System

Abstract

To solve a burden for a user to think about a daily dish.SOLUTION: A system includes: an interactive AI for providing a recipe on the basis of an input of a user; means for instructing a cooking procedure of the recipe through a dialogue with a user; means for receiving feedback from the user and learning the feedback; means for proposing a dish which suits taste of the user, and ordering delivery; means for providing a cooking process of the proposed dish and detailed information on ingredients by an image and video; means for inquiring of the interactive AI about a question as to the dish from the user; and means for providing the user with the recipe updated by the interactive AI as an answer to the question.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] In daily cooking, it is a big burden for users to think about what to make and how to make it. Also, it is difficult to provide recipes that match the user's preferences. [Means for solving the problem]

[0005] As a means of solving this problem, we provide a conversational AI that provides recipes and cooking instructions based on user input. Furthermore, by learning from user feedback and evolving the recipe based on that, it becomes possible to provide recipes that are tailored to the user's preferences. [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 the 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 the 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] One embodiment of the present invention is a smartphone application equipped with a conversational AI. The user converses with the AI ​​through the application and provides information about the day's meals. For example, the user can tell the AI ​​their preferences, such as "I want to eat Japanese food today" or "Please tell me about some dishes that use chicken." Based on this information, the AI ​​provides recipes that match the user's preferences. The recipes are not just text, but also use images and videos to explain the cooking steps in detail. Furthermore, if the user has any questions or problems during the cooking process, they can directly contact the AI.

[0029] "Example 2"

[0030] Furthermore, another embodiment of the present invention is a function that allows the AI ​​to learn from feedback from users. After cooking, the user can tell the AI ​​their thoughts and suggestions for improvement, and the AI ​​will learn from this information. For example, if the AI ​​receives feedback such as "I would like it less spicy" or "I would like to use more vegetables," the AI ​​will provide recipes that reflect these requests from the next time. This makes it possible to provide recipes that are tailored to the user's preferences.

[0031] "Example 3"

[0032] In addition, as a further embodiment of the present invention, there is a function where the AI ​​refers to the user's past conversation history regarding food. The AI ​​learns the user's tastes and preferences from past conversations with the user. For example, if the user has previously mentioned that they "like spicy food," the AI ​​can remember that information and provide recipes for spicy dishes. This makes it possible to provide recipes that more closely correspond to the user's tastes.

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

[0034] "Example 1"

[0035] Step 1: The user opens the smartphone application and starts a conversation with the conversational AI.

[0036] Step 2: The user provides the AI ​​with information about the day's meals. For example, the user may tell the AI ​​a request such as, "I want to eat Japanese food today," or "Please tell me about dishes that use chicken."

[0037] Step 3: Based on the information provided by the user, the AI ​​will select an appropriate recipe and provide the recipe and cooking instructions to the user.

[0038] Step 4: If the user has any questions or problems during the cooking process, they can directly contact the AI, which will provide an appropriate answer to the question.

[0039] "Example 2"

[0040] Step 1: After cooking, the user tells the AI ​​what they think and what improvements they would like to make.

[0041] Step 2: The AI ​​learns from user feedback and incorporates that information into its next recipe selection.

[0042] "Example 3"

[0043] Step 1: The AI ​​learns the user's preferences and tastes from past conversations with them.

[0044] Step 2: When a user requests a new recipe, the AI ​​will refer to past conversation history to provide recipes that match the user's preferences.

[0045] Example 1

[0046] 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".

[0047] Conventional recipe provision systems have difficulty in providing recipes that meet the specific needs of users, and are unable to respond quickly and appropriately to follow-up questions during the cooking process. Furthermore, they do not adequately consider the user's preferences or past conversation history when providing recipes. This leads to a decrease in user satisfaction and a decrease in the frequency of system use.

[0048] 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.

[0049] In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for generating a recipe that meets the user's needs using a generative AI model, a means for providing the generated recipe to the user in text, image, or video format, and a means for generating an answer using the generative AI model again when the user asks an additional question during the cooking process. This makes it possible to provide a recipe according to the user's specific needs, and to quickly and appropriately respond to additional questions during the cooking process. It also makes it possible to provide a recipe that takes into account the user's preferences and past conversation history, thereby improving user satisfaction.

[0050] "User" means an individual or organization that uses the system to receive recipes.

[0051] "Input" refers to the provision of information, such as text or voice, by a user to a system.

[0052] A "recipe" is information that lists the ingredients and steps for making a particular dish.

[0053] "Conversational AI" is an artificial intelligence system that provides information to users through natural language dialogue.

[0054] "Feedback" refers to reactions such as evaluations and opinions that users provide to a system.

[0055] A "generative AI model" is an artificial intelligence model that generates new information based on user requests.

[0056] A "prompt" is an input sentence that prompts a generative AI model to generate specific information.

[0057] "Text" is data that includes character information.

[0058] An "image" is data that contains visual information.

[0059] "Video" is data that contains dynamic visual information.

[0060] A "cooking procedure" is the sequence of specific steps to prepare a dish.

[0061] A "server" is a computer that is the central part of a system and is a device that processes and stores data.

[0062] A "terminal" is a device that is directly operated by a user and is a device that connects to the system to exchange information.

[0063] "Analysis" is the process of understanding input information and extracting meaning from it.

[0064] An "HTTP request" is a protocol for sending and receiving data over the Internet.

[0065] This invention is a system that uses a smartphone application equipped with a conversational AI to provide recipes according to the user's request. A specific embodiment of this system will be described below.

[0066] First, a user launches a smartphone application. The device connects to a server via the Internet and loads a conversational AI model. This conversational AI model includes a natural language processing library (e.g., NLTK or spaCy) and a generative AI model (e.g., GPT-3 (registered trademark)).

[0067] The user inputs a request, such as "I want to eat Japanese food today" or "Tell me about some dishes that use chicken," into the text input field of the application. The device sends this input as text data to the server. An HTTP request is used for transmission.

[0068] The server analyzes the received text data using a natural language processing library to understand the user's intent. Based on the analysis results, it generates and sends an appropriate prompt to the generative AI model. For example, it generates the following prompt sentence:

[0069] User request: I want to eat Japanese food today

[0070] Prompt for generative AI model: Generate a recipe for a Japanese dish. Please provide detailed instructions for cooking and a list of ingredients needed to make it easy for users to follow. Also include links to images and videos that show the cooking steps.

[0071] The generative AI model receives the prompts and generates a recipe that matches the user's needs. The generated recipe includes cooking instructions, a list of ingredients needed, and links to related images and videos. The server sends the generated recipe information to the device, again using an HTTP request.

[0072] The device displays the received recipe information on the user's smartphone screen. The user can cook the dish while checking the text, images, and videos. For example, if the user enters "I want to eat Japanese food today," the server sends the following prompt to the generative AI model:

[0073] Prompt: The user says they want to eat Japanese food. Generate an easy-to-make Japanese recipe. Describe the cooking steps in detail and provide a list of ingredients needed. Also include links to images and videos that show the cooking steps.

[0074] The generative AI model generates a recipe for "Teriyaki Chicken" and the server sends it to the device, which displays it to the user as follows:

[0075] Recipe: Chicken Teriyaki

[0076] Ingredients: chicken thighs, soy sauce, mirin, sugar, sake

[0077] procedure:

[0078] 1. Cut the chicken thighs into bite-sized pieces.

[0079] 2. Heat oil in a frying pan and fry the chicken.

[0080] 3. Add the sauce made from soy sauce, mirin, sugar and sake and bring to a boil.

[0081] 4. When it's done, arrange it on a plate.

[0082] Image link: [Chicken Teriyaki Image]

[0083] Video link: [Chicken Teriyaki Cooking Video]

[0084] If the user has any questions or difficulties during the cooking process, they again enter their question into the application's text input field, and the server sends the following prompt to the generative AI model:

[0085] Prompt: The user says they don't know the amount of sauce. Please explain in detail the amount of sauce in the chicken teriyaki.

[0086] The generative AI model generates the answer "2 tablespoons soy sauce, 2 tablespoons mirin, 1 tablespoon sugar, 1 tablespoon sake" and the server sends it to the device. The device displays it to the user as follows:

[0087] Answer: The amount of sauce for chicken teriyaki is 2 tablespoons of soy sauce, 2 tablespoons of mirin, 1 tablespoon of sugar, and 1 tablespoon of sake.

[0088] In this way, users can easily get cooking recipes through the application and also receive support during the cooking process.

[0089] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0090] Step 1:

[0091] The user launches an application.

[0092] Specific operation: The user taps the application icon on the smartphone's home screen to launch it.

[0093] Input: User actions

[0094] Output: Application launch

[0095] Step 2:

[0096] The device connects to the server and loads the conversational AI model.

[0097] Specific operation: The device connects to a server via the internet and loads a conversational AI model (e.g. GPT-3).

[0098] Input: Start application

[0099] Output: Conversational AI model loaded

[0100] Step 3:

[0101] The user inputs a request.

[0102] Specific operation: The user inputs a request such as "I want to eat Japanese food today" or "Please tell me some recipes that use chicken" into a text input field of the application.

[0103] Input: User's request text

[0104] Output: The requested text

[0105] Step 4:

[0106] The terminal transmits the input request to the server.

[0107] Specific operation: The terminal sends the user's input as text data to the server using an HTTP request.

[0108] Input: The requested text

[0109] Output: HTTP request to the server

[0110] Step 5:

[0111] The server analyzes the request and sends prompts to the generative AI model.

[0112] Specific operation: The server analyzes the received text data using a natural language processing library (e.g., spaCy) to understand the user's intent. Based on the analysis results, it generates and sends an appropriate prompt to the generative AI model.

[0113] Input: Request text received in HTTP request

[0114] Output: prompts to the generative AI model

[0115] Step 6:

[0116] A generative AI model generates the recipe.

[0117] What it does: A generative AI model (e.g. GPT-3) takes the prompt and generates a recipe tailored to the user's needs, including cooking instructions, a list of needed ingredients, and links to related images and videos.

[0118] Input: A prompt for the generative AI model

[0119] Output: Generated recipe information

[0120] Step 7:

[0121] The server sends the generated recipe to the terminal.

[0122] Specific operation: The server sends the generated recipe information to the terminal, again using an HTTP request.

[0123] Input: Generated recipe information

[0124] Output: HTTP request to the terminal

[0125] Step 8:

[0126] The terminal displays the recipe to the user.

[0127] Specific operation: The device displays the received recipe information on the user's smartphone screen. The user can then prepare the dish while checking the text, images, and videos.

[0128] Input: Recipe information received via HTTP request

[0129] Output: The recipe displayed on the user's smartphone screen

[0130] Step 9:

[0131] If the user has any questions or problems while cooking, they can contact the AI ​​again.

[0132] Specific operation: If the user has any questions or problems while cooking, they enter their questions into the application's text input field again.

[0133] Input: User's additional question text

[0134] Output: The additional question text entered

[0135] Step 10:

[0136] The server generates answers to the additional questions and sends them to the terminal.

[0137] Specific operation: The server receives the user's additional question, again uses the generative AI model to generate an appropriate answer, and sends the generated answer to the terminal, which displays it to the user.

[0138] Input: Additional question text received in HTTP request

[0139] Output: Generated answer information and HTTP request to the terminal

[0140] (Application example 1)

[0141] 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."

[0142] Conventional recipe provision systems using conversational AI can provide recipes that match the user's preferences, but they do not provide sufficient support when the user actually cooks the recipe. In addition, there is no way for the user to immediately enjoy the suggested dishes. Furthermore, there is a lack of a way for the user to respond in real time when they have questions or problems about cooking. This reduces user convenience and leads to issues with satisfaction.

[0143] 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.

[0144] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for suggesting dishes that suit the user's preferences and ordering them for delivery, a means for providing detailed information on the cooking process and ingredients of the suggested dishes in the form of images and videos, and a means for the user to inquire of the AI ​​if they have any questions or problems regarding cooking. This allows the user to not only receive suggestions for dishes that suit their preferences, but also to immediately order the dishes for delivery, and furthermore, to visually check the cooking process and detailed information on the ingredients, thereby enhancing cooking support. In addition, since questions can be answered in real time, the convenience and satisfaction of the user is improved.

[0145] "Conversational AI" is an artificial intelligence system that provides information through natural language dialogue with users and generates responses based on the user's requests.

[0146] A "recipe" is information that describes the ingredients and cooking steps for making a particular dish.

[0147] "Cooking instructions" are instructions that show the specific operations and steps required to prepare a dish.

[0148] "Feedback" refers to information provided by users, such as opinions, ratings, and impressions after use.

[0149] A "delivery order" is an ordering procedure for delivering the food selected by the user to a specified location.

[0150] A "cooking process" is a series of cooking operations or steps performed to complete a dish.

[0151] "Detailed information about ingredients" refers to specific information such as the type, amount, nutritional value, and origin of the ingredients used in the dish.

[0152] An "image" is a still image used to visually convey information.

[0153] A "video" is a moving image that conveys information visually.

[0154] An "inquiry" is a question or inquiry made by a user to resolve a concern or problem.

[0155] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, a means for suggesting dishes that suit the user's preferences and placing a delivery order, a means for providing images and videos of the cooking process and detailed information about ingredients of the suggested dishes, and a means for the user to contact the AI ​​if they have any questions or problems regarding cooking.

[0156] Program processing explanation

[0157] Hardware used

[0158] Smartphone

[0159] server

[0160] Software used

[0161] Python

[0162] OpenAI(R) API

[0163] Data processing and calculation

[0164] The server receives user input and generates an appropriate recipe using a generative AI model. Specifically, when a user inputs "I want to eat Japanese food today" through their smartphone, the information is sent to the server. The server uses the OpenAI API to generate a recipe based on the user's request. This generated recipe is provided to the user in the form of text, images, and videos.

[0165] When a user places a delivery order for the suggested food, the server sends the order information to the delivery service, allowing the user to enjoy the suggested food immediately. If the user has any problems during the cooking process, they can contact the AI ​​via their smartphone. The AI ​​will respond to questions in real time and support the user.

[0166] Examples

[0167] If a user types "I want to eat Japanese food today," the server uses the OpenAI API to generate a recipe for "Teriyaki Chicken." This recipe is provided to the user in the form of text, images, and videos. If the user answers "Yes," the server sends the order information to a delivery service, and the food is delivered to the user's location.

[0168] Examples of prompt statements

[0169] What recipes would you suggest when a user says, "I want to eat Japanese food today"?

[0170] In this way, users can not only receive suggestions for dishes that suit their tastes, but also order those dishes for delivery immediately, and can visually check detailed information about the cooking process and ingredients, providing more support for cooking.In addition, users can have their questions answered in real time, improving convenience and satisfaction for users.

[0171] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0172] Step 1:

[0173] A user inputs a request for a meal using a smartphone terminal. For example, the user inputs, "I want to eat Japanese food today." This input data is text data including the user's preferences and requests.

[0174] Step 2:

[0175] The device sends the user's input data to the server. The server receives this input data and inputs it as a prompt sentence into the generative AI model. An example of a prompt sentence is, "Please tell me the recipe of the dish that you would suggest when the user says, 'I want to eat Japanese food today.'"

[0176] Step 3:

[0177] The server uses a generative AI model (OpenAI API) to generate a recipe based on the prompt text. This generated recipe is output as text data. For example, a recipe for "Teriyaki Chicken" is generated.

[0178] Step 4:

[0179] The server retrieves relevant media data from a database to complement the generated recipe in the form of images and videos, thus providing visual information to the user.

[0180] Step 5:

[0181] The server sends the generated recipe and related images and videos to the terminal. The terminal displays this data to the user. The user checks the proposed recipe and selects whether to place a delivery order.

[0182] Step 6:

[0183] When a user places a delivery order, the terminal sends the order information to the server. The server then sends the order information to the delivery service and arranges for the food to be delivered. The order information includes the user's address and the order details.

[0184] Step 7:

[0185] If a user has any problems during the cooking process, they can ask the AI ​​through their device. The server receives the user's question and uses the generative AI model to generate an appropriate response. This response is output as text data and sent to the device.

