for making suggestions" refers to the process of suggesting appropriate meals to the user based on the results of learning.

[1682] "Customization features" refers to the ability to modify suggested meals to suit a user's preferences.

[1683] "Choice Data" refers to information about meals selected by a User through the System.

[1684] "Database" means a system for storing and managing User Selection Data.

[1685] "Generative AI model" refers to the artificial intelligence algorithms used to analyze user data and make predictions.

[1686] "Means of analyzing" refers to the process of using a generative AI model to analyze a user's selection history.

[1687] "Prediction means" refers to the process of estimating the next meal a user is likely to choose based on the analysis results.

[1688] "Presentation means" refers to the process of displaying the prediction results to the user.

[1689] This invention is an artificial intelligence system for assisting users in choosing meals, learning from the user's preferences and past selection history, and making meal suggestions based on the results. The system collects user selection data, stores it in a database, analyzes it using a generative AI model, predicts the next meal that the user is likely to choose, and presents it to the user.

[1690] Hardware and Software Used

[1691] Hardware: Servers, terminals (smartphones and PCs used by users)

[1692] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), web application frameworks (e.g. Django)

[1693] Details of data processing and calculation

[1694] 1. Data collection:

[1695] The user uses the device to make meal selections.

[1696] The terminal transmits the user's selection data to the server.

[1697] The server stores the received data in a database.

[1698] 2. Data Analysis:

[1699] The server retrieves the user's selection history from the database.

[1700] The server uses the generative AI model to analyze the user's selection history.

[1701] A generative AI model learns your preferences based on your selection history.

[1702] 3. Meal suggestions:

[1703] The server uses a generative AI model to predict what meal the user is likely to choose next.

[1704] The server transmits the prediction results to the terminal.

[1705] The device will show meal suggestions to the user.

[1706] Examples

[1707] User selection history: The user has frequently chosen Japanese food in the past.

[1708] Recommendation content: The AI ​​will suggest Japanese dishes that the user is likely to like, such as "sushi" or "tempura."

[1709] Examples of prompt statements

[1710] "Please suggest meals that the user is likely to choose next based on the user's past meal choices. The user has often chosen Japanese food in the past."

[1711] The system supports users in choosing meals by learning their preferences and suggesting the meal they are likely to choose next. Specifically, it uses a generative AI model to analyze user data and suggest appropriate meals, thereby increasing user satisfaction. In addition, the suggested meals are automatically customized, reducing the effort required by the user. Furthermore, by effectively utilizing the user's selection data, it becomes possible to accurately predict the meal they are likely to choose next.

[1712] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1713] Step 1:

[1714] The user selects a meal

[1715] The user selects a meal using a device (smartphone or PC). For example, the user opens the app and selects "Japanese food."

[1716] Input: User's meal selection (e.g. Japanese food)

[1717] Output: Selection data (e.g. Japanese food)

[1718] Step 2:

[1719] The terminal transmits the selection data to the server.

[1720] The terminal transmits the user's selection data to the server.

[1721] Input: User selection data (e.g. Japanese food)

[1722] Output: Selection data sent to the server

[1723] Step 3:

[1724] The server stores the selection in a database.

[1725] The server stores the received selection data in a database, including the user ID, the type of meal selected, and the date and time.

[1726] Input: Selection data sent to the server (e.g. User ID: 12345, Selection: Japanese food, Date and time: 2023-10-01 12:00)

[1727] Output: Selection data stored in a database

[1728] Step 4:

[1729] The server retrieves the selection history from the database.

[1730] The server periodically retrieves the user's selection history from the database.

[1731] Input: Selection data stored in the database

[1732] Output: User selection history (e.g. User ID: 12345's selection history)

[1733] Step 5:

[1734] The server analyzes the selection history using a generative AI model.

[1735] The server uses a generative AI model to analyze the user's selection history. The generative AI model learns the user's preferences based on the user's selection history.

[1736] Input: User selection history (e.g. User ID: 12345's past selection history)

[1737] Output: User preference pattern (e.g. User ID: 12345 likes Japanese food)

[1738] Step 6:

[1739] Predict the meal your server is likely to choose next.