[0186] Step 8:

[0187] The device displays the response received from the server to the user, who can then continue cooking while referring to the instructions and advice from the AI.

[0188] In this way, users can not only receive suggestions for dishes that suit their tastes, but also order those dishes for delivery immediately, and can visually check detailed information about the cooking process and ingredients, providing more support for cooking. In addition, users can have their questions answered in real time, improving convenience and satisfaction for users.

[0189] Example 2

[0190] 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."

[0191] Conventional recipe provision systems have difficulty in fully reflecting individual user preferences and feedback, and have been unable to increase user satisfaction. In addition, there was a lack of a mechanism to effectively learn from user feedback and reflect it in the next recipe, making it difficult to provide recipes that match the user's preferences.

[0192] 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.

[0193] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning it, a means for analyzing the user's feedback using a generative AI model and reflecting it in the next recipe, a terminal means for transmitting the user's feedback to the server, a means for the server to input the feedback to the generative AI model and generate a customized recipe, and a means for transmitting the recipe generated by the server to the terminal. This makes it possible to effectively learn the individual preferences and feedback of the user and reflect them in the next recipe.

[0194] "User input" refers to information such as text or voice provided by a user through a terminal.

[0195] A "recipe" is information that includes instructions for making a dish and a list of ingredients.

[0196] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through dialogue with a user and responds according to the user's requests.

[0197] The "means for instructing cooking steps" is a function for explaining and instructing the user on cooking steps.

[0198] "Feedback" refers to impressions and suggestions for improvement provided by users after cooking.

[0199] A "generative AI model" is an artificial intelligence model that analyzes user feedback and reflects it in the next recipe.

[0200] The "terminal means" is a device that allows a user to input feedback and transmit it to the server.

[0201] The "server" is a computer system that receives user feedback, generates customized recipes using a generative AI model, and sends them to the terminal.

[0202] A "customized recipe" is a cooking procedure or list of ingredients that has been adjusted based on user feedback and preferences.

[0203] The present invention is a system that provides recipes based on user input, learns feedback from the user, and reflects the feedback in the next recipe. A specific embodiment of this system will be described below.

[0204] First, a conversational AI means is used to provide a recipe for the user to cook. The AI ​​means instructs the user on the cooking steps of the recipe through dialogue with the user. For example, when the user inputs "I want to make spaghetti," the AI ​​means provides a recipe for spaghetti and instructs the user on the cooking steps step by step.

[0205] Next, after the user has finished cooking, they input their feedback using a device (such as a smartphone or tablet). The feedback includes specific impressions and suggestions for improvement, such as "I wish it was less spicy" or "I wish it had more vegetables." The device then sends this feedback to the server.

[0206] The server inputs the received feedback into a generative AI model, such as OpenAI's GPT-4 (registered trademark). The model analyzes the feedback and learns the user's preferences and requests. The next time the user requests a recipe, the server generates a customized recipe based on the learned information and sends it to the device.

[0207] For example, if a user provides feedback such as "I would like it less spicy," the device sends this feedback to the server. The server inputs the feedback into a generative AI model, which then generates a prompt such as "Please make the recipe less spicy next time." The server generates a recipe for spaghetti with less spiciness based on this prompt and sends it to the device. The user can then use this customized recipe to cook next time.

[0208] In this way, the system can effectively learn from users' individual preferences and feedback and incorporate them into future recipes, thereby increasing user satisfaction.

[0209] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0210] Step 1:

[0211] The user cooks

[0212] A user cooks a dish using a provided recipe. For example, to make spaghetti, the user prepares the ingredients and cooks the dish according to the recipe. The input is the provided recipe and the output is the finished dish.

[0213] Step 2:

[0214] User enters feedback

[0215] After cooking is finished, the user uses the terminal to input feedback. For example, they input specific impressions and suggestions for improvement, such as "I'd like it to be less spicy" or "I'd like to use more vegetables." The input is the user's impressions and suggestions for improvement, and the output is the feedback data.

[0216] Step 3:

[0217] The device sends feedback to the server

[0218] The terminal transmits the feedback input by the user to the server. At this time, the feedback is transmitted as text data. The input is the feedback data, and the output is the feedback transmitted to the server.

[0219] Step 4:

[0220] The server inputs the feedback into the generative AI model

[0221] The server inputs the received feedback into a generative AI model. For example, when using OpenAI's GPT-4, the feedback is passed to the model as a prompt sentence. The input is the feedback data, and the output is the prompt sentence.

[0222] Step 5:

[0223] Generative AI models learn from feedback

[0224] The generative AI model analyzes the feedback and learns the user's preferences and requests. For example, if it receives feedback such as "Make it less spicy," it will learn to make the recipe less spicy the next time it generates a recipe. The input is the prompt, and the output is the learning result.

[0225] Step 6:

[0226] The server generates a customized recipe

[0227] The server uses the information from the generative AI model to generate a customized recipe based on the user's preferences. For example, it creates a recipe for spaghetti with less spice. The input is the learning result, and the output is the customized recipe.

[0228] Step 7:

[0229] Send the recipe generated by the server to the terminal

[0230] The server sends the generated customized recipe to the terminal. The recipe is sent as text data. The input is the customized recipe, and the output is the recipe sent to the terminal.

[0231] Step 8:

[0232] User receives customized recipe

[0233] The user receives the customized recipe on the terminal and uses it for the next cooking. For example, the user can check the recipe for spaghetti with less spice to help with the next cooking. The input is the recipe sent to the terminal, and the output is the customized recipe received by the user.

[0234] (Application example 2)

[0235] 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."

[0236] Conventional food delivery systems have a problem in that it is difficult to suggest recipes that fully reflect the user's preferences and feedback. In particular, even if a user provides specific feedback on a dish, there is a lack of a mechanism to reflect that feedback in the next order, making it difficult to improve user satisfaction.

[0237] 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. In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, and a means for proposing the next recipe based on the user's feedback. This makes it possible to propose recipes that reflect the user's preferences and feedback.

[0238] "User input" means information or instructions provided by a user to a system.

[0239] A "recipe" is a list of ingredients and steps for making a particular dish.

[0240] "Conversational AI" is an artificial intelligence system that uses natural language processing technology to interact with users and provide them with information.

[0241] A "cooking procedure" is the order or method of specific steps required to prepare a dish.

[0242] "Feedback" refers to thoughts and suggestions for improvement provided by users regarding the system.

[0243] "Means of learning" refers to the techniques and methods by which the system analyzes user feedback and incorporates it into its next suggestions.

[0244] "Recipe suggestions" refers to suggestions of cooking ingredients and steps provided to the user by the system.

[0245] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, and a means for suggesting the next recipe based on the user's feedback.

[0246] System Configuration

[0247] The system uses the following hardware and software:

[0248] Hardware: Smartphone

[0249] Software: Python, scikit-learn

[0250] Program Processing

[0251] The server receives input from the user and generates a recipe based on it. It instructs the user on the cooking steps through dialogue. After the user finishes cooking, they can provide feedback. This feedback is collected by the server and saved as learning data. The next time a recipe is proposed, a recipe tailored to the user's preferences will be suggested based on this feedback.

[0252] Examples

[0253] For example, if a user provides feedback such as "I want it less spicy," the next recipe will be suggested with less spiciness. This makes it possible to suggest recipes that are tailored to the user's preferences.

[0254] Examples of prompt statements

[0255] Below are examples of prompts to be input to a generative AI model.

[0256] User ID: user123

[0257] Feedback: I would like it to be less spicy

[0258] Next suggestion: A less spicy recipe

[0259] In this way, recipe suggestions that reflect user feedback are realized.

[0260] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0261] Step 1:

[0262] A user accesses the system using a smartphone and requests a recipe. The user's input (request content) is sent to the server. The server receives this input and starts processing it as initial data for generating a recipe.

[0263] Step 2:

[0264] The server uses conversational AI to generate specific recipes through dialogue with the user. The conversational AI uses natural language processing technology to analyze the user's request and select an appropriate recipe. In this process, the user's preferences and past history are also taken into account. The input is the user's request, and the output is the generated recipe.

[0265] Step 3:

[0266] The server provides the generated recipe to the user and instructs the cooking procedure. The user cooks while checking the cooking procedure through their smartphone. The input is the generated recipe, and the output is the cooking procedure instruction to the user.

[0267] Step 4:

[0268] After the user finishes cooking, he / she provides feedback. The feedback is sent to the server via the smartphone. The input is the user's feedback, and the output is the feedback data stored in the server.

[0269] Step 5:

[0270] The server analyzes the received feedback and saves it as training data. Specifically, the feedback content is divided into features and evaluation, and training is performed using a linear regression model. The input is the feedback data, and the output is the updated training model.

[0271] Step 6:

[0272] The next time the user requests a recipe, the server uses the learned model to suggest recipes that match the user's preferences. The server takes into account the previous feedback and generates a recipe optimized for the user's preferences. The input is the user's new request and the learned model, and the output is the optimized recipe.

[0273] In this way, recipe suggestions that reflect user feedback are realized.

[0274] Example 3

[0275] 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."

[0276] Conventional conversational AI systems have had difficulty providing recipes that fully reflect the user's tastes and preferences. In addition, they lacked the means to effectively utilize the user's past conversation history, and the recipes they provide based on the user's preferences were limited. Furthermore, they lacked the ability to effectively learn from user feedback and evolve recipes.

[0277] 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.

[0278] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for storing the user's past conversation history about meals in a database, a means for analyzing the stored conversation history and learning the user's preferences, a means for using a generative AI model for generating recipes based on the user's preferences, and a means for providing the generated recipes to the user. This makes it possible to provide recipes that deeply correspond to the user's preferences and tastes.

[0279] A "conversational artificial intelligence means" is a technology that provides information through natural language dialogue with a user and generates responses according to the user's requests.

[0280] A "means for instructing cooking procedures" is a technology that provides users with step-by-step instructions on each step of a recipe, supporting them in the progress of cooking.

[0281] "Means of receiving feedback and learning from it" refers to technology that collects opinions and evaluations provided by users and uses them to improve the system's performance and content.

[0282] "Means for storing past conversation history in a database" refers to a technology that records the content of past conversations with a user and stores it in a database for future reference.

[0283] "Means for analyzing conversation history and learning user preferences" refers to technology that analyzes stored conversation history to identify user preferences and tastes and reflects them in the system.

[0284] "Means of using generative AI models" refers to technologies that utilize machine learning models to generate appropriate information and content based on user preferences and requests.

[0285] A "means for generating recipes" is a technology that automatically creates cooking recipes that meet a user's requests and preferences.

[0286] A "means for providing recipes" is a technology that presents the generated recipes to users and enables them to view and use them.

[0287] This invention relates to a conversational AI system that provides recipes based on user input. The system has the function of instructing the cooking steps of a recipe through conversation with the user, receiving feedback from the user, and learning from it. The system also stores the user's past conversation history about meals in a database, and analyzes the stored conversation history to learn the user's preferences. Furthermore, the system uses a generative AI model that generates recipes based on the user's preferences, and provides the generated recipes to the user.

[0288] Hardware and Software Used

[0289] Hardware:

[0290] Server: Manages the database, runs the generative AI models, and provides the interface with the user.

[0291] Terminal: A device such as a smartphone or PC used by a user.

[0292] software:

[0293] Database management system: Stores user conversation history and preference information.

[0294] Generative AI models: For example, using natural language processing models such as GPT-4.

[0295] Conversational artificial intelligence system: manages dialogue with the user, providing recipes and cooking instructions.

[0296] Program Processing

[0297] The server executes a conversational artificial intelligence means that provides recipes based on user input. When the user says "I like spicy food" through the terminal, the terminal transmits the statement to the server. The server stores the statement in a database and analyzes it using a generative AI model. As a result of the analysis, it is determined that the user likes spicy food.

[0298] Later, when the user requests, "What recipes do you recommend for dinner tonight?", the server inputs the following prompt into the generative AI model:

[0299] Example prompt:

[0300] You know from the user's past conversation history that they like spicy food, so provide them with some recommended spicy food recipes.

[0301] The generative AI model generates a recipe that matches the user's preferences based on this prompt. For example, a recipe for spicy chicken curry is generated. The server sends the generated recipe to the terminal, and the terminal displays to the user, "Here is the recipe for spicy chicken curry."

[0302] In this way, a system is realized in which the server, the terminal, and the user work together to provide recipes that match the user's preferences. The flow of the specification process in the third embodiment will be described with reference to FIG.

[0303] Step 1:

[0304] A user starts a dialogue with the conversational AI system through a terminal. For example, the user says, "I like spicy food."

[0305] Input: What the user says

[0306] Output: User utterance data

[0307] Step 2:

[0308] The terminal transmits the user's utterance data to the server, which then stores the received utterance data in a database.

[0309] Input: User utterance data

[0310] Output: Speech data stored in a database

[0311] Step 3:

[0312] The server periodically analyzes the conversation history in the database and uses a generative AI model to extract keywords and phrases from users' utterances using natural language processing techniques.

[0313] Input: Conversation history in a database

[0314] Output: Extracted keywords and phrases

[0315] Step 4:

[0316] The server learns the user's preferences based on the extracted keywords and phrases. For example, based on the information "I like spicy food," the server recognizes that the user likes spicy food. This information is stored in a database as a user profile.

[0317] Input: Extracted keywords or phrases

[0318] Output: Updated user profile

[0319] Step 5:

[0320] The user requests a new recipe through the terminal, for example, by saying, "Tell me a recommended recipe for tonight's dinner."

[0321] Input: User request

[0322] Output: User request data

[0323] Step 6:

[0324] The server receives the user's request data and inputs a prompt sentence into the generative AI model. The prompt sentence includes the user's preference information. For example, the prompt sentence might say, "We know from the user's past conversation history that the user likes spicy food. Please provide some recommended spicy food recipes for the user."

[0325] Input: User request data, user preferences

[0326] Output: Prompt statement

[0327] Step 7:

[0328] The generative AI model uses the prompt text to generate recipes that match the user's preferences, for example a recipe for spicy chicken curry.

[0329] Input: prompt statement

[0330] Output: The generated recipe

[0331] Step 8:

[0332] The server transmits the generated recipe to the terminal. The terminal displays the recipe to the user. For example, the recipe is provided in the form of "Here is the recipe for spicy chicken curry."

[0333] Input: The generated recipe

[0334] Output: The recipe displayed to the user

[0335] In this way, a system is realized in which the server, terminal, and user work together to provide recipes that suit the user's preferences.

[0336] (Application example 3)

[0337] 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."

[0338] Conventional recipe provision systems have difficulty in providing recipes that fully reflect the user's tastes and preferences, and therefore have been unable to increase user satisfaction. In addition, they lacked the functionality to suggest restaurants and menus that the user would like, and therefore were unable to comprehensively support the user's dining experience.

[0339] 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.

[0340] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing recipes based on the history, and a means for suggesting restaurants based on the user's preferences. This makes it possible to provide recipes and suggest restaurants that reflect the user's preferences and tastes.

[0341] "User input" means information or instructions provided by a user to a system.

[0342] A "conversational AI that provides recipes" is an artificial intelligence that suggests cooking recipes through dialogue with the user.

[0343] A "means for instructing the cooking steps of a recipe" refers to a function that provides specific guidance to the user on the cooking steps.

[0344] "Means of receiving and learning from user feedback" refers to the ability of the system to collect opinions and evaluations provided by users and make improvements based on them.

[0345] "A means of referencing the user's past conversation history about food and providing recipes based on that" refers to a function that uses records of the user's past conversations to suggest recipes that suit the user's preferences.

[0346] "A means of suggesting restaurants based on user preferences" refers to a function that recommends suitable restaurants based on the user's preferences and past conversation history.

[0347] A "generative AI model" is a model that uses artificial intelligence to generate new information and patterns from data.

[0348] A "prompt sentence" is an instruction sentence entered into an artificial intelligence to elicit a specific response.

[0349] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referencing the user's past conversation history about meals and providing recipes based thereon, and a means for suggesting restaurants based on the user's preferences.

[0350] System Configuration

[0351] The server uses the following hardware and software:

[0352] Hardware: powerful processor, sufficient memory, database server

[0353] Software: conversational AI engine (e.g. OpenAI's GPT-3), database management system (e.g. SQLite), user interface (e.g. smartphone app)

[0354] Data processing and calculation

[0355] The server processes the data in the following steps:

[0356] 1. Receive user input: Receive information entered by the user through the smartphone app.

[0357] 2. Recipe provision by conversational AI: A conversational AI engine generates appropriate recipes based on user input.

[0358] 3. Cooking Instructions: Provide the user with step-by-step instructions for cooking a recipe.

[0359] 4. Collect feedback and learn: Collect feedback from users and let the AI ​​engine learn from that information.

[0360] 5. Referencing past conversation history: Retrieve the user’s past conversation history from the database and provide recipes based on the user’s preferences.

[0361] 6. Restaurant Suggestions: Suggest suitable restaurants based on user preferences.

[0362] Examples

[0363] For example, if a user has previously mentioned that they like spicy food, the system will prioritize spicy recipes based on that information and also suggest restaurants that serve spicy food.

[0364] Example prompt:

[0365] User conversation history: "I like spicy food. I especially like curry and mapo tofu."

[0366] Extract user preferences.

[0367] Using this prompt, a generative AI model extracts the user's preferences and uses the results to provide recipes and restaurant suggestions.

[0368] The flow of the specific process in the application example 3 will be described with reference to FIG.

[0369] Step 1:

[0370] Users input meal information through a smartphone app.

[0371] Input: User's dietary information (e.g. favorite dishes, allergies, etc.)

[0372] Output: The user's input information is sent to the server.

[0373] Specific behavior: A user enters information into an input form in the app and presses the submit button.

[0374] Step 2:

[0375] The server receives the user's input information and sends it to the conversational AI engine.

[0376] Input: User input information

[0377] Output: User information sent to the conversational AI engine

[0378] Specific operation: The server analyzes the user information it receives, converts it into an appropriate format, and sends it to the AI ​​engine.

[0379] Step 3:

[0380] A conversational AI engine generates recipes based on user input.

[0381] Input: User input information

[0382] Output: The generated recipe

[0383] What it does: The AI ​​engine generates an appropriate recipe taking into account the user's preferences and constraints.

[0384] Step 4:

[0385] The server sends the generated recipe to the user.

[0386] Input: The generated recipe

[0387] Output: A recipe displayed on the user's smartphone app

[0388] Specific operation: The server sends the generated recipe to the user's smartphone app and displays it.

[0389] Step 5:

[0390] The user follows the recipe, cooks the dish and then provides feedback in the app.

[0391] Input: User feedback (e.g. recipe rating, suggestions for improvement, etc.)

[0392] Output: The feedback information is sent to the server.

[0393] Specific action: After cooking, the user enters information into the app's feedback form and presses the submit button.

[0394] Step 6:

[0395] The server receives user feedback and sends it to the AI ​​engine for learning.

[0396] Input: User feedback information

[0397] Output: The trained AI model

[0398] Specific operation: The server analyzes the received feedback information and sends it to the AI ​​engine to update the model.

[0399] Step 7:

[0400] The server retrieves the user's past conversation history from the database and sends it to the AI ​​engine.

[0401] Input: User's past conversation history

[0402] Output: Conversation history sent to the AI ​​engine

[0403] Specific operation: The server retrieves the user's past conversation history from the database and sends it to the AI ​​engine.

[0404] Step 8:

[0405] The AI ​​engine extracts the user's preferences based on past conversation history and provides recipes.

[0406] Input: User's past conversation history

[0407] Output: A recipe based on the user's preferences.

[0408] How it works: The AI ​​engine analyzes conversation history, extracts user preferences, and generates appropriate recipes.

[0409] Step 9:

[0410] The server suggests restaurants based on the user's preferences.

[0411] Input: User preferences

[0412] Output: Suggested restaurant information

[0413] Specific operation: The server searches the database for suitable restaurants based on the user's preferences and makes suggestions.

[0414] 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.

[0415] "Example 1"

[0416] One embodiment of the present invention is a system that provides a conversational AI that includes an emotion engine that recognizes the emotions of a user. The system provides a recipe based on the user's input and instructs the user on how to cook the recipe through a conversation. The system also receives feedback from the user and learns from it. Specifically, if the user inputs, "I'm feeling sad today, so I want to eat something that will cheer me up," the emotion engine recognizes the user's "sad" mood and the AI ​​provides a recipe that is said to be uplifting.

[0417] "Example 2"

[0418] In addition, the emotion engine adjusts the recipe offerings based on the user's emotions. For example, if a user feels "very tired today," the emotion engine will recognize that information and offer recipes that are easy to make even for tired users. This makes it possible to offer optimal recipes according to the user's emotional state.

[0419] "Example 3"

[0420] The emotion engine also adjusts recipe cooking instructions based on the user's emotions. For example, if the emotion engine recognizes that the user is in a hurry, the AI ​​will provide concise and easy-to-understand cooking instructions. This makes it possible to provide optimal cooking instructions according to the user's emotional state.

[0421] The process flow of each embodiment will be described below.

[0422] "Example 1"

[0423] Step 1: The user gives input to the conversational AI. For example, the user might say, "I'm feeling sad today, so I want to eat something that will cheer me up."

[0424] Step 2: The emotion engine recognizes emotions from the user's input. In this example, the "sad" mood is recognized.

[0425] Step 3: The AI ​​provides a recipe based on the output of the emotion engine. In this example, it provides a recipe that is said to be uplifting.

[0426] "Example 2"

[0427] Step 1: A user gives input to the conversational AI. For example, the user might say, "I'm very tired today."

[0428] Step 2: The emotion engine recognizes the emotion from the user's input. In this example, the state "tired" is recognized.

[0429] Step 3: The AI ​​provides a recipe based on the output of the emotion engine. In this example, it provides a recipe that is easy to make even for tired users.

[0430] "Example 3"

[0431] Step 1: The user gives input to the conversational AI. For example, the user might say, "I'm anxious."

[0432] Step 2: The emotion engine recognizes emotions from the user's input. In this example, the state of "anxious" is recognized.

[0433] Step 3: The AI ​​gives instructions for cooking based on the output of the emotion engine. In this example, the instructions are given succinctly and clearly.

[0434] Example 1

[0435] 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".

[0436] Conventional conversational AI systems were able to provide recipes based on user input, but they lacked the ability to suggest recipes that took the user's emotions into account, or to learn from the user's feedback and reflect it in future suggestions. This made it difficult to provide optimal recipes based on the user's mood and preferences. In addition, they did not provide sufficient real-time support to users during the cooking process.

[0437] 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.

[0438] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking procedure of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion recognition means for recognizing the emotion of the user, and a means for providing a recipe based on the emotion of the user. This makes it possible to provide an optimal recipe according to the user's mood and preferences.

[0439] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through a dialogue with a user and generates appropriate responses based on the user's input.

[0440] The "means for instructing cooking steps" is a function that explains the specific cooking steps of a recipe to the user and sequentially instructs the user on the necessary steps.

[0441] "Means of learning feedback" refers to a function that collects evaluations and opinions from users and uses them to improve the system's performance and the content of its suggestions.

[0442] An "emotion recognition means" is a technology for analyzing emotions from a user's input or dialogue and recognizing those emotions.

[0443] A "means for providing recipes" is a function that generates appropriate cooking recipes based on the user's requests and feelings and provides them to the user.

[0444] This invention is a system that uses a smartphone application equipped with conversational artificial intelligence to provide optimal recipes based on the user's meal-related desires and emotions. The system analyzes the user's input, recognizes emotions, and generates recipes using a generative AI model. It also learns feedback from the user and reflects it in future suggestions.

[0445] Hardware and Software Used

[0446] Hardware

[0447] Device: Mobile devices such as smartphones and tablets

[0448] Server: Cloud or on-premise server

[0449] software

[0450] Natural Language Processing (NLP) Technology: Google(R) Cloud Natural Language API, IBM Watson(R) Natural Language Understanding

[0451] Emotion recognition engine: Microsoft(R) Azure(R) Emotion API

[0452] Generative AI model: OpenAI's GPT-4

[0453] Machine learning algorithms: TENSORFLOW(registered trademark), PyTorch

[0454] Program Processing

[0455] User Input

[0456] The user starts the smartphone application and inputs their meal-related request into a text box, for example, "I'd like to eat Japanese food today" or "Tell me about some dishes that use chicken."

[0457] Analyzing data

[0458] The terminal sends the user's input to the server, which then analyzes the received text data using natural language processing technology. This analysis allows the server to understand the user's request.

[0459] Emotion recognition

[0460] Based on the analysis results, the server uses an emotion recognition engine to recognize the user's emotions. For example, if the user enters "I'm feeling sad today, so I want to eat something that cheers me up," the server recognizes the emotion "sad."

[0461] Recipe Suggestions

[0462] The server uses a generative AI model to generate recipes that match the user's preferences based on the analysis results and emotion recognition results. The generated recipes are sent to the device in the form of text, images, and videos.

[0463] User interaction

[0464] If a user has a question or difficulty while cooking, they can ask the AI ​​through the application. The device sends the question to the server, which then generates a response in real time and returns it to the device.

[0465] Learning Feedback

[0466] The user provides feedback on the provided recipe, and the device sends this feedback to the server, which then learns from the feedback using machine learning algorithms.

[0467] Examples

[0468] If a user types, "I want to eat Japanese food today," the device will send this request to the server. The server will analyze the request and suggest Japanese recipes using a generative AI model. For example, a recipe for "teriyaki chicken" will be provided in text, images, and video. If a user types, "I'm feeling sad today, so I want to eat something that will cheer me up," the server will recognize the emotion of "sadness" and suggest a recipe for "curry rice," which is said to be uplifting.

[0469] Examples of prompt statements

[0470] "A user says they want to eat Japanese food. Please suggest some Japanese recipes."

[0471] "A user says they're feeling sad. Can you suggest some cheer-up recipes?"

[0472] In this way, the system provides the optimal recipe based on the user's desires and feelings.

[0473] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0474] Step 1:

[0475] The user starts the smartphone application and inputs their meal-related requests into a text box. For example, they may make requests such as, "I'd like to eat Japanese food today," or "Tell me about some dishes that use chicken." The input data is stored in text format on the device.

[0476] Step 2:

[0477] The device sends the user's input data to the server. The server analyzes the received text data using natural language processing technology (Google Cloud Natural Language API and IBM Watson Natural Language Understanding). The user's request is extracted as a result of the analysis.

[0478] Step 3:

[0479] Based on the analysis results, the server uses an emotion recognition engine (Microsoft Azure Emotion API) to recognize the user's emotions. For example, if the user enters "I'm feeling sad today, so I want to eat something that cheers me up," the emotion "sad" is recognized. The emotion recognition results are stored on the server.

[0480] Step 4:

[0481] The server uses a generative AI model (OpenAI's GPT-4) to generate recipes that match the user's preferences based on the analysis results and emotion recognition results. The generated recipes are stored on the server in text, image, and video formats.

[0482] Step 5:

[0483] The server sends the generated recipe to the terminal. The terminal displays the received recipe to the user. The user confirms the recipe and starts cooking.

[0484] Step 6:

[0485] If a user has a question or difficulty while cooking, they can ask the AI ​​through the application. The device sends the question to the server, which generates a response in real time and returns it to the device. The user receives the response and continues cooking.

[0486] Step 7:

[0487] The user provides feedback on the provided recipe. The device sends this feedback to the server, which then uses machine learning algorithms (TensorFlow and PyTorch) to learn from the feedback. The learning results are reflected in future suggestions.

[0488] (Application example 1)

[0489] 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."

[0490] Conventional conversational AI systems could not suggest recipes based on the user's emotions or mood of the day, and could not sufficiently increase user satisfaction. In addition, there was no function to order delivery of the suggested recipes as they were, which reduced user convenience.

[0491] 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. In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine means for recognizing the user's emotions, a means for providing a recipe according to the user's emotions, and a means for ordering the suggested dish for delivery. This makes it possible to suggest recipes according to the user's emotions and mood of the day, and further allows the suggested dish to be ordered for delivery as is, thereby improving user satisfaction and convenience.

[0492] "User input" is information, such as text or voice, that a user provides to the system.

[0493] "Conversational AI" is artificial intelligence that provides information and gives instructions through natural conversation with the user.

[0494] A "cooking procedure" is a specific step or method for preparing a dish.

[0495] "Feedback" refers to reactions such as evaluations and opinions that users provide to a system.

[0496] An "emotion engine" is a technology that recognizes a user's emotions and responds appropriately based on those emotions.

[0497] A "recipe" is a detailed description of the ingredients and steps for making a particular dish.

[0498] A "delivery order" is an ordering procedure for delivering the food selected by the user to a specified location.

[0499] "Preferences" refer to personal preferences such as the ingredients and types of cuisine that the user particularly likes.

[0500] A "conversation history" is a record of past conversations between a user and the system.

[0501] A "server" is a computer that manages the entire system and processes and stores various data.

[0502] A system for implementing the present invention includes a conversational AI that provides a recipe based on user input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine that recognizes the user's emotions, a means for providing a recipe according to the user's emotions, and a means for ordering delivery of the suggested dishes.

[0503] System Configuration

[0504] Hardware

[0505] Server: Manages the entire system and processes and stores various data. Specifically, a cloud server or an on-premise server is used.

[0506] User device: A device that is directly operated by a user, such as a smartphone or tablet.

[0507] software

[0508] Conversational AI: Provides information and instructions through natural conversation with the user, using generative AI models such as OpenAI's GPT-3.

[0509] Sentiment engine: Recognizes the user's emotions and takes appropriate action based on those emotions, using sentiment analysis libraries such as TextBlob.

[0510] Delivery ordering system: A system for ordering delivery of the suggested dishes. Specifically, it will be linked to existing delivery service APIs.

[0511] Data processing and calculation

[0512] 1. Processing user input:

[0513] Users input requests such as "I want to eat Japanese food today" or "Please tell me some recipes that use chicken" via their smartphone.

[0514] This input is sent to the server in text format.

[0515] 2. Emotion recognition:

[0516] The server uses a sentiment analysis library such as TextBlob to analyze the sentiment of the user's input text.

[0517] For example, if a user types, "I'm feeling sad today, so I want to eat something that will cheer me up," the emotion engine will recognize this as "sad."

[0518] 3. Recipe suggestions:

[0519] The server uses a generative AI model (e.g. GPT-3) to generate recipes based on the user's emotions and preferences.

[0520] An example of a prompt sentence is "The user says that he is feeling sad and wants to eat something that will cheer him up. Please suggest some suitable dishes."

[0521] 4. Delivery Order:

[0522] The server works with existing delivery service APIs to order delivery of the suggested recipes as is.

[0523] Order food from the nearest restaurant based on the user's location.

[0524] Examples

[0525] If a user types in "I'm tired today, so I want to eat something that will relax me" through their smartphone, the emotion engine will recognize this as "tired" and use the generative AI model to suggest relaxing dishes. The suggested dishes will then be ordered for delivery from the nearest restaurant based on the user's location.

[0526] An example of a prompt sentence is, "The user says that he feels tired and wants to eat something that will relax him. Please suggest some suitable dishes."

[0527] In this way, it is possible to suggest recipes based on the user's emotions and mood of the day, and the suggested dishes can be ordered for delivery, thereby increasing user satisfaction and convenience.

[0528] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0529] Step 1:

[0530] The user inputs requests such as "I want to eat Japanese food today" or "Please tell me about some dishes that use chicken" via smartphone. The input is in text format and is sent from the smartphone to the server. The input data is text information that includes the user's request.

[0531] Step 2:

[0532] The server analyzes the received user input text. Specifically, it uses a text analysis engine to extract the user's request. The input data is the user's request text, and the output data is the analyzed request information.

[0533] Step 3:

[0534] The server analyzes the sentiment of the user's input text using a sentiment analysis library such as TextBlob. The input data is the user's request text, and the output data is the sentiment information (e.g., "sad," "tired," etc.).