[1740] The server uses the generative AI model to predict what meal the user is likely to choose next.

[1741] Input: User preference pattern (e.g. User ID: 12345 likes Japanese food)

[1742] Output: The next most likely meal (e.g. sushi)

[1743] Step 7:

[1744] The server transmits the prediction results to the terminal.

[1745] The server transmits the prediction results to the terminal.

[1746] Input: The next most likely meal (e.g. sushi)

[1747] Output: Prediction results sent to the device

[1748] Step 8:

[1749] The terminal displays meal suggestions to the user.

[1750] The device displays meal suggestions to the user, for example, "How about sushi for your next meal?"

[1751] Input: Prediction result sent to the device (e.g. sushi)

[1752] Output: Dining suggestions displayed to the user (e.g., How about sushi for your next meal?)

[1753] In this way, the system collects and analyzes the user's selection data, and supports the user in choosing meals by suggesting the meal they are likely to choose next.

[1754] (Application example 2)

[1755] Next, a description will be given of Application Example 2 of Form 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".

[1756] Conventional food delivery services lack the ability to suggest meals based on the user's preferences and past selection history, which means that users have to take the trouble of choosing the menu themselves every time. In addition, the suggested meals are not automatically customized based on the user's past selection history or feedback, making it difficult to increase user satisfaction.

[1757] The identification process by the identification processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for learning the user's preferences and past selection history, a means for proposing meals based on the learning results, a means for enabling customization of the proposed meals, and a means for vectorizing the user's past order history and calculating cosine similarity to predict the next meal to be proposed. This makes it possible to propose optimal meals based on the user's preferences and past selection history, and further to automatically customize the proposals.

[1758] "User preferences" are the characteristics of meals that a user is likely to like, inferred based on the user's past meal selections and preferences.

[1759] "Past selection history" is a record of meals previously selected or ordered by a user.

[1760] "Means of learning" refers to the methods and algorithms that allow the AI ​​to analyze data and understand the user's preferences based on the user's preferences and past selection history.

[1761] A "means for providing meal suggestions" is a method or system for presenting to a user the next meal that the user is likely to select based on the results of learning.

[1762] A "means for enabling customization of meal suggestions" is a method or system that allows a user to modify the suggested meals to suit their personal preferences.

[1763] "Vectorization" is the process of representing a user's past order history as numerical data and converting it into a computable form.

[1764] "Cosine similarity" is an index for calculating the similarity between vectorized data, and is a method for measuring the similarity based on the angle between two vectors.

[1765] A "means for predicting the next meal suggestion" is a method or system for predicting the next meal a user is likely to select based on the user's past ordering history and preferences.

[1766] The system for implementing this invention is an AI system that learns the user's preferences and past selection history and suggests meals based on them. A specific embodiment of this system will be described below.

[1767] First, the server collects the user's past order history and vectorizes it using the Python library CountVectorizer, which converts the user's order history into a numerical data format that can be calculated.

[1768] Next, the server calculates the cosine similarity based on the vectorized data. Cosine similarity is a method of measuring similarity based on the angle between two vectors, and is used to calculate the similarity between the user's past order history and the suggested meals.

[1769] The server predicts the next meal the user is likely to choose based on the calculated cosine similarity, and the prediction result is sent to the user's device and suggested to the user.

[1770] Users can customize the suggested meals, taking into account the user's past selections and feedback, resulting in meals that are optimized to the user's preferences.

[1771] For example, if a user has frequently ordered "sushi" and "tempura" in the past, the next menu item similar to "sushi" or "tempura" will be suggested. Users can take these suggestions and further customize them to suit their preferences.

[1772] Example of a prompt to be input to a generative AI model:

[1773] "Predict the next meal suggestion based on the user's past ordering history. The user's ordering history is: "Sushi", "Tempura", and "Ramen". What meal should be suggested next?"

[1774] This system can be implemented on a regular PC or smartphone. The software requires Python and its libraries (scikit-learn, numpy, json). This makes it possible to suggest optimal meals based on the user's preferences and past selection history, and further customize the suggestions automatically.