[0535] Step 4:

[0536] The server uses a generative AI model (e.g., GPT-3) to generate recipes based on the user's emotions and preferences. The prompt sentence is input to the generative AI model, which outputs an appropriate recipe. The input data are emotion information and user request information, and the output data is the suggested recipe.

[0537] Step 5:

[0538] The server provides the suggested recipe to the user. Specifically, it transmits recipe text information, images, videos, etc. to the smartphone terminal. The input data is the suggested recipe information, and the output data is the recipe information displayed on the user terminal.

[0539] Step 6:

[0540] When a user checks the suggested recipes and places a delivery order, the user sends a delivery order request from the smartphone terminal to the server. The input data is the delivery order request, and the output data is the order information sent to the server.

[0541] Step 7:

[0542] The server works with existing delivery service APIs to order food from the nearest restaurant based on the user's location information. The input data is the user's location information and order information, and the output data is an order request to the delivery service.

[0543] Step 8:

[0544] A delivery service takes orders and delivers food to a specified location. The input data is the order request to the delivery service, and the output data is the delivery of the food to the user.

[0545] In this way, it is possible to suggest recipes based on the user's emotions and mood of the day, and the suggested dishes can be ordered for delivery, thereby increasing user satisfaction and convenience.

[0546] Example 2

[0547] 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."

[0548] Conventional recipe provision systems had difficulty providing recipes that corresponded to the individual preferences and emotional state of the user. In addition, they lacked the ability to effectively learn from user feedback and reflect it in future recipes. This made it difficult to increase user satisfaction.

[0549] 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.

[0550] In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine means for inputting the emotional state of the user and adjusting the recipe based thereon, and a means for inputting the feedback and the emotional state to a generative AI model and generating a recipe. This makes it possible to provide an optimal recipe according to the user's preferences and emotional state.

[0551] A "conversational artificial intelligence means" is an artificial intelligence having a function of providing a recipe based on a user's input and instructing the user on the cooking steps of the recipe through conversation with the user.

[0552] A "means for learning feedback" is a means that has the function of receiving feedback from the user, analyzing it, and reflecting it in subsequent recipe generation.

[0553] The "emotion engine means" is a means having a function of inputting the user's emotional state and adjusting the recipe based on that.

[0554] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to take user feedback and emotional state as input and generate recipes based on that.

[0555] A "means for generating recipes" is a means having a function for generating optimal recipes according to the user's preferences and emotional state.

[0556] The invention is a system including a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine means for inputting the user's emotional state and adjusting the recipe based thereon, and a means for inputting the feedback and the emotional state into a generative artificial intelligence model and generating a recipe.

[0557] System Configuration

[0558] This system is realized using the following hardware and software.

[0559] Hardware: The device on which the user enters information (smartphone, tablet, PC, etc.), and the server that processes the data

[0560] Software: Conversational AI, Emotion Engine, Generative AI Models

[0561] Program Processing

[0562] Learning Feedback

[0563] 1. After cooking, the user enters feedback into the device, such as "I would like it to be less spicy."

[0564] 2. The device receives the user feedback and transmits it to a server via the Internet.

[0565] 3. The server analyzes the received feedback and processes it as text data. For example, it converts the feedback into a specific instruction such as "Make it less spicy."

[0566] 4. The server inputs the analyzed feedback into the generative AI model. Specifically, it sends a prompt to the generative AI model, such as "Make the recipe less spicy next time."

[0567] 5. The generative AI model receives the prompt and learns from the feedback, so that it can reflect the user's requests in future recipe generation.

[0568] Recipe adjustment using emotion engine

[0569] 1. The user inputs his / her emotional state into the device, such as "I feel very tired today."

[0570] 2. The device receives the user's emotional state and transmits it to a server via the Internet.

[0571] 3. The server passes the received emotional state to the emotion engine. Specifically, it sends information that "the user is tired" to the emotion engine.

[0572] 4. The emotion engine analyzes the user’s emotional state and requests “easy recipes for tired users” from the generative AI model.

[0573] 5. The emotion engine sends a prompt to the generative AI model saying, "Please provide a recipe that is easy for tired users to make."

[0574] 6. A generative AI model receives the prompt and generates recipes according to the user’s emotional state, for example providing recipes that require less cooking time and effort.

[0575] Examples

[0576] Feedback learning prompt:

[0577] User feedback: "I wish it was less spicy."

[0578] Learn how to tone down the spiciness to incorporate into your next recipe.

[0579] Emotion engine prompt:

[0580] User's emotional state: "I'm very tired today"

[0581] Provide recipes that are easy to make even for tired users.

[0582] This system makes it possible to provide optimal recipes based on the user's preferences and emotional state.

[0583] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0584] Learning Feedback

[0585] Step 1:

[0586] Users enter feedback after cooking.

[0587] Input: The user types into the terminal, "I would like it less spicy."

[0588] Output: Feedback data is saved to the device.

[0589] Specific action: The user enters feedback in text format using a device such as a smartphone or computer.

[0590] Step 2:

[0591] The device sends the feedback to the server.

[0592] Input: Feedback data stored on the device.

[0593] Output: Feedback data is sent to the server.

[0594] Specific operation: The terminal transmits feedback data to the server via the Internet.

[0595] Step 3:

[0596] The server parses the feedback.

[0597] Input: The feedback data sent to the server.

[0598] Output: Parsed feedback information (e.g. "Make it less spicy").

[0599] Specific Action: The server performs text analysis on the feedback data and converts it into specific instructions.

[0600] Step 4:

[0601] The server inputs feedback into the generative AI model.

[0602] Input: The parsed feedback information.

[0603] Output: The prompt sent to the generative AI model.

[0604] Specific operation: The server sends a prompt to the generative AI model, such as "Make the recipe less spicy next time."

[0605] Step 5:

[0606] Generative AI models learn from feedback.

[0607] Input: The prompt sent to the generative AI model.

[0608] Output: The learned feedback information.

[0609] Specific operation: The generative AI model receives the prompt text and reflects the user's requests the next time it generates a recipe.

[0610] Recipe adjustment using emotion engine

[0611] Step 1:

[0612] The user inputs an emotional state.

[0613] Input: A user types into a terminal, "I'm very tired today."

[0614] Output: Emotional state data is saved on the device.

[0615] Specific operation: The user inputs their emotional state in text format using a device such as a smartphone or computer.

[0616] Step 2:

[0617] The device transmits the emotional state to the server.

[0618] Input: Emotional state data stored on the device.

[0619] Output: Emotional state data is sent to the server.

[0620] Specific operation: The terminal transmits emotional state data to a server via the Internet.

[0621] Step 3:

[0622] The server passes the information to the emotion engine.

[0623] Input: Emotional state data sent to the server.

[0624] Output: Emotional state information sent to the emotion engine.

[0625] Specific operation: The server sends information that "the user is tired" to the emotion engine.

[0626] Step 4:

[0627] The emotion engine analyzes the user's emotional state.

[0628] Input: Emotional state information sent to the emotion engine.

[0629] Output: Parsed emotional state information (e.g. "Quick recipes for tired users").

[0630] Specific operation: The emotion engine analyzes the emotional state information and requests an appropriate recipe from the generative AI model.

[0631] Step 5:

[0632] The emotion engine requests the appropriate recipe from the generative AI model.

[0633] Input: Parsed emotional state information.

[0634] Output: The prompt sent to the generative AI model.

[0635] Specific behavior: The emotion engine sends a prompt to the generative AI model saying, "Please provide a recipe that is easy for tired users to make."

[0636] Step 6:

[0637] A generative AI model provides recipes based on the user's emotional state.

[0638] Input: The prompt sent to the generative AI model.

[0639] Output: A recipe depending on the user's emotional state.

[0640] How it works: A generative AI model receives a prompt and generates recipes that are quick and hassle-free to prepare.

[0641] (Application example 2)

[0642] 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."

[0643] Conventional recipe provision systems have difficulty in proposing recipes according to the individual preferences and emotional state of the user, and therefore have been unable to increase user satisfaction. In addition, there is a lack of a means to effectively utilize feedback from users, making it difficult to reflect this in the next recipe proposal.

[0644] 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.

[0645] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for analyzing the emotional state of the user and adjusting the recipe based on it, and a means for suggesting the next recipe based on the user's feedback and emotional state. This makes it possible to suggest optimal recipes according to the individual preferences and emotional state of the user, thereby increasing user satisfaction.

[0646] "User input" means information or instructions provided by a user to a system.

[0647] "Conversational AI" refers to artificial intelligence that uses natural language processing technology to interact with users, provide information, and give instructions.

[0648] A "means for instructing cooking steps" refers to a function that guides the user through cooking steps and methods.

[0649] "Means of receiving and learning from feedback" refers to the ability to collect opinions and feedback from users and use that information to make improvements and adjustments to the system.

[0650] "Means for analyzing the user's emotional state and adjusting the recipe based on that" refers to a function that analyzes the user's emotions and changes the recipe content or suggestions based on the results.

[0651] "Means to suggest next recipe" refers to a feature that takes into account the user's past feedback and emotional state to suggest the next dish they want to make.

[0652] A system for implementing the present invention includes a conversational AI that provides a recipe based on user input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for analyzing the user's emotional state and adjusting the recipe based thereon, and a means for suggesting the next recipe based on the user's feedback and emotional state.

[0653] Program processing explanation

[0654] The server executes a program using Python. Specifically, it performs the following processes:

[0655] 1. Receive user input:

[0656] Users input information into the system using their smartphones, including cooking requests, emotional states, and feedback.

[0657] 2. Conversational AI dialogue:

[0658] Conversational AI uses natural language processing techniques to converse with users to suggest recipes and provide cooking instructions, while using generative AI models to analyze user input and generate appropriate responses.

[0659] 3. Gather feedback and learn:

[0660] The feedback provided by users after cooking is collected and stored in a database, which is then learned from to improve recipe suggestions for the next time.

[0661] 4. Emotional state analysis:

[0662] It uses an emotion engine to analyze the user's emotional state. If a user types, "I'm very tired today," the emotion engine analyzes that information and suggests easy recipes.

[0663] 5. Recipe suggestions for next time:

[0664] It takes into account the user's past feedback and emotional state to suggest the next recipe, providing the best recipe based on the user's preferences and emotional state.

[0665] Hardware and Software Used

[0666] Hardware: Smartphone

[0667] Software: Python, natural language processing technology, emotion engine, generative AI model

[0668] Examples

[0669] When a user uses their smartphone to input "I'm very tired today," the emotion engine analyzes that information and suggests easy-to-make recipes. For example, "Stir-fried chicken and vegetables" is suggested. If the user inputs feedback after cooking, such as "I wish it was less spicy," a less spicy recipe will be suggested next time.

[0670] Examples of prompt statements

[0671] User: I'm really tired today.

[0672] AI: Well, what about a simple recipe? For example, how about "chicken and vegetable stir fry"?

[0673] User: I'd like it less spicy.

[0674] AI: I'll suggest a less spicy recipe next time. Thank you for your feedback.

[0675] In this way, it is possible to suggest optimal recipes based on the user's individual preferences and emotional state, thereby increasing user satisfaction.

[0676] The flow of the identification process in the application example 2 will be described with reference to FIG.

[0677] Step 1:

[0678] A user inputs information into the system using a smartphone. The input information can be a food request, an emotional state, feedback, etc. For example, a user might input, "I'm very tired today." This input data is sent to the server.

[0679] Step 2:

[0680] The server analyzes the received user input data. Specifically, it uses natural language processing technology to understand the user's input and generate an appropriate response. At this time, a generative AI model is used to analyze the user's input and identify the user's emotional state. For example, the emotion "tired" is analyzed.

[0681] Step 3:

[0682] The server proposes recipes suitable for the user based on the analysis results. Using an emotion engine, it selects recipes according to the user's emotional state. For example, it suggests "chicken and vegetable stir fry" as an "easy recipe to make." The proposal is displayed on the smartphone.

[0683] Step 4:

[0684] The user checks the suggested recipe and cooks the food. After cooking, the user inputs feedback via their smartphone. For example, they can input feedback such as "I would like it less spicy." This feedback data is sent to the server.

[0685] Step 5:

[0686] The server stores the received feedback in a database. The feedback data is then learned so that it can be reflected in the next recipe suggestions. Specifically, a machine learning algorithm is used to analyze the feedback data and learn the user's preferences.

[0687] Step 6:

[0688] The next time the user enters input into the system again, the server will suggest recipes taking into account the user's past feedback and emotional state. For example, it may suggest a "less spicy recipe" based on the user's previous feedback. The suggestions are displayed on the smartphone.

[0689] In this way, it is possible to suggest optimal recipes according to the individual preferences and emotional state of the user, thereby increasing user satisfaction.

[0690] Example 3

[0691] 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."

[0692] Conventional conversational AI systems do not adequately provide recipes that correspond to the user's preferences and emotional state. In addition, there is a lack of means to effectively utilize the user's past conversation history, making it difficult to provide personalized recipes based on the user's preferences. Furthermore, cooking procedures are not adjusted according to the user's emotional state, which makes it difficult to increase user satisfaction.

[0693] 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.

[0694] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing a recipe based thereon, and a means for analyzing the user's emotional state and adjusting the cooking steps based thereon. This makes it possible to provide personalized recipes according to the user's preferences and emotional state, thereby increasing user satisfaction.

[0695] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through a dialogue with a user and generates appropriate responses based on the user's input.

[0696] "User input" is data, such as text or voice, that a user provides to the system.

[0697] A "recipe" is information that lists the ingredients and steps for making a particular dish.

[0698] A "cooking procedure" is the specific steps or methods to follow when preparing a dish.

[0699] "Feedback" refers to reactions such as evaluations and opinions that users provide to a system.

[0700] "Means of learning" are ways that the system can improve itself based on user feedback and past data.

[0701] "Past conversation history" is a record of previous interactions between a user and the system.

[0702] "Emotional state" refers to a user's current psychological state or mood.

[0703] A "means for analyzing" is a method for analyzing the content of data input by a user and extracting specific information.

[0704] A "generative artificial intelligence model" is a machine learning model used to generate appropriate responses or information based on user input.

[0705] The present invention relates to a conversational artificial intelligence system that provides recipes based on user input. A specific embodiment of this system will be described below.

[0706] System Configuration

[0707] The server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, a means for referring to the user's past conversation history regarding meals and providing a recipe based thereon, and a means for analyzing the user's emotional state and adjusting the cooking steps based thereon.

[0708] Hardware and Software Used

[0709] The server uses the following hardware and software:

[0710] Hardware: A high-performance server machine (e.g. a cloud server equipped with CPU, memory, and storage)

[0711] software:

[0712] Database management system (e.g. MySQL (registered trademark), PostgreSQL)

[0713] Generative AI models (e.g. GPT-4)

[0714] Sentiment analysis engine (e.g. Emotion API)

[0715] Data processing and calculation

[0716] The server processes and calculates data in the following procedure.

[0717] 1. Collecting user conversation history:

[0718] The server retrieves the user's past conversation history from a database, for example by running a SQL query using MySQL to search for the conversation history using the user ID as a key.

[0719] 2. Learning user preferences:

[0720] The server uses the acquired conversation history to train a generative AI model (GPT-4) to learn the user's preferences and tastes. For example, if a user mentions that they like spicy food, the model memorizes that information.

[0721] 3. Generate the recipe:

[0722] When a user requests a recipe, the server uses the learned information to generate a recipe that matches the user's preferences using a generative AI model. For example, if the user likes spicy food, the server will suggest spicy food recipes.

[0723] 4. Analysis of the user's emotional state:

[0724] The server analyzes the user's current emotional state using an emotion engine (Emotion API). If the server detects that the user is in a hurry, it provides simple and easy-to-understand instructions on how to cook the food.

[0725] 5. Adjust cooking procedure:

[0726] The server adjusts the cooking instructions based on the emotion engine's analysis results. For example, if the user is in a hurry, the instructions will be simplified.

[0727] 6. Recipe provided by:

[0728] The server transmits the adjusted recipe to the user's terminal, and the user receives the recipe.