[1775] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1776] Step 1:

[1777] The server collects the user's past order history. Specifically, it retrieves data on meals the user has ordered in the past from the database. This data includes the order date and time, menu name, quantity, etc. The input is the user ID, and the output is the user's order history data.

[1778] Step 2:

[1779] The server vectorizes the collected order history data. Specifically, it uses Python's CountVectorizer to represent the order history as numerical data. This allows each menu to be represented as a numerical vector. The input is the order history data, and the output is the vectorized data.

[1780] Step 3:

[1781] The server calculates the cosine similarity based on the vectorized data. Specifically, it uses the cosine_similarity function to calculate the similarity between the user's past order history and the suggested meals. The input is the vectorized data, and the output is the similarity score.

[1782] Step 4:

[1783] The server predicts the next meal the user is likely to choose based on the calculated cosine similarity. Specifically, it selects the menu with the highest similarity score. The input is the similarity score, and the output is the predicted menu.

[1784] Step 5:

[1785] The server sends the predicted menu to the user's device. Specifically, it sends the prediction result to the user's smartphone app via API. The input is the predicted menu, and the output is the suggested menu displayed on the user's device.

[1786] Step 6:

[1787] The user customizes the proposed menu by viewing the proposed menu on a smartphone app and modifying it to suit their preferences. The input is the proposed menu, and the output is the customized menu.

[1788] Step 7:

[1789] The server receives the user's customization results and stores them in a database. Specifically, it records the customizations made by the user and reflects them in the next proposal. The input is the customized menu, and the output is the updated user selection history.

[1790] Example 3

[1791] Next, a third embodiment of the third embodiment will be described. 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 meal recommendation systems make suggestions based on the user's preferences and past selection history, but the problem is that the suggested meals are not easy to customize. In addition, when a user tries to customize the suggested meals to suit their preferences, the system may not be able to properly reflect the request. Another issue is that the customization process is complicated, resulting in a poor user experience.

[1793] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1794] In this invention, the server includes a means for receiving a user request, a means for sending a prompt sentence to the generative AI model to generate meal suggestions, a means for sending the generated meal suggestions to a user terminal and displaying them, a means for receiving a user customization request, a means for sending a prompt sentence to the generative AI model again based on the customization request to generate updated meal suggestions, and a means for sending the updated meal suggestions to the user terminal and displaying them, thereby enabling the user to easily customize the suggested meals to suit their preferences.

[1795] The "means for receiving a user request" is a function that enables the user to send a request for meal suggestions from the terminal to the server.

[1796] "Means for sending prompt sentences to the generative AI model to generate meal suggestions" refers to a function by which the server sends appropriate prompt sentences to the generative AI model based on a user request and generates meal suggestions.

[1797] "Means for transmitting and displaying the generated meal suggestions to the user terminal" refers to a function by which the server transmits meal suggestions obtained from the generative AI model to the user terminal and displays them on the terminal.

[1798] The "means for receiving a user's customization request" is a function that enables a user to request customization of a proposed meal, such as adding or deleting ingredients or changing the cooking method.

[1799] "Means for sending a prompt sentence to the regenerative AI model based on the customization request to generate updated meal suggestions" refers to a function by which the server receives a user's customization request, sends a prompt sentence to the regenerative AI model based on the customization request, and generates updated meal suggestions.

[1800] "Means for transmitting and displaying updated meal suggestions to the user terminal" refers to a function by which the server transmits updated meal suggestions obtained from the generative AI model to the user terminal and displays them on the terminal.

[1801] This invention is an AI system for assisting users in choosing meals, learning the user's preferences and past selection history, making meal suggestions based on the user's preferences and past selection history, and enabling the suggested meals to be customized. A specific embodiment of this system will be described below.

[1802] Hardware and Software Used

[1803] Hardware: Servers, user devices (PCs, smartphones, tablets, etc.)

[1804] Software: Generative AI models (e.g. GPT-4), database management systems (e.g. MySQL), web application frameworks (e.g. Django)

[1805] System Overview

[1806] 1. Receiving a User Request

[1807] A user uses the device to request meal suggestions.