[0729] Examples

[0730] If the user has previously mentioned that they like spicy food, the following prompt sentence can be input into the generative AI model:

[0731] Example prompt:

[0732] You know from the user's past conversation history that they like spicy food, so provide them with a simple and easy-to-follow recipe for spicy food since they are in a rush.

[0733] An example of a generated recipe:

[0734] Spicy Chicken Curry Recipe:

[0735] 1. Cut the chicken into bite-sized pieces.

[0736] 2. Heat oil in a frying pan and fry the chicken.

[0737] 3. Add the onion, garlic and ginger and continue to stir fry.

[0738] 4. Add curry powder, tomato paste, and chili powder.

[0739] 5. Add water and simmer.

[0740] 6. Season with salt.

[0741] In this way, the server can provide the optimal recipe according to the user's preferences and emotional state. The flow of the specification process in the third embodiment will be described with reference to FIG.

[0742] Step 1:

[0743] The server retrieves the user's conversation history from the database. It receives the user ID as input and executes a SQL query to search the conversation history. As output, it obtains the user's past conversation history data. Specifically, the server executes the following SQL query:

[0744] sql

[0745] SELECT FROM conversation_history WHERE user_id = '12345';

[0746] Step 2:

[0747] The server uses the acquired conversation history to train a generative AI model (GPT-4) and learn the user's preferences and tastes. It receives the conversation history data as input and inputs it into the generative AI model. As output, it obtains a model that reflects the user's preferences. In concrete terms, the server inputs the conversation history as text data into the generative AI model.

[0748] python

[0749] conversation_history = "User: I like spicy food."

[0750] model.train(conversation_history)

[0751] Step 3:

[0752] When a user requests a recipe, the server uses the learned information to generate a recipe that matches the user's preferences using a generative AI model. It receives the user's request and the learned model as input, and inputs a prompt sentence into the generative AI model. As output, it obtains a recipe that matches the user's preferences. In concrete terms, the server inputs the following prompt sentence into the generative AI model.

[0753] python

[0754] prompt = "Can you give me some spicy food recipes?"

[0755] recipe = model.generate(prompt)

[0756] Step 4:

[0757] The server analyzes the user's current emotional state using an emotion engine (Emotion API). It receives the user's text and voice data as input and inputs it into the emotion engine. It obtains the user's emotional state as output. In concrete terms, the server inputs the user's input data into the emotion engine.

[0758] python

[0759] emotion = emotion_api.analyze(user_input)

[0760] if emotion == "anxious":

[0761] instructions = simplify_instructions(recipe)

[0762] Step 5:

[0763] The server adjusts the cooking instructions based on the analysis results of the emotion engine. It receives the emotional state and the generated recipe as input and applies an algorithm to simplify the cooking instructions. It obtains the adjusted recipe as output. Specifically, the server simplifies the instructions as follows:

[0764] python

[0765] def simplify_instructions(recipe):

[0766] Process to simplify the procedure

[0767] simplified_recipe = ...

[0768] return simplified_recipe

[0769] Step 6:

[0770] The server sends the adjusted recipe to the user's terminal. The server receives the adjusted recipe as input and sends it to the user's terminal. The user receives the recipe as output. In concrete terms, the server sends the recipe as follows:

[0771] python

[0772] send_to_user(simplified_recipe)

[0773] (Application example 3)

[0774] 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."

[0775] Conventional food delivery systems have the problem that it is difficult to suggest optimal dishes according to the individual preferences and emotional state of the user. In addition, since a uniform service is provided without considering the user's past conversation history or emotional state, user satisfaction may decrease. Furthermore, when the user is in a hurry, the cooking procedure is not adjusted according to the emotional state, which may increase the user's stress.

[0776] 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.

[0777] In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing a recipe based thereon, a means for analyzing the user's emotional state and adjusting the cooking steps based thereon, and a means for suggesting optimal dishes based on the user's preferences and emotional state and supporting ordering. This makes it possible to suggest and order optimal dishes according to the user's individual preferences and emotional state.

[0778] "Conversational AI" is an artificial intelligence system that provides information through natural language dialogue with users and services according to their requests.

[0779] A "recipe" is a set of instructions that lists the ingredients and steps for making a particular dish.

[0780] "Cooking instructions" are instructions that show the specific operations and steps required to prepare a dish.

[0781] "Feedback" refers to information provided by users, such as opinions, ratings, and impressions after use.

[0782] "Past conversation history" refers to a record of past interactions between a user and the system.

[0783] "Emotional state" refers to a user's current psychological state or mood.

[0784] "Preferences" refers to the foods and flavors that a user particularly likes.

[0785] "Dish suggestions" refers to the act of selecting and presenting appropriate dishes based on the user's preferences and emotional state.

[0786] "Means for supporting ordering" refers to a function that assists the user in the process of actually ordering the selected dish.

[0787] A system for implementing this invention includes a conversational AI that provides a recipe based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing a recipe based thereon, a means for analyzing the user's emotional state and adjusting the cooking steps based thereon, and a means for suggesting optimal dishes based on the user's preferences and emotional state and supporting ordering.

[0788] System Configuration

[0789] This system is realized using the following hardware and software.

[0790] Hardware: Smartphone

[0791] Software: Python, OpenAI API, sentiment analysis library (e.g. sentiment_analysis), food delivery API (e.g. food_delivery_api)

[0792] Data processing and calculation

[0793] Get a user's conversation history

[0794] The server retrieves the user's past conversation history from the database, including what dishes the user has liked in the past and what feedback he or she has provided.

[0795] Learn your preferences

[0796] The server analyzes the conversation history and learns the user's preferences. For example, if the user mentions that they like spicy food, the server remembers that information and prioritizes spicy food suggestions.

[0797] Analyze your emotional state

[0798] The server analyzes the user's emotional state from the current conversation, using a sentiment analysis library to determine whether the user is "anxious" or "relaxed."

[0799] Propose the best dish

[0800] The server suggests the best recipes based on the user's preferences and emotional state. For example, if the user is in a hurry, it suggests recipes with simple cooking steps.

[0801] Order food

[0802] The server then orders the suggested dishes through a food delivery API, allowing users to receive their food without any hassle.

[0803] Examples

[0804] If a user says, "I'm busy and in a hurry today," the system will learn from the user's past conversation history that they like spicy food, suggest a simple spicy dish, and then order that dish through a food delivery API.

[0805] Examples of prompt statements

[0806] User ID: 12345

[0807] Current conversation: I'm busy and anxious today.

[0808] By inputting this prompt into a generative AI model, the system can suggest and order the most appropriate dish based on the user's emotional state.

[0809] The flow of the specific process in the application example 3 will be described with reference to FIG.

[0810] Step 1:

[0811] The server receives user input, which includes the user ID and the current conversation. Based on this input, the server retrieves the user's past conversation history from the database. The output is the user's past conversation history.

[0812] Step 2:

[0813] The server analyzes the acquired conversation history and learns the user's preferences. Specifically, it extracts information about preferences, such as "I like spicy food," from the conversation history and stores it in a user preference database. The input is the user's past conversation history, and the output is the user's preference data.

[0814] Step 3:

[0815] The server analyzes the current conversation content using a sentiment analysis library (e.g., sentiment_analysis) to determine the user's emotional state. For example, from the input "I'm busy today and I'm anxious," the emotional state "anxious" is extracted. The input is the current conversation content, and the output is the user's emotional state.

[0816] Step 4:

[0817] The server proposes the best dish based on the user's preference data and emotional state. For example, if the user "likes spicy food" and is "in a hurry," the server proposes a simple spicy dish. The input is the user's preference data and emotional state, and the output is the proposed dish.

[0818] Step 5:

[0819] The server orders the suggested dishes through a food delivery API (e.g. food_delivery_api) by sending the suggested dishes to the API and completing the order process. The input is the suggested dishes and the output is the order confirmation.

[0820] Step 6:

[0821] The server receives feedback from users and learns from it. Specifically, it analyzes the feedback provided by users and incorporates it into the next suggestions. The input is the user feedback and the output is the updated preference data.

[0822] 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.

[0823] 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.

[0824] Another example of generative AI is Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0825] 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.

[0826] [Second embodiment]

[0827] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0828] 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.

[0829] 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).

[0830] 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.

[0831] 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.

[0832] 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).

[0833] 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.

[0834] 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.

[0835] 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.

[0836] 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.

[0837] 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.

[0838] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0839] "Example 1"

[0840] One embodiment of the present invention is a smartphone application equipped with a conversational AI. The user converses with the AI ​​through the application and provides information about the day's meals. For example, the user can tell the AI ​​their preferences, such as "I want to eat Japanese food today" or "Please tell me about some dishes that use chicken." Based on this information, the AI ​​provides recipes that match the user's preferences. The recipes are not just text, but also use images and videos to explain the cooking steps in detail. Furthermore, if the user has any questions or problems during the cooking process, they can directly contact the AI.

[0841] "Example 2"

[0842] Furthermore, another embodiment of the present invention is a function that allows the AI ​​to learn from feedback from users. After cooking, the user can tell the AI ​​their thoughts and suggestions for improvement, and the AI ​​will learn from this information. For example, if the AI ​​receives feedback such as "I would like it less spicy" or "I would like to use more vegetables," the AI ​​will provide recipes that reflect these requests from the next time. This makes it possible to provide recipes that are tailored to the user's preferences.

[0843] "Example 3"

[0844] In addition, as a further embodiment of the present invention, there is a function where the AI ​​refers to the user's past conversation history regarding food. The AI ​​learns the user's tastes and preferences from past conversations with the user. For example, if the user has previously mentioned that they "like spicy food," the AI ​​can remember that information and provide recipes for spicy dishes. This makes it possible to provide recipes that more closely correspond to the user's tastes.

[0845] The process flow of each embodiment will be described below.

[0846] "Example 1"

[0847] Step 1: The user opens the smartphone application and starts a conversation with the conversational AI.

[0848] Step 2: The user provides the AI ​​with information about the day's meals. For example, the user may tell the AI ​​a request such as, "I want to eat Japanese food today," or "Please tell me about dishes that use chicken."

[0849] Step 3: Based on the information provided by the user, the AI ​​will select an appropriate recipe and provide the recipe and cooking instructions to the user.

[0850] Step 4: If the user has any questions or problems during the cooking process, they can directly contact the AI, which will provide an appropriate answer to the question.

[0851] "Example 2"

[0852] Step 1: After cooking, the user tells the AI ​​what they think and what improvements they would like to make.

[0853] Step 2: The AI ​​learns from user feedback and incorporates that information into its next recipe selection.

[0854] "Example 3"

[0855] Step 1: The AI ​​learns the user's preferences and tastes from past conversations with them.

[0856] Step 2: When a user requests a new recipe, the AI ​​will refer to past conversation history to provide recipes that match the user's preferences.

[0857] Example 1

[0858] 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".

[0859] Conventional recipe provision systems have difficulty in providing recipes that meet the specific needs of users, and are unable to respond quickly and appropriately to follow-up questions during the cooking process. Furthermore, they do not adequately consider the user's preferences or past conversation history when providing recipes. This leads to a decrease in user satisfaction and a decrease in the frequency of system use.

[0860] 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.

[0861] In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for generating a recipe that meets the user's needs using a generative AI model, a means for providing the generated recipe to the user in text, image, or video format, and a means for generating an answer using the generative AI model again when the user asks an additional question during the cooking process. This makes it possible to provide a recipe according to the user's specific needs, and to quickly and appropriately respond to additional questions during the cooking process. It also makes it possible to provide a recipe that takes into account the user's preferences and past conversation history, thereby improving user satisfaction.

[0862] "User" means an individual or organization that uses the system to receive recipes.

[0863] "Input" refers to the provision of information, such as text or voice, by a user to a system.

[0864] A "recipe" is information that lists the ingredients and steps for making a particular dish.

[0865] "Conversational AI" is an artificial intelligence system that provides information to users through natural language dialogue.

[0866] "Feedback" refers to reactions such as evaluations and opinions that users provide to a system.

[0867] A "generative AI model" is an artificial intelligence model that generates new information based on user requests.

[0868] A "prompt" is an input sentence that prompts a generative AI model to generate specific information.

[0869] "Text" is data that includes character information.

[0870] An "image" is data that contains visual information.

[0871] "Video" is data that contains dynamic visual information.

[0872] A "cooking procedure" is the sequence of specific steps to prepare a dish.

[0873] A "server" is a computer that is the central part of a system and is a device that processes and stores data.

[0874] A "terminal" is a device that is directly operated by a user and is a device that connects to the system to exchange information.

[0875] "Analysis" is the process of understanding input information and extracting meaning from it.

[0876] An "HTTP request" is a protocol for sending and receiving data over the Internet.

[0877] This invention is a system that uses a smartphone application equipped with a conversational AI to provide recipes according to the user's request. A specific embodiment of this system will be described below.

[0878] First, a user launches a smartphone application. The device connects to a server over the Internet and loads a conversational AI model. This conversational AI model includes a natural language processing library (e.g., NLTK or spaCy) and a generative AI model (e.g., GPT-3).

[0879] The user inputs a request, such as "I want to eat Japanese food today" or "Tell me about some dishes that use chicken," into the text input field of the application. The device sends this input as text data to the server. An HTTP request is used for transmission.

[0880] The server analyzes the received text data using a natural language processing library to understand the user's intent. Based on the analysis results, it generates and sends an appropriate prompt to the generative AI model. For example, it generates the following prompt sentence:

[0881] User request: I want to eat Japanese food today

[0882] Prompt for generative AI model: Generate a recipe for a Japanese dish. Please provide detailed instructions for cooking and a list of ingredients needed to make it easy for users to follow. Also include links to images and videos that show the cooking steps.

[0883] The generative AI model receives the prompts and generates a recipe that matches the user's needs. The generated recipe includes cooking instructions, a list of ingredients needed, and links to related images and videos. The server sends the generated recipe information to the device, again using an HTTP request.

[0884] The device displays the received recipe information on the user's smartphone screen. The user can cook the dish while checking the text, images, and videos. For example, if the user enters "I want to eat Japanese food today," the server sends the following prompt to the generative AI model:

[0885] Prompt: The user says they want to eat Japanese food. Generate an easy-to-make Japanese recipe. Describe the cooking steps in detail and provide a list of ingredients needed. Also include links to images and videos that show the cooking steps.

[0886] The generative AI model generates a recipe for "Teriyaki Chicken" and the server sends it to the device, which displays it to the user as follows:

[0887] Recipe: Chicken Teriyaki

[0888] Ingredients: chicken thighs, soy sauce, mirin, sugar, sake

[0889] procedure:

[0890] 1. Cut the chicken thighs into bite-sized pieces.

[0891] 2. Heat oil in a frying pan and fry the chicken.

[0892] 3. Add the sauce made from soy sauce, mirin, sugar and sake and bring to a boil.

[0893] 4. When it's done, arrange it on a plate.

[0894] Image link: [Chicken Teriyaki Image]

[0895] Video link: [Chicken Teriyaki Cooking Video]

[0896] If the user has any questions or difficulties during the cooking process, they again enter their question into the application's text input field, and the server sends the following prompt to the generative AI model:

[0897] Prompt: The user says they don't know the amount of sauce. Please explain in detail the amount of sauce in the chicken teriyaki.

[0898] The generative AI model generates the answer "2 tablespoons soy sauce, 2 tablespoons mirin, 1 tablespoon sugar, 1 tablespoon sake" and the server sends it to the device. The device displays it to the user as follows:

[0899] Answer: The amount of sauce for chicken teriyaki is 2 tablespoons of soy sauce, 2 tablespoons of mirin, 1 tablespoon of sugar, and 1 tablespoon of sake.

[0900] In this way, users can easily get cooking recipes through the application and also receive support during the cooking process.

[0901] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0902] Step 1:

[0903] The user launches an application.

[0904] Specific operation: The user taps the application icon on the smartphone's home screen to launch it.

[0905] Input: User actions

[0906] Output: Application launch

[0907] Step 2:

[0908] The device connects to the server and loads the conversational AI model.

[0909] Specific operation: The device connects to a server via the internet and loads a conversational AI model (e.g. GPT-3).

[0910] Input: Start application

[0911] Output: Conversational AI model loaded

[0912] Step 3:

[0913] The user inputs a request.

[0914] Specific operation: The user inputs a request such as "I want to eat Japanese food today" or "Please tell me some recipes that use chicken" into a text input field of the application.

[0915] Input: User's request text

[0916] Output: The requested text

[0917] Step 4:

[0918] The terminal transmits the input request to the server.

[0919] Specific operation: The terminal sends the user's input as text data to the server using an HTTP request.

[0920] Input: The requested text

[0921] Output: HTTP request to the server

[0922] Step 5:

[0923] The server analyzes the request and sends prompts to the generative AI model.

[0924] Specific operation: The server analyzes the received text data using a natural language processing library (e.g., spaCy) to understand the user's intent. Based on the analysis results, it generates and sends an appropriate prompt to the generative AI model.

[0925] Input: Request text received in HTTP request

[0926] Output: prompts to the generative AI model

[0927] Step 6:

[0928] A generative AI model generates the recipe.

[0929] What it does: A generative AI model (e.g. GPT-3) takes the prompt and generates a recipe tailored to the user's needs, including cooking instructions, a list of needed ingredients, and links to related images and videos.

[0930] Input: A prompt for the generative AI model

[0931] Output: Generated recipe information

[0932] Step 7:

[0933] The server sends the generated recipe to the terminal.

[0934] Specific operation: The server sends the generated recipe information to the terminal, again using an HTTP request.

[0935] Input: Generated recipe information

[0936] Output: HTTP request to the terminal

[0937] Step 8:

[0938] The terminal displays the recipe to the user.

[0939] Specific operation: The device displays the received recipe information on the user's smartphone screen. The user can then prepare the dish while checking the text, images, and videos.

[0940] Input: Recipe information received via HTTP request

[0941] Output: The recipe displayed on the user's smartphone screen

[0942] Step 9:

[0943] If the user has any questions or problems while cooking, they can contact the AI ​​again.

[0944] Specific operation: If the user has any questions or problems while cooking, they enter their questions into the application's text input field again.

[0945] Input: User's additional question text

[0946] Output: The additional question text entered

[0947] Step 10:

[0948] The server generates answers to the additional questions and sends them to the terminal.

[0949] Specific operation: The server receives the user's additional question, again uses the generative AI model to generate an appropriate answer, and sends the generated answer to the terminal, which displays it to the user.

[0950] Input: Additional question text received in HTTP request

[0951] Output: Generated answer information and HTTP request to the terminal

[0952] (Application example 1)

[0953] 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".

[0954] Conventional recipe provision systems using conversational AI can provide recipes that match the user's preferences, but they do not provide sufficient support when the user actually cooks the recipe. In addition, there is no way for the user to immediately enjoy the suggested dishes. Furthermore, there is a lack of a way for the user to respond in real time when they have questions or problems about cooking. This reduces user convenience and leads to issues with satisfaction.

[0955] 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.

[0956] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for suggesting dishes that suit the user's preferences and ordering them for delivery, a means for providing detailed information on the cooking process and ingredients of the suggested dishes in the form of images and videos, and a means for the user to inquire of the AI ​​if they have any questions or problems regarding cooking. This allows the user to not only receive suggestions for dishes that suit their preferences, but also to immediately order the dishes for delivery, and furthermore, to visually check the cooking process and detailed information on the ingredients, thereby enhancing cooking support. In addition, since questions can be answered in real time, the convenience and satisfaction of the user is improved.

[0957] "Conversational AI" is an artificial intelligence system that provides information through natural language dialogue with users and generates responses based on the user's requests.

[0958] A "recipe" is information that describes the ingredients and cooking steps for making a particular dish.

[0959] "Cooking instructions" are instructions that show the specific operations and steps required to prepare a dish.

[0960] "Feedback" refers to information provided by users, such as opinions, ratings, and impressions after use.

[0961] A "delivery order" is an ordering procedure for delivering the food selected by the user to a specified location.

[0962] A "cooking process" is a series of cooking operations or steps performed to complete a dish.

[0963] "Detailed information about ingredients" refers to specific information such as the type, amount, nutritional value, and origin of the ingredients used in the dish.

[0964] An "image" is a still image used to visually convey information.

[0965] A "video" is a moving image that conveys information visually.

[0966] An "inquiry" is a question or inquiry made by a user to resolve a concern or problem.

[0967] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, a means for suggesting dishes that suit the user's preferences and placing a delivery order, a means for providing images and videos of the cooking process and detailed information about ingredients of the suggested dishes, and a means for the user to contact the AI ​​if they have any questions or problems regarding cooking.

[0968] Program processing explanation

[0969] Hardware used

[0970] Smartphone

[0971] server

[0972] Software used

[0973] Python

[0974] OpenAI API

[0975] Data processing and calculation

[0976] The server receives user input and generates an appropriate recipe using a generative AI model. Specifically, when a user inputs "I want to eat Japanese food today" through their smartphone, the information is sent to the server. The server uses the OpenAI API to generate a recipe based on the user's request. This generated recipe is provided to the user in the form of text, images, and videos.

[0977] When a user places a delivery order for the suggested food, the server sends the order information to the delivery service, allowing the user to enjoy the suggested food immediately. If the user has any problems during the cooking process, they can contact the AI ​​via their smartphone. The AI ​​will respond to questions in real time and support the user.

[0978] Examples

[0979] If a user types "I want to eat Japanese food today," the server uses the OpenAI API to generate a recipe for "Teriyaki Chicken." This recipe is provided to the user in the form of text, images, and videos. If the user answers "Yes," the server sends the order information to a delivery service, and the food is delivered to the user's location.

[0980] Examples of prompt statements

[0981] What recipes would you suggest when a user says, "I want to eat Japanese food today"?

[0982] In this way, users can not only receive suggestions for dishes that suit their tastes, but also order those dishes for delivery immediately, and can visually check detailed information about the cooking process and ingredients, providing more support for cooking.In addition, users can have their questions answered in real time, improving convenience and satisfaction for users.

[0983] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0984] Step 1:

[0985] A user inputs a request for a meal using a smartphone terminal. For example, the user inputs, "I want to eat Japanese food today." This input data is text data including the user's preferences and requests.

[0986] Step 2:

[0987] The device sends the user's input data to the server. The server receives this input data and inputs it as a prompt sentence into the generative AI model. An example of a prompt sentence is, "Please tell me the recipe of the dish that you would suggest when the user says, 'I want to eat Japanese food today.'"

[0988] Step 3:

[0989] The server uses a generative AI model (OpenAI API) to generate a recipe based on the prompt text. This generated recipe is output as text data. For example, a recipe for "Teriyaki Chicken" is generated.

[0990] Step 4:

[0991] The server retrieves relevant media data from a database to complement the generated recipe in the form of images and videos, thus providing visual information to the user.

[0992] Step 5:

[0993] The server sends the generated recipe and related images and videos to the terminal. The terminal displays this data to the user. The user checks the proposed recipe and selects whether to place a delivery order.

[0994] Step 6:

[0995] When a user places a delivery order, the terminal transmits the order information to the server. The server transmits the order information to the delivery service and arranges for the food to be delivered. The order information includes the user's address and the order details.

[0996] Step 7:

[0997] If a user has any problems during the cooking process, they can ask the AI ​​through their device. The server receives the user's question and uses the generative AI model to generate an appropriate response. This response is output as text data and sent to the device.

[0998] Step 8:

[0999] The device displays the response received from the server to the user, who can then continue cooking while referring to the instructions and advice from the AI.

[1000] In this way, users can not only receive suggestions for dishes that suit their tastes, but also order those dishes for delivery immediately, and can visually check detailed information about the cooking process and ingredients, providing more support for cooking. In addition, users can have their questions answered in real time, improving convenience and satisfaction for users.

[1001] Example 2

[1002] 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".

[1003] Conventional recipe provision systems have difficulty in fully reflecting individual user preferences and feedback, and have been unable to increase user satisfaction. In addition, there was a lack of a mechanism to effectively learn from user feedback and reflect it in the next recipe, making it difficult to provide recipes that match the user's preferences.

[1004] 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.

[1005] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning it, a means for analyzing the user's feedback using a generative AI model and reflecting it in the next recipe, a terminal means for transmitting the user's feedback to the server, a means for the server to input the feedback to the generative AI model and generate a customized recipe, and a means for transmitting the recipe generated by the server to the terminal. This makes it possible to effectively learn the individual preferences and feedback of the user and reflect them in the next recipe.

[1006] "User input" refers to information such as text or voice provided by a user through a terminal.

[1007] A "recipe" is information that includes instructions for making a dish and a list of ingredients.

[1008] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through dialogue with a user and responds according to the user's requests.

[1009] The "means for instructing cooking steps" is a function for explaining and instructing the user on cooking steps.

[1010] "Feedback" refers to impressions and suggestions for improvement provided by users after cooking.

[1011] A "generative AI model" is an artificial intelligence model that analyzes user feedback and reflects it in the next recipe.

[1012] The "terminal means" is a device that allows a user to input feedback and transmit it to the server.

[1013] The "server" is a computer system that receives user feedback, generates customized recipes using a generative AI model, and sends them to the terminal.

[1014] A "customized recipe" is a cooking procedure or list of ingredients that has been adjusted based on user feedback and preferences.

[1015] The present invention is a system that provides recipes based on user input, learns feedback from the user, and reflects the feedback in the next recipe. A specific embodiment of this system will be described below.

[1016] First, a conversational AI means is used to provide a recipe for the user to cook. The AI ​​means instructs the user on the cooking steps of the recipe through dialogue with the user. For example, when the user inputs "I want to make spaghetti," the AI ​​means provides a recipe for spaghetti and instructs the user on the cooking steps step by step.

[1017] Next, after the user has finished cooking, they input their feedback using a device (such as a smartphone or tablet). The feedback includes specific impressions and suggestions for improvement, such as "I wish it was less spicy" or "I wish it had more vegetables." The device then sends this feedback to the server.

[1018] The server inputs the received feedback into a generative AI model, such as OpenAI's GPT-4. The model analyzes the feedback and learns the user's preferences and requests. The next time the user requests a recipe, the server generates a customized recipe based on the learned information and sends it to the device.

[1019] For example, if a user provides feedback such as "I would like it less spicy," the device sends this feedback to the server. The server inputs the feedback into a generative AI model, which then generates a prompt such as "Please make the recipe less spicy next time." The server generates a recipe for spaghetti with less spiciness based on this prompt and sends it to the device. The user can then use this customized recipe to cook next time.

[1020] In this way, the system can effectively learn from users' individual preferences and feedback and incorporate them into future recipes, thereby increasing user satisfaction.

[1021] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1022] Step 1:

[1023] The user cooks

[1024] A user cooks a dish using a provided recipe. For example, to make spaghetti, the user prepares the ingredients and cooks the dish according to the recipe. The input is the provided recipe and the output is the finished dish.

[1025] Step 2:

[1026] User enters feedback

[1027] After cooking is finished, the user uses the terminal to input feedback. For example, they input specific impressions and suggestions for improvement, such as "I'd like it to be less spicy" or "I'd like to use more vegetables." The input is the user's impressions and suggestions for improvement, and the output is the feedback data.

[1028] Step 3:

[1029] The device sends feedback to the server

[1030] The terminal transmits the feedback input by the user to the server. At this time, the feedback is transmitted as text data. The input is the feedback data, and the output is the feedback transmitted to the server.

[1031] Step 4:

[1032] The server inputs the feedback into the generative AI model

[1033] The server inputs the received feedback into a generative AI model. For example, when using OpenAI's GPT-4, the feedback is passed to the model as a prompt sentence. The input is the feedback data, and the output is the prompt sentence.

[1034] Step 5:

[1035] Generative AI models learn from feedback

[1036] The generative AI model analyzes the feedback and learns the user's preferences and requests. For example, if it receives feedback such as "Make it less spicy," it will learn to make the recipe less spicy the next time it generates a recipe. The input is the prompt, and the output is the learning result.

[1037] Step 6:

[1038] The server generates a customized recipe

[1039] The server uses the information from the generative AI model to generate a customized recipe based on the user's preferences. For example, it creates a recipe for spaghetti with less spice. The input is the learning result, and the output is the customized recipe.

[1040] Step 7:

[1041] Send the recipe generated by the server to the terminal

[1042] The server sends the generated customized recipe to the terminal. The recipe is sent as text data. The input is the customized recipe, and the output is the recipe sent to the terminal.

[1043] Step 8:

[1044] User receives customized recipe

[1045] The user receives the customized recipe on the terminal and uses it for the next cooking. For example, the user can check the recipe for spaghetti with less spice to help with the next cooking. The input is the recipe sent to the terminal, and the output is the customized recipe received by the user.

[1046] (Application example 2)

[1047] 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".

[1048] Conventional food delivery systems have a problem in that it is difficult to suggest recipes that fully reflect the user's preferences and feedback. In particular, even if a user provides specific feedback on a dish, there is a lack of a mechanism to reflect that feedback in the next order, making it difficult to improve user satisfaction.

[1049] 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. In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, and a means for proposing the next recipe based on the user's feedback. This makes it possible to propose recipes that reflect the user's preferences and feedback.

[1050] "User input" means information or instructions provided by a user to a system.

[1051] A "recipe" is a list of ingredients and steps for making a particular dish.

[1052] "Conversational AI" is an artificial intelligence system that uses natural language processing technology to interact with users and provide them with information.

[1053] A "cooking procedure" is the order or method of specific steps required to prepare a dish.

[1054] "Feedback" refers to thoughts and suggestions for improvement provided by users regarding the system.

[1055] "Means of learning" refers to the techniques and methods by which the system analyzes user feedback and incorporates it into its next suggestions.

[1056] "Recipe suggestions" refers to suggestions of cooking ingredients and steps provided to the user by the system.

[1057] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, and a means for suggesting the next recipe based on the user's feedback.

[1058] System Configuration

[1059] The system uses the following hardware and software:

[1060] Hardware: Smartphone

[1061] Software: Python, scikit-learn

[1062] Program Processing

[1063] The server receives input from the user and generates a recipe based on it. It instructs the user on the cooking steps through dialogue. After the user finishes cooking, they can provide feedback. This feedback is collected by the server and saved as learning data. The next time a recipe is proposed, a recipe tailored to the user's preferences will be suggested based on this feedback.

[1064] Examples

[1065] For example, if a user provides feedback such as "I want it less spicy," the next recipe will be suggested with less spiciness. This makes it possible to suggest recipes that are tailored to the user's preferences.

[1066] Examples of prompt statements

[1067] Below are examples of prompts to be input to a generative AI model.

[1068] User ID: user123

[1069] Feedback: I would like it to be less spicy

[1070] Next suggestion: A less spicy recipe

[1071] In this way, recipe suggestions that reflect user feedback are realized.

[1072] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1073] Step 1:

[1074] A user accesses the system using a smartphone and requests a recipe. The user's input (request content) is sent to the server. The server receives this input and starts processing it as initial data for generating a recipe.

[1075] Step 2:

[1076] The server uses conversational AI to generate specific recipes through dialogue with the user. The conversational AI uses natural language processing technology to analyze the user's request and select an appropriate recipe. In this process, the user's preferences and past history are also taken into account. The input is the user's request, and the output is the generated recipe.

[1077] Step 3:

[1078] The server provides the generated recipe to the user and instructs the cooking procedure. The user cooks while checking the cooking procedure through their smartphone. The input is the generated recipe, and the output is the cooking procedure instruction to the user.

[1079] Step 4:

[1080] After the user finishes cooking, he / she provides feedback. The feedback is sent to the server via the smartphone. The input is the user's feedback, and the output is the feedback data stored in the server.

[1081] Step 5:

[1082] The server analyzes the received feedback and saves it as training data. Specifically, the feedback content is divided into features and evaluation, and training is performed using a linear regression model. The input is the feedback data, and the output is the updated training model.

[1083] Step 6:

[1084] The next time the user requests a recipe, the server uses the learned model to suggest recipes that match the user's preferences. The server takes into account the previous feedback and generates a recipe optimized for the user's preferences. The input is the user's new request and the learned model, and the output is the optimized recipe.