[1808] Example: User types, "Dinner suggestions please."

[1809] 2. Meal suggestions using generative AI models

[1810] The server sends prompts to a generative AI model (e.g., GPT-4) to generate meal suggestions.

[1811] Example: Prompt: "Please provide dinner suggestions to the user. Offer healthy and balanced menu options."

[1812] 3. Display of suggested meals

[1813] The server transmits the generated meal suggestions to the user terminal and displays them to the user.

[1814] Example: "Dinner suggestion: grilled chicken and vegetable salad."

[1815] 4. Customize your meals

[1816] Users can customize the suggested meals by adding or removing ingredients, changing the cooking method, etc.

[1817] Example: User types, "Add tomatoes to my salad please."

[1818] 5. Reflecting customizations

[1819] The server receives the user's customizations and again sends prompts to the generative AI model to generate updated meal suggestions.

[1820] Example: Prompt: "The user would like to add tomatoes to the suggested salad. Please provide an updated menu."

[1821] 6. View updated meal suggestions

[1822] The server sends the updated meal suggestions to the user's terminal and displays them to the user.

[1823] Example: "Updated dinner suggestion: Grilled chicken and vegetable salad with tomatoes."

[1824] Examples

[1825] Examples of prompt statements

[1826] Initial suggestion: "Give users dinner suggestions. Offer healthy, balanced meals."

[1827] After customization: "The user would like to add tomatoes to the suggested salad. Please provide an updated menu."

[1828] User operation example

[1829] 1. User types, "Can I have some dinner suggestions?"

[1830] 2. The server sends a prompt to the generative AI model, suggesting "grilled chicken and vegetable salad."

[1831] 3. User types, "Add tomatoes to my salad."

[1832] 4. The server sends the updated prompt to the generative AI model, suggesting "grilled chicken and vegetable salad with tomatoes."

[1833] In this way, the user can customize the suggested meals to suit his / her preferences. This system allows the user to receive more satisfying meal suggestions, and improves the efficiency of meal selection. The flow of the identification process in the third embodiment will be described with reference to FIG. 15.

[1834] Step 1:

[1835] Receiving a user request

[1836] A user uses a terminal to request meal suggestions. The user types "Give me dinner suggestions" and presses the send button. The terminal sends this request to the server. The input is the user's request, and the output is the transmission of the request data to the server.

[1837] Step 2:

[1838] Generating meal suggestions with generative AI models

[1839] The server receives the user's request and sends a prompt to the generative AI model. The prompt is "Please give the user dinner suggestions. Please provide healthy and balanced menus." The generative AI model generates meal suggestions based on this prompt. The input is the prompt and the output is the generated meal suggestions.

[1840] Step 3:

[1841] Show suggested meals

[1842] The server receives the generated meal suggestion and sends it to the user terminal. The terminal displays the suggestion to the user. Specifically, the server sends the suggestion "grilled chicken and vegetable salad" to the user terminal, which displays it on the screen. The input is the generated meal suggestion, and the output is the display on the user terminal.

[1843] Step 4:

[1844] Receiving customised meal requests

[1845] A user requests customization of a proposed meal. The user types "add tomatoes to the salad" and presses the submit button. The device sends this customization request to the server. The input is the user's customization request, and the output is the transmission of the request data to the server.

[1846] Step 5:

[1847] Reflecting customizations

[1848] The server receives the user's customization request and sends a prompt sentence to the generative AI model again. The prompt sentence is "The user wants to add tomatoes to the proposed salad. Please provide an updated menu." The generative AI model generates updated meal suggestions based on this prompt sentence. The input is the prompt sentence based on the customization request, and the output is the updated meal suggestions.

[1849] Step 6:

[1850] View Updated Proposals

[1851] The server receives the updated meal suggestions and sends them to the user's device. The device displays the updated suggestions to the user. Specifically, the server sends the updated suggestion "Grilled chicken and vegetable salad with tomatoes" to the user's device, which displays it on its screen. The input is the updated meal suggestions and the output is the display on the user's device.

[1852] (Application example 3)

[1853] Next, a description will be given of Application Example 3 of Form Example 3. 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".