[1085] In this way, recipe suggestions that reflect user feedback are realized.

[1086] Example 3

[1087] 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".

[1088] Conventional conversational AI systems have had difficulty providing recipes that fully reflect the user's tastes and preferences. In addition, they lacked the means to effectively utilize the user's past conversation history, and the recipes they provide based on the user's preferences were limited. Furthermore, they lacked the ability to effectively learn from user feedback and evolve recipes.

[1089] 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.

[1090] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for storing the user's past conversation history about meals in a database, a means for analyzing the stored conversation history and learning the user's preferences, a means for using a generative AI model for generating recipes based on the user's preferences, and a means for providing the generated recipes to the user. This makes it possible to provide recipes that deeply correspond to the user's preferences and tastes.

[1091] A "conversational artificial intelligence means" is a technology that provides information through natural language dialogue with a user and generates responses according to the user's requests.

[1092] A "means for instructing cooking procedures" is a technology that provides users with step-by-step instructions on each step of a recipe, supporting them in the progress of cooking.

[1093] "Means of receiving feedback and learning from it" refers to technology that collects opinions and evaluations provided by users and uses them to improve the system's performance and content.

[1094] "Means for storing past conversation history in a database" refers to a technology that records the content of past conversations with a user and stores it in a database for future reference.

[1095] "Means for analyzing conversation history and learning user preferences" refers to technology that analyzes stored conversation history to identify user preferences and tastes and reflects them in the system.

[1096] "Means of using generative AI models" refers to technologies that utilize machine learning models to generate appropriate information and content based on user preferences and requests.

[1097] A "means for generating recipes" is a technology that automatically creates cooking recipes that meet a user's requests and preferences.

[1098] A "means for providing recipes" is a technology that presents the generated recipes to users and enables them to view and use them.

[1099] This invention relates to a conversational AI system that provides recipes based on user input. The system has the function of instructing the cooking steps of a recipe through conversation with the user, receiving feedback from the user, and learning from it. The system also stores the user's past conversation history about meals in a database, and analyzes the stored conversation history to learn the user's preferences. Furthermore, the system uses a generative AI model that generates recipes based on the user's preferences, and provides the generated recipes to the user.

[1100] Hardware and Software Used

[1101] Hardware:

[1102] Server: Manages the database, runs the generative AI models, and provides the interface with the user.

[1103] Terminal: A device such as a smartphone or PC used by a user.

[1104] software:

[1105] Database management system: Stores user conversation history and preference information.

[1106] Generative AI models: For example, using natural language processing models such as GPT-4.

[1107] Conversational artificial intelligence system: manages dialogue with the user, providing recipes and cooking instructions.

[1108] Program Processing

[1109] The server executes a conversational artificial intelligence means that provides recipes based on user input. When the user says "I like spicy food" through the terminal, the terminal transmits the statement to the server. The server stores the statement in a database and analyzes it using a generative AI model. As a result of the analysis, it is determined that the user likes spicy food.

[1110] Later, when the user requests, "What recipes do you recommend for dinner tonight?", the server inputs the following prompt into the generative AI model:

[1111] Example prompt:

[1112] You know from the user's past conversation history that they like spicy food, so provide them with some recommended spicy food recipes.

[1113] The generative AI model generates a recipe that matches the user's preferences based on this prompt. For example, a recipe for spicy chicken curry is generated. The server sends the generated recipe to the terminal, and the terminal displays to the user, "Here is the recipe for spicy chicken curry."

[1114] In this way, a system is realized in which the server, the terminal, and the user work together to provide recipes that match the user's preferences. The flow of the specification process in the third embodiment will be described with reference to FIG.

[1115] Step 1:

[1116] A user starts a dialogue with the conversational AI system through a terminal. For example, the user says, "I like spicy food."

[1117] Input: What the user says

[1118] Output: User utterance data

[1119] Step 2:

[1120] The terminal transmits the user's utterance data to the server, which then stores the received utterance data in a database.

[1121] Input: User utterance data

[1122] Output: Speech data stored in a database

[1123] Step 3:

[1124] The server periodically analyzes the conversation history in the database and uses a generative AI model to extract keywords and phrases from users' utterances using natural language processing techniques.

[1125] Input: Conversation history in a database

[1126] Output: Extracted keywords and phrases

[1127] Step 4:

[1128] The server learns the user's preferences based on the extracted keywords and phrases. For example, based on the information "I like spicy food," the server recognizes that the user likes spicy food. This information is stored in a database as a user profile.

[1129] Input: Extracted keywords or phrases

[1130] Output: Updated user profile

[1131] Step 5:

[1132] The user requests a new recipe through the terminal, for example, by saying, "Tell me a recommended recipe for tonight's dinner."

[1133] Input: User request

[1134] Output: User request data

[1135] Step 6:

[1136] The server receives the user's request data and inputs a prompt sentence into the generative AI model. The prompt sentence includes the user's preference information. For example, the prompt sentence might say, "We know from the user's past conversation history that the user likes spicy food. Please provide some recommended spicy food recipes for the user."

[1137] Input: User request data, user preferences

[1138] Output: Prompt statement

[1139] Step 7:

[1140] The generative AI model uses the prompt text to generate recipes that match the user's preferences, for example a recipe for spicy chicken curry.

[1141] Input: prompt statement

[1142] Output: The generated recipe

[1143] Step 8:

[1144] The server transmits the generated recipe to the terminal. The terminal displays the recipe to the user. For example, the recipe is provided in the form of "Here is the recipe for spicy chicken curry."

[1145] Input: The generated recipe

[1146] Output: The recipe displayed to the user

[1147] In this way, a system is realized in which the server, terminal, and user work together to provide recipes that suit the user's preferences.

[1148] (Application example 3)

[1149] 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".

[1150] Conventional recipe provision systems have difficulty in providing recipes that fully reflect the user's tastes and preferences, and therefore have been unable to increase user satisfaction. In addition, they lacked the functionality to suggest restaurants and menus that the user would like, and therefore were unable to comprehensively support the user's dining experience.

[1151] 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.

[1152] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing recipes based on the history, and a means for suggesting restaurants based on the user's preferences. This makes it possible to provide recipes and suggest restaurants that reflect the user's preferences and tastes.

[1153] "User input" means information or instructions provided by a user to a system.

[1154] A "conversational AI that provides recipes" is an artificial intelligence that suggests cooking recipes through dialogue with the user.

[1155] A "means for instructing the cooking steps of a recipe" refers to a function that provides specific guidance to the user on the cooking steps.

[1156] "Means of receiving and learning from user feedback" refers to the ability of the system to collect opinions and evaluations provided by users and make improvements based on them.

[1157] "A means of referencing the user's past conversation history about food and providing recipes based on that" refers to a function that uses records of the user's past conversations to suggest recipes that suit the user's preferences.

[1158] "A means of suggesting restaurants based on user preferences" refers to a function that recommends suitable restaurants based on the user's preferences and past conversation history.

[1159] A "generative AI model" is a model that uses artificial intelligence to generate new information and patterns from data.

[1160] A "prompt sentence" is an instruction sentence entered into an artificial intelligence to elicit a specific response.

[1161] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referencing the user's past conversation history about meals and providing recipes based thereon, and a means for suggesting restaurants based on the user's preferences.

[1162] System Configuration

[1163] The server uses the following hardware and software:

[1164] Hardware: powerful processor, sufficient memory, database server

[1165] Software: conversational AI engine (e.g. OpenAI's GPT-3), database management system (e.g. SQLite), user interface (e.g. smartphone app)

[1166] Data processing and calculation

[1167] The server processes the data in the following steps:

[1168] 1. Receive user input: Receive information entered by the user through the smartphone app.

[1169] 2. Recipe provision by conversational AI: A conversational AI engine generates appropriate recipes based on user input.

[1170] 3. Cooking Instructions: Provide the user with step-by-step instructions for cooking a recipe.

[1171] 4. Collect feedback and learn: Collect feedback from users and let the AI ​​engine learn from that information.

[1172] 5. Referencing past conversation history: Retrieve the user’s past conversation history from the database and provide recipes based on the user’s preferences.

[1173] 6. Restaurant Suggestions: Suggest suitable restaurants based on user preferences.

[1174] Examples

[1175] For example, if a user has previously mentioned that they like spicy food, the system will prioritize spicy recipes based on that information and also suggest restaurants that serve spicy food.

[1176] Example prompt:

[1177] User conversation history: "I like spicy food. I especially like curry and mapo tofu."

[1178] Extract user preferences.

[1179] Using this prompt, a generative AI model extracts the user's preferences and uses the results to provide recipes and restaurant suggestions.

[1180] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1181] Step 1:

[1182] Users input meal information through a smartphone app.

[1183] Input: User's dietary information (e.g. favorite dishes, allergies, etc.)

[1184] Output: The user's input information is sent to the server.

[1185] Specific behavior: A user enters information into an input form in the app and presses the submit button.

[1186] Step 2:

[1187] The server receives the user's input information and sends it to the conversational AI engine.

[1188] Input: User input information

[1189] Output: User information sent to the conversational AI engine

[1190] Specific operation: The server analyzes the user information it receives, converts it into an appropriate format, and sends it to the AI ​​engine.

[1191] Step 3:

[1192] A conversational AI engine generates recipes based on user input.

[1193] Input: User input information

[1194] Output: The generated recipe

[1195] What it does: The AI ​​engine generates an appropriate recipe taking into account the user's preferences and constraints.

[1196] Step 4:

[1197] The server sends the generated recipe to the user.

[1198] Input: The generated recipe

[1199] Output: A recipe displayed on the user's smartphone app

[1200] Specific operation: The server sends the generated recipe to the user's smartphone app and displays it.

[1201] Step 5:

[1202] The user follows the recipe, cooks the dish and then provides feedback in the app.

[1203] Input: User feedback (e.g. recipe rating, suggestions for improvement, etc.)

[1204] Output: The feedback information is sent to the server.

[1205] Specific action: After cooking, the user enters information into the app's feedback form and presses the submit button.

[1206] Step 6:

[1207] The server receives user feedback and sends it to the AI ​​engine for learning.

[1208] Input: User feedback information

[1209] Output: The trained AI model

[1210] Specific operation: The server analyzes the received feedback information and sends it to the AI ​​engine to update the model.

[1211] Step 7:

[1212] The server retrieves the user's past conversation history from the database and sends it to the AI ​​engine.

[1213] Input: User's past conversation history

[1214] Output: Conversation history sent to the AI ​​engine

[1215] Specific operation: The server retrieves the user's past conversation history from the database and sends it to the AI ​​engine.

[1216] Step 8:

[1217] The AI ​​engine extracts the user's preferences based on past conversation history and provides recipes.

[1218] Input: User's past conversation history

[1219] Output: A recipe based on the user's preferences.

[1220] How it works: The AI ​​engine analyzes conversation history, extracts user preferences, and generates appropriate recipes.

[1221] Step 9:

[1222] The server suggests restaurants based on the user's preferences.

[1223] Input: User preferences

[1224] Output: Suggested restaurant information

[1225] Specific operation: The server searches the database for suitable restaurants based on the user's preferences and makes suggestions.

[1226] 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.

[1227] "Example 1"

[1228] One embodiment of the present invention is a system that provides a conversational AI that includes an emotion engine that recognizes the emotions of a user. The system provides a recipe based on the user's input and instructs the user on how to cook the recipe through a conversation. The system also receives feedback from the user and learns from it. Specifically, if the user inputs, "I'm feeling sad today, so I want to eat something that will cheer me up," the emotion engine recognizes the user's "sad" mood and the AI ​​provides a recipe that is said to be uplifting.

[1229] "Example 2"

[1230] In addition, the emotion engine adjusts the recipe offerings based on the user's emotions. For example, if a user feels "very tired today," the emotion engine will recognize that information and offer recipes that are easy to make even for tired users. This makes it possible to offer optimal recipes according to the user's emotional state.

[1231] "Example 3"

[1232] The emotion engine also adjusts recipe cooking instructions based on the user's emotions. For example, if the emotion engine recognizes that the user is in a hurry, the AI ​​will provide concise and easy-to-understand cooking instructions. This makes it possible to provide optimal cooking instructions according to the user's emotional state.

[1233] The process flow of each embodiment will be described below.

[1234] "Example 1"

[1235] Step 1: The user gives input to the conversational AI. For example, the user might say, "I'm feeling sad today, so I want to eat something that will cheer me up."

[1236] Step 2: The emotion engine recognizes emotions from the user's input. In this example, the "sad" mood is recognized.

[1237] Step 3: The AI ​​provides a recipe based on the output of the emotion engine. In this example, it provides a recipe that is said to be uplifting.

[1238] "Example 2"

[1239] Step 1: A user gives input to the conversational AI. For example, the user might say, "I'm very tired today."

[1240] Step 2: The emotion engine recognizes the emotion from the user's input. In this example, the state "tired" is recognized.

[1241] Step 3: The AI ​​provides a recipe based on the output of the emotion engine. In this example, it provides a recipe that is easy to make even for tired users.

[1242] "Example 3"

[1243] Step 1: The user gives input to the conversational AI. For example, the user might say, "I'm anxious."

[1244] Step 2: The emotion engine recognizes emotions from the user's input. In this example, the state of "anxious" is recognized.

[1245] Step 3: The AI ​​gives instructions for cooking based on the output of the emotion engine. In this example, the instructions are given succinctly and clearly.

[1246] Example 1

[1247] 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".

[1248] Conventional conversational AI systems were able to provide recipes based on user input, but they lacked the ability to suggest recipes that took the user's emotions into account, or to learn from the user's feedback and reflect it in future suggestions. This made it difficult to provide optimal recipes based on the user's mood and preferences. In addition, they did not provide sufficient real-time support to users during the cooking process.

[1249] 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.

[1250] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking procedure of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion recognition means for recognizing the emotion of the user, and a means for providing a recipe based on the emotion of the user. This makes it possible to provide an optimal recipe according to the user's mood and preferences.

[1251] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through a dialogue with a user and generates appropriate responses based on the user's input.

[1252] The "means for instructing cooking steps" is a function that explains the specific cooking steps of a recipe to the user and sequentially instructs the user on the necessary steps.

[1253] "Means of learning feedback" refers to a function that collects evaluations and opinions from users and uses them to improve the system's performance and the content of its suggestions.

[1254] An "emotion recognition means" is a technology for analyzing emotions from a user's input or dialogue and recognizing those emotions.

[1255] A "means for providing recipes" is a function that generates appropriate cooking recipes based on the user's requests and feelings and provides them to the user.

[1256] This invention is a system that uses a smartphone application equipped with conversational artificial intelligence to provide optimal recipes based on the user's meal-related desires and emotions. The system analyzes the user's input, recognizes emotions, and generates recipes using a generative AI model. It also learns feedback from the user and reflects it in future suggestions.

[1257] Hardware and Software Used

[1258] Hardware

[1259] Device: Mobile devices such as smartphones and tablets

[1260] Server: Cloud or on-premise server

[1261] software

[1262] Natural Language Processing (NLP) technology: Google Cloud Natural Language API, IBM Watson Natural Language Understanding

[1263] Emotion recognition engine: Microsoft Azure Emotion API

[1264] Generative AI model: OpenAI's GPT-4

[1265] Machine learning algorithms: TensorFlow, PyTorch

[1266] Program Processing

[1267] User Input

[1268] The user starts the smartphone application and inputs their meal-related request into a text box, for example, "I'd like to eat Japanese food today" or "Tell me about some dishes that use chicken."

[1269] Analyzing data

[1270] The terminal sends the user's input to the server, which then analyzes the received text data using natural language processing technology. This analysis allows the server to understand the user's request.

[1271] Emotion recognition

[1272] Based on the analysis results, the server uses an emotion recognition engine to recognize the user's emotions. For example, if the user enters "I'm feeling sad today, so I want to eat something that cheers me up," the server recognizes the emotion "sad."

[1273] Recipe Suggestions

[1274] The server uses a generative AI model to generate recipes that match the user's preferences based on the analysis results and emotion recognition results. The generated recipes are sent to the device in the form of text, images, and videos.