[1854] Conventional food delivery services have the problem that it is difficult for users to freely customize the suggested menu, and they are unable to fully meet the user's preferences and needs. In addition, the customization process requires users to manually enter information, which makes ordering less efficient. Furthermore, there was a need for a system that can make more personalized suggestions by utilizing the user's past selection history and feedback.

[1855] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1856] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for enabling customization of the suggested meals, a means for the user to input customization content, a means for sending the customization content as a prompt sentence to the generative AI model, a means for receiving a customized menu from the generative AI model, and a means for displaying the customized menu to the user. This allows the user to freely customize the suggested menu and order efficiently. In addition, the server can make more personalized suggestions by utilizing the user's past selection history and feedback.

[1857] "User preferences" refers to information about the user's preferences for meals, ingredients, cooking methods, etc. that the user has previously selected.

[1858] "Past selection history" is a record of the user's previous meal and customization selections.

[1859] A "means of learning" is an algorithm or system that analyzes user preferences and past choices to extract patterns and trends.

[1860] The "means for making meal suggestions" is a function for presenting suitable meal menus to the user based on the results of learning.

[1861] "Means to customize suggested meals" refers to functionality that allows users to add or remove ingredients from suggested menus, change cooking methods, etc.

[1862] The "means for inputting customization details" refers to an interface for inputting the customization details desired by the user into the system.

[1863] A "generative AI model" is an artificial intelligence model that generates customized menus based on user input.

[1864] A "prompt sentence" is a text-based input sentence that instructs the generative AI model on customization details.

[1865] The "means for receiving a customized menu" is a function for receiving a customized menu generated from a generative AI model.

[1866] The "means for displaying the customized menu to the user" refers to an interface for visually presenting the generated customized menu to the user.

[1867] The system for implementing the present invention learns the user's preferences and past selection history, makes meal suggestions based on the results of the learning, and enables customization of the suggested meals. A specific embodiment of this system is described below.

[1868] System Configuration

[1869] The system includes a user terminal, a server, and a generative AI model. The user terminal is a device such as a smartphone or tablet, and the server is located on the cloud. The generative AI model uses an advanced natural language processing model such as OpenAI's GPT-3.

[1870] Program Processing

[1871] User Device

[1872] The user device provides an interface for inputting the user's preferences and past selection history. Once the user receives meal suggestions, a screen is displayed for inputting customization details. Once the user has input the customization details, they are sent to the server.

[1873] server

[1874] The server has a database for learning user preferences and past selection history, receives the customization content sent by the user and sends it as a prompt to the generative AI model, receives the customized menu from the generative AI model and sends it to the user's device.

[1875] Generative AI Models

[1876] The generative AI model receives the prompt text sent by the server and generates a customized menu, which is returned to the server and finally displayed on the user's device.

[1877] Examples

[1878] For example, if a user is prompted for "pizza," the user can input "add toppings, more cheese." This customization is sent to the generative AI model as a prompt like this:

[1879] Examples of prompt statements

[1880] Menu suggestions: Pizza

[1881] Customizations: Add toppings, increase cheese

[1882] Generate your customized menu.

[1883] The generative AI model receives this prompt and generates a customized menu, for example, "Pizza with extra toppings and extra cheese." This menu is then displayed on the user's device via the server.

[1884] Hardware and Software Used

[1885] User devices: Smartphones, tablets

[1886] Server: Cloud server (e.g. AWS, Google Cloud)

[1887] Generative AI model: OpenAI GPT-3

[1888] Database: A database to store user preferences and past selections (e.g. MySQL, PostgreSQL)

[1889] In this way, users can customize the menu suggestions to their liking and order efficiently, while also leveraging the user's past selections and feedback to provide more personalized suggestions.

[1890] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1891] Step 1:

[1892] The user uses the device to receive meal suggestions.

[1893] Input: User preferences and past selections

[1894] Data processing: The server analyzes this data and generates a meal menu suitable for the user.

[1895] Output: The suggested meal menu is displayed on the user's device.

[1896] Step 2:

[1897] The user inputs customization details for the proposed menu.

[1898] Input: Suggested meal, user customizations (e.g., add toppings, increase cheese)

[1899] Data processing: The user device receives the customization content and generates data to send to the server.

[1900] Output: The customizations are sent to the server.

[1901] Step 3:

[1902] The server sends the customization content to the generative AI model as a prompt.

[1903] Input: User customizations

[1904] Data processing: The server converts the customization content into a prompt text and sends it to the generative AI model.

[1905] Output: The prompt is sent to the generative AI model.

[1906] Step 4:

[1907] A generative AI model takes the prompt and generates a customized menu.

[1908] Input: prompt statement

[1909] Data processing: The generative AI model analyzes the prompt text and generates a customized menu.

[1910] Output: The customized menu is returned to the server.

[1911] Step 5:

[1912] The server transmits the generated customized menu to the user terminal.

[1913] Input: Customized menu

[1914] Data processing: The server converts the generated menu into data to be sent to the user's terminal.

[1915] Output: The customized menu is sent to the user's device.

[1916] Step 6:

[1917] The user device displays the customized menu to the user.

[1918] Input: Customized menu

[1919] Data processing: The user device converts the received menu into data for visual display.

[1920] Output: The customized menu is displayed to the user.

[1921] In addition, 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.

[1922] "Example 1"

[1923] As one embodiment of the present invention, an AI system incorporating an emotion engine is provided. The system recognizes the user's emotions and makes meal suggestions based on the emotions. Specifically, if the user is expressing emotions of joy or excitement, the system suggests dishes that the user would like, such as sweets or celebratory meals. On the other hand, if the user is expressing emotions of depression or stress, the system suggests dishes that the user would find comforting, such as hot soup or familiar home-cooked meals.

[1924] "Example 2"

[1925] In another embodiment of the present invention, a system is provided in which the emotion engine adjusts the suggested meal customizations. Specifically, if the user is expressing emotions of happiness or excitement, the system may suggest customizations that the user would like, such as adding chocolate sauce or upgrading the dessert. On the other hand, if the user is expressing emotions of depression or stress, the system may suggest customizations that the user would like, such as adding ingredients or adjusting the thickness of the soup.

[1926] "Example 3"

[1927] Further, as another embodiment of the present invention, a system is provided in which an emotion engine simultaneously adjusts meal suggestions and customization based on the user's emotions. Specifically, when a user shows emotions of joy or excitement, the system simultaneously suggests a sweet dish that the user would like, such as adding chocolate sauce, while when a user shows emotions of depression or stress, the system simultaneously suggests a hot soup that the user would find reassuring, such as adding ingredients.

[1928] The process flow of each embodiment will be described below.

[1929] "Example 1"

[1930] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[1931] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[1932] Step 3: Suggested meals are presented to the user and the user selects from them.

[1933] "Example 2"

[1934] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[1935] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[1936] Step 3: Further tailor the suggested meals based on emotions.

[1937] Step 4: Tailored meal suggestions are presented to the user for them to choose from.

[1938] "Example 3"

[1939] Step 1: The AI ​​system recognizes the user's emotions using an emotion engine.

[1940] Step 2: Based on the recognized emotions, provide meal suggestions that the user might like.

[1941] Step 3: Further tailor the suggested meals based on emotions.

[1942] Step 4: Tailored meal suggestions and customization suggestions are presented simultaneously to the user, who can then choose from them.

[1943] Example 1

[1944] Next, a description will be given of Example 1 of the embodiment 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".

[1945] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they cannot suggest meals that take into account the user's emotional state, which means that the system is unable to sufficiently increase user satisfaction. In addition, the suggested meals are often customized manually, which is time-consuming for users.

[1946] 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.

[1947] In this invention, the server includes a means for learning the user's preferences and past selection history, a means for suggesting meals based on the results of the learning, a means for recognizing the user's emotions and suggesting meals based on the emotions, and a means for enabling customization of the suggested meals. This allows meals to be suggested taking into account the user's emotional state, thereby increasing the user's satisfaction. In addition, the suggested meals are automatically customized, reducing the user's effort.

[1948] "User preferences" refers to the user's particular food preferences based on the user's past meal selections and / or meal ratings.

[1949] "Past selection history" refers to data such as the type of meal a user has selected, the date and time, and their rating.

[1950] "Means of learning" refers to machine learning algorithms and databases that analyze users' preferences and past selection history to understand their eating trends.

[1951] "Means for making meal suggestions" refers to algorithms or systems that suggest appropriate meals to users based on the results of their learning.

[1952] "Means of recognizing emotions" refers to emotion analysis engines or software used to analyze a user's facial expressions and voice to identify the user's emotional state.

[1953] "Means enabling customization of meal suggestions" means an interface or system that allows a user to modify the suggested meals to suit their personal preferences.

[1954] "Feedback" refers to ratings and comments that users make on suggested meals.

[1955] This invention is an artificial intelligence system for assisting a user in selecting meals, learning the user's preferences and past selection history, and recognizing the user's emotions to suggest meals. A specific embodiment of this system will be described below.

[1956] System Configuration

[1957] Hardware and Software

[1958] The system consists of a server, a terminal, and a user. The server is equipped with a database and a machine learning algorithm, and the terminal is equipped with an emotion analysis engine. Specific software used is Microsoft Azure's Emotion API and Google Cloud's Natural Language API.

[1959] Data collection and learning

[1960] User data collection

[1961] A user inputs their daily meal selections into the terminal. For example, if the user chooses pasta for lunch, the user inputs that information into the terminal. The input data includes the type of meal, date and time, and rating (e.g., "delicious").

[1962] Accumulating and learning data

[1963] The server receives the user's meal data sent from the device and stores it in a database. The server uses the stored data to train a machine learning model. This model learns the user's eating habits and uses them to make future suggestions. For example, if the user tends to frequently choose pasta, the server learns that information.

[1964] Emotion recognition and meal suggestions

[1965] Emotion recognition

[1966] The device is equipped with an emotion engine to recognize the user's emotions. When the user speaks to the device, the device analyzes the user's facial expressions and voice to identify the user's emotional state. For example, if the user is smiling while speaking, the device recognizes that the user is expressing the emotion of happiness.

[1967] Meal suggestions

[1968] The server generates appropriate meal suggestions based on the emotion data sent from the device and the user's past selection history. For example, if the user expresses joy, the server will suggest sweets and celebratory meals that the user likes based on past data. Specifically, chocolate cakes and fruit tarts that the user has given high ratings in the past are suggested.

[1969] Examples

[1970] Consider a case where a user has been successful at work and is expressing emotions of joy. The user speaks to the device, saying, "I'm so happy today!" The device analyzes the user's facial expressions and voice to recognize the emotion of joy. The recognition results are sent to the server. The server then suggests meals that the user might like based on the received emotion data and past selection history. For example, it might suggest chocolate cake or fruit tart that the user has given a high rating in the past. The suggestions are sent to the device and displayed to the user.

[1971] Examples of prompt statements

[1972] Examples of prompts to be input to a generative AI model include:

[1973] "What are some examples of meals you would suggest when the user is expressing the emotion of pleasure? The user's past selections include chocolate cake and fruit tart."

[1974] Using this prompt, the generative AI model can generate appropriate meal suggestions based on the user's emotions and past selection history.

[1975] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1976] Step 1: Collect user data

[1977] A user inputs their daily meal selections into the terminal. For example, if the user chooses pasta for lunch, the user inputs that information into the terminal. The input data includes the type of meal, date and time, and rating (e.g., "delicious").

[1978] Input: Meal type, date, time, rating

[1979] Output: Meal data stored on the device

[1980] Step 2: Sending and storing data

[1981] The terminal transmits the collected meal data of the user to the server.

[1982] The server stores the received data in a database.

[1983] Input: Meal data sent from the device

[1984] Output: Meal data stored in a database

[1985] Step 3: Train the data

[1986] The server uses the user's food data stored in the database to train a machine learning model, which learns the user's eating habits and uses them to make future suggestions.

[1987] Input: Meal data stored in a database

[1988] Output: A trained machine learning model

[1989] Step 4: Recognize emotions

[1990] When a user speaks to the device, the device analyzes the user's facial expressions and voice to identify the user's emotional state. For example, if the user is smiling while speaking, the device recognizes that the user is expressing the emotion of joy. The emotion engine uses, for example, the Emotion API and Natural Language API.

[1991] Input: User's facial expression, voice

[1992] Output: Recognized emotion data

[1993] Step 5: Sending emotion data

[1994] The terminal transmits the recognized emotion data to the server.

[1995] Input: Recognized emotion data

[1996] Output: Emotion data sent to the server

[1997] Step 6: Meal Suggestions

[1998] The server generates appropriate meal suggestions based on the emotion data sent from the device and the user's past selection history. For example, if the user expresses joy, the server will suggest sweets and celebratory meals that the user likes based on past data.

[1999] Input: Emotion data, past selection history

[2000] Output: Suggested meals

[2001] Step 7: Viewing Proposals

[2002] The terminal displays the suggestions sent by the server to the user, who can then review the suggested meals and customize them if necessary.

[2003] Input: Suggested meals

[2004] Output: The suggestions displayed to the user

[2005] Examples

[2006] Consider a case where a user has been successful at work and is expressing emotions of joy. The user speaks to the device, saying, "I'm so happy today!" The device analyzes the user's facial expressions and voice to recognize the emotion of joy. The recognition results are sent to the server. The server then suggests meals that the user might like based on the received emotion data and past selection history. For example, it might suggest chocolate cake or fruit tart that the user has given a high rating in the past. The suggestions are sent to the device and displayed to the user.

[2007] (Application example 1)

[2008] Next, a description will be given of Application Example 1 of Form 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".

[2009] Conventional meal selection support systems can suggest meals based on the user's preferences and past selection history, but they have the problem that it is difficult to provide meals that match the user's current mood because they cannot make suggestions that take into account the user's emotional state. In addition, there is a problem that it is not possible to improve user satisfaction because there is a lack of means to recognize the user's emotions in real time and suggest appropriate meals based on them.

[2010] 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.

[2011] In this invention, the server includes means for learning the user's preferences and past selection history, means for suggesting meals based on the results of the learning, means for enabling customization of the suggested meals, means for recognizing the user's emotions and suggesting meals based on the emotions, and means for recognizing the user's emotions in real time using the smartphone camera, thereby enabling meal suggestions according to the user's current emotional state and improving user satisfaction.

[2012] "User preferences" refers to the meals selected by the user in the past and the user's particular preferences for meals.

[2013] "Past selection history" refers to a record of meals previously selected by a user.

[2014] "Means of learning" refers to the function of accumulating data on user preferences and past selection history, and analyzing this to find patterns.

[2015] "Means for making meal suggestions" refers to a function that suggests suitable meals to the user based on the results of learning.

[2016] "Means enabling customization of meal suggestions" refers to functionality that allows a user to modify the suggested meals to suit their personal preferences.

[2017] "Means of recognizing the user's emotions" refers to the function of analyzing the user's emotions from their facial expressions, voice, etc., and determining their state.

[2018] "Means for making meal suggestions based on emotions" refers to a function that suggests appropriate meals based on the recognized emotional state of the user.

[2019] "Means of recognizing a user's emotions in real time using a smartphone camera" refers to a function that uses a smartphone camera to analyze a user's facial expressions and determine their emotions in real time.

[2020] As an embodiment ...

Claims

1. A system for assisting a user in selecting a meal, comprising: A means for performing machine learning of a learning model using a user's food preferences and past meal menu selection history; A means for generating a meal menu to be suggested to a user using the learning model; means for recognizing the emotional state of the user by analyzing the user's facial expressions and voice using an emotion engine; means for generating a prompt sentence indicating a meal suggestion that the current user would like based on the recognized current emotional state of the user and the meal menu generated by the learning model; A means for suggesting meals to a user using the generated prompt sentence and the generative AI model; A system including:

2. and means for automatically customizing the suggested meals based on a user's past selection history. The system of claim 1 .

3. and means for customizing the suggested meals based on feedback from a user. The system of claim 1 .

Images