[1275] User interaction

[1276] If a user has a question or difficulty while cooking, they can ask the AI ​​through the application. The device sends the question to the server, which then generates a response in real time and returns it to the device.

[1277] Learning Feedback

[1278] The user provides feedback on the provided recipe, and the device sends this feedback to the server, which then learns from the feedback using machine learning algorithms.

[1279] Examples

[1280] If a user types, "I want to eat Japanese food today," the device will send this request to the server. The server will analyze the request and suggest Japanese recipes using a generative AI model. For example, a recipe for "teriyaki chicken" will be provided in text, images, and video. If a user types, "I'm feeling sad today, so I want to eat something that will cheer me up," the server will recognize the emotion of "sadness" and suggest a recipe for "curry rice," which is said to be uplifting.

[1281] Examples of prompt statements

[1282] "A user says they want to eat Japanese food. Please suggest some Japanese recipes."

[1283] "A user says they're feeling sad. Can you suggest some cheer-up recipes?"

[1284] In this way, the system provides the optimal recipe based on the user's desires and feelings.

[1285] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1286] Step 1:

[1287] The user starts the smartphone application and inputs their meal-related requests into a text box. For example, they may make requests such as, "I'd like to eat Japanese food today," or "Tell me about some dishes that use chicken." The input data is stored in text format on the device.

[1288] Step 2:

[1289] The device sends the user's input data to the server. The server analyzes the received text data using natural language processing technology (Google Cloud Natural Language API and IBM Watson Natural Language Understanding). The user's request is extracted as a result of the analysis.

[1290] Step 3:

[1291] Based on the analysis results, the server uses an emotion recognition engine (Microsoft Azure Emotion API) to recognize the user's emotions. For example, if the user enters "I'm feeling sad today, so I want to eat something that cheers me up," the emotion "sad" is recognized. The emotion recognition results are stored on the server.

[1292] Step 4:

[1293] The server uses a generative AI model (OpenAI's GPT-4) to generate recipes that match the user's preferences based on the analysis results and emotion recognition results. The generated recipes are stored on the server in text, image, and video formats.

[1294] Step 5:

[1295] The server sends the generated recipe to the terminal. The terminal displays the received recipe to the user. The user confirms the recipe and starts cooking.

[1296] Step 6:

[1297] If a user has a question or difficulty while cooking, they can ask the AI ​​through the application. The device sends the question to the server, which generates a response in real time and returns it to the device. The user receives the response and continues cooking.

[1298] Step 7:

[1299] The user provides feedback on the provided recipe. The device sends this feedback to the server, which then uses machine learning algorithms (TensorFlow and PyTorch) to learn from the feedback. The learning results are reflected in future suggestions.

[1300] (Application example 1)

[1301] 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".

[1302] Conventional conversational AI systems could not suggest recipes based on the user's emotions or mood of the day, and could not sufficiently increase user satisfaction. In addition, there was no function to order delivery of the suggested recipes as they were, which reduced user convenience.

[1303] 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. In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine means for recognizing the user's emotions, a means for providing a recipe according to the user's emotions, and a means for ordering the suggested dish for delivery. This makes it possible to suggest recipes according to the user's emotions and mood of the day, and further allows the suggested dish to be ordered for delivery as is, thereby improving user satisfaction and convenience.

[1304] "User input" is information, such as text or voice, that a user provides to the system.

[1305] "Conversational AI" is artificial intelligence that provides information and gives instructions through natural conversation with the user.

[1306] A "cooking procedure" is a specific step or method for preparing a dish.

[1307] "Feedback" refers to reactions such as evaluations and opinions that users provide to a system.

[1308] An "emotion engine" is a technology that recognizes a user's emotions and responds appropriately based on those emotions.

[1309] A "recipe" is a detailed description of the ingredients and steps for making a particular dish.

[1310] A "delivery order" is an ordering procedure for delivering the food selected by the user to a specified location.

[1311] "Preferences" refer to personal preferences such as the ingredients and types of cuisine that the user particularly likes.

[1312] A "conversation history" is a record of past conversations between a user and the system.

[1313] A "server" is a computer that manages the entire system and processes and stores various data.

[1314] A system for implementing the present invention includes a conversational AI that provides a recipe based on user input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine that recognizes the user's emotions, a means for providing a recipe according to the user's emotions, and a means for ordering delivery of the suggested dishes.

[1315] System Configuration

[1316] Hardware

[1317] Server: Manages the entire system and processes and stores various data. Specifically, a cloud server or an on-premise server is used.

[1318] User device: A device that is directly operated by a user, such as a smartphone or tablet.

[1319] software

[1320] Conversational AI: Provides information and instructions through natural conversation with the user, using generative AI models such as OpenAI's GPT-3.

[1321] Sentiment engine: Recognizes the user's emotions and takes appropriate action based on those emotions, using sentiment analysis libraries such as TextBlob.

[1322] Delivery ordering system: A system for ordering delivery of the suggested dishes. Specifically, it will be linked to existing delivery service APIs.

[1323] Data processing and calculation

[1324] 1. Processing user input:

[1325] Users input requests such as "I want to eat Japanese food today" or "Please tell me some recipes that use chicken" via their smartphone.

[1326] This input is sent to the server in text format.

[1327] 2. Emotion recognition:

[1328] The server uses a sentiment analysis library such as TextBlob to analyze the sentiment of the user's input text.

[1329] For example, if a user types, "I'm feeling sad today, so I want to eat something that will cheer me up," the emotion engine will recognize this as "sad."

[1330] 3. Recipe suggestions:

[1331] The server uses a generative AI model (e.g. GPT-3) to generate recipes based on the user's emotions and preferences.

[1332] An example of a prompt sentence is "The user says that he is feeling sad and wants to eat something that will cheer him up. Please suggest some suitable dishes."

[1333] 4. Delivery Order:

[1334] The server works with existing delivery service APIs to order delivery of the suggested recipes as is.

[1335] Order food from the nearest restaurant based on the user's location.

[1336] Examples

[1337] If a user types in "I'm tired today, so I want to eat something that will relax me" through their smartphone, the emotion engine will recognize this as "tired" and use the generative AI model to suggest relaxing dishes. The suggested dishes will then be ordered for delivery from the nearest restaurant based on the user's location.

[1338] An example of a prompt sentence is, "The user says that he feels tired and wants to eat something that will relax him. Please suggest some suitable dishes."

[1339] In this way, it is possible to suggest recipes based on the user's emotions and mood of the day, and the suggested dishes can be ordered for delivery, thereby increasing user satisfaction and convenience.

[1340] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1341] Step 1:

[1342] The user inputs requests such as "I want to eat Japanese food today" or "Tell me about some dishes that use chicken" via smartphone. The input is in text format and is sent from the smartphone to the server. The input data is text information that includes the user's request.

[1343] Step 2:

[1344] The server analyzes the received user input text. Specifically, it uses a text analysis engine to extract the user's request. The input data is the user's request text, and the output data is the analyzed request information.

[1345] Step 3:

[1346] The server analyzes the sentiment of the user's input text using a sentiment analysis library such as TextBlob. The input data is the user's request text, and the output data is the sentiment information (e.g., "sad," "tired," etc.).

[1347] Step 4:

[1348] The server uses a generative AI model (e.g., GPT-3) to generate recipes based on the user's emotions and preferences. The prompt sentence is input to the generative AI model, which outputs an appropriate recipe. The input data are emotion information and user request information, and the output data is the suggested recipe.

[1349] Step 5:

[1350] The server provides the suggested recipe to the user. Specifically, it transmits recipe text information, images, videos, etc. to the smartphone terminal. The input data is the suggested recipe information, and the output data is the recipe information displayed on the user terminal.

[1351] Step 6:

[1352] When a user checks the suggested recipes and places a delivery order, the user sends a delivery order request from the smartphone terminal to the server. The input data is the delivery order request, and the output data is the order information sent to the server.

[1353] Step 7:

[1354] The server works with existing delivery service APIs to order food from the nearest restaurant based on the user's location information. The input data is the user's location information and order information, and the output data is an order request to the delivery service.

[1355] Step 8:

[1356] A delivery service takes orders and delivers food to a specified location. The input data is the order request to the delivery service, and the output data is the delivery of the food to the user.

[1357] In this way, it is possible to suggest recipes based on the user's emotions and mood of the day, and the suggested dishes can be ordered for delivery, thereby increasing user satisfaction and convenience.

[1358] Example 2

[1359] 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".

[1360] Conventional recipe provision systems had difficulty providing recipes that corresponded to the individual preferences and emotional state of the user. In addition, they lacked the ability to effectively learn from user feedback and reflect it in future recipes. This made it difficult to increase user satisfaction.

[1361] 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.

[1362] In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine means for inputting the emotional state of the user and adjusting the recipe based thereon, and a means for inputting the feedback and the emotional state to a generative AI model and generating a recipe. This makes it possible to provide an optimal recipe according to the user's preferences and emotional state.

[1363] A "conversational artificial intelligence means" is an artificial intelligence having a function of providing a recipe based on a user's input and instructing the user on the cooking steps of the recipe through conversation with the user.

[1364] A "means for learning feedback" is a means that has the function of receiving feedback from a user, analyzing it, and reflecting it in subsequent recipe generation.

[1365] The "emotion engine means" is a means having a function of inputting the user's emotional state and adjusting the recipe based on that.

[1366] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to take user feedback and emotional state as input and generate recipes based on that.

[1367] A "means for generating recipes" is a means having a function for generating optimal recipes according to the user's preferences and emotional state.

[1368] The invention is a system including a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion engine means for inputting the user's emotional state and adjusting the recipe based thereon, and a means for inputting the feedback and the emotional state into a generative artificial intelligence model and generating a recipe.

[1369] System Configuration

[1370] This system is realized using the following hardware and software.

[1371] Hardware: The device on which the user enters information (smartphone, tablet, PC, etc.), and the server that processes the data

[1372] Software: Conversational AI, Emotion Engine, Generative AI Models

[1373] Program Processing

[1374] Learning Feedback

[1375] 1. After cooking, the user enters feedback into the device, such as "I would like it to be less spicy."

[1376] 2. The device receives the user feedback and transmits it to a server via the Internet.

[1377] 3. The server analyzes the received feedback and processes it as text data. For example, it converts the feedback into a specific instruction such as "Make it less spicy."

[1378] 4. The server inputs the analyzed feedback into the generative AI model. Specifically, it sends a prompt to the generative AI model, such as "Make the recipe less spicy next time."

[1379] 5. The generative AI model receives the prompt and learns from the feedback, so that it can reflect the user's requests in future recipe generation.

[1380] Recipe adjustment using emotion engine

[1381] 1. The user inputs his / her emotional state into the device, such as "I feel very tired today."

[1382] 2. The device receives the user's emotional state and transmits it to a server via the Internet.

[1383] 3. The server passes the received emotional state to the emotion engine. Specifically, it sends information that "the user is tired" to the emotion engine.

[1384] 4. The emotion engine analyzes the user’s emotional state and requests “easy recipes for tired users” from the generative AI model.

[1385] 5. The emotion engine sends a prompt to the generative AI model saying, "Please provide a recipe that is easy for tired users to make."

[1386] 6. A generative AI model receives the prompt and generates recipes according to the user’s emotional state, for example providing recipes that require less cooking time and effort.

[1387] Examples

[1388] Feedback learning prompt:

[1389] User feedback: "I wish it was less spicy."

[1390] Learn how to tone down the spiciness to incorporate into your next recipe.

[1391] Emotion engine prompt:

[1392] User's emotional state: "I'm very tired today"

[1393] Provide recipes that are easy to make even for tired users.

[1394] This system makes it possible to provide optimal recipes based on the user's preferences and emotional state.

[1395] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1396] Learning Feedback

[1397] Step 1:

[1398] Users enter feedback after cooking.

[1399] Input: The user types into the terminal, "I would like it less spicy."

[1400] Output: Feedback data is saved to the device.

[1401] Specific action: The user enters feedback in text format using a device such as a smartphone or computer.

[1402] Step 2:

[1403] The device sends the feedback to the server.

[1404] Input: Feedback data stored on the device.

[1405] Output: Feedback data is sent to the server.

[1406] Specific operation: The terminal transmits feedback data to the server via the Internet.

[1407] Step 3:

[1408] The server parses the feedback.

[1409] Input: The feedback data sent to the server.

[1410] Output: Parsed feedback information (e.g. "Make it less spicy").

[1411] Specific Action: The server performs text analysis on the feedback data and converts it into specific instructions.

[1412] Step 4:

[1413] The server inputs feedback into the generative AI model.

[1414] Input: The parsed feedback information.

[1415] Output: The prompt sent to the generative AI model.

[1416] Specific operation: The server sends a prompt to the generative AI model, such as "Make the recipe less spicy next time."

[1417] Step 5:

[1418] Generative AI models learn from feedback.

[1419] Input: The prompt sent to the generative AI model.

[1420] Output: The learned feedback information.

[1421] Specific operation: The generative AI model receives the prompt text and reflects the user's requests the next time it generates a recipe.

[1422] Recipe adjustment using emotion engine

[1423] Step 1:

[1424] The user inputs an emotional state.

[1425] Input: A user types into a terminal, "I'm very tired today."

[1426] Output: Emotional state data is saved on the device.

[1427] Specific operation: The user inputs their emotional state in text format using a device such as a smartphone or computer.

[1428] Step 2:

[1429] The device transmits the emotional state to the server.

[1430] Input: Emotional state data stored on the device.

[1431] Output: Emotional state data is sent to the server.

[1432] Specific operation: The terminal transmits emotional state data to a server via the Internet.

[1433] Step 3:

[1434] The server passes the information to the emotion engine.

[1435] Input: Emotional state data sent to the server.

[1436] Output: Emotional state information sent to the emotion engine.

[1437] Specific operation: The server sends information that "the user is tired" to the emotion engine.

[1438] Step 4:

[1439] The emotion engine analyzes the user's emotional state.

[1440] Input: Emotional state information sent to the emotion engine.

[1441] Output: Parsed emotional state information (e.g. "Quick recipes for tired users").

[1442] Specific operation: The emotion engine analyzes the emotional state information and requests an appropriate recipe from the generative AI model.

[1443] Step 5:

[1444] The emotion engine requests the appropriate recipe from the generative AI model.

[1445] Input: Parsed emotional state information.

[1446] Output: The prompt sent to the generative AI model.

[1447] Specific behavior: The emotion engine sends a prompt to the generative AI model saying, "Please provide a recipe that is easy for tired users to make."

[1448] Step 6:

[1449] A generative AI model provides recipes based on the user's emotional state.

[1450] Input: The prompt sent to the generative AI model.

[1451] Output: A recipe depending on the user's emotional state.

[1452] How it works: A generative AI model receives a prompt and generates recipes that are quick and hassle-free to prepare.

[1453] (Application example 2)

[1454] 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".

[1455] Conventional recipe provision systems have difficulty in proposing recipes according to the individual preferences and emotional state of the user, and therefore have been unable to increase user satisfaction. In addition, there is a lack of a means to effectively utilize feedback from users, making it difficult to reflect this in the next recipe proposal.

[1456] 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.

[1457] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for analyzing the emotional state of the user and adjusting the recipe based on it, and a means for suggesting the next recipe based on the user's feedback and emotional state. This makes it possible to suggest optimal recipes according to the individual preferences and emotional state of the user, thereby increasing user satisfaction.

[1458] "User input" means information or instructions provided by a user to a system.

[1459] "Conversational AI" refers to artificial intelligence that uses natural language processing technology to interact with users, provide information, and give instructions.

[1460] A "means for instructing cooking steps" refers to a function that guides the user through cooking steps and methods.

[1461] "Means of receiving and learning from feedback" refers to the ability to collect opinions and feedback from users and use that information to make improvements and adjustments to the system.

[1462] "Means for analyzing the user's emotional state and adjusting the recipe based on that" refers to a function that analyzes the user's emotions and changes the recipe content or suggestions based on the results.

[1463] "Means to suggest next recipe" refers to a feature that takes into account the user's past feedback and emotional state to suggest the next dish they want to make.

[1464] A system for implementing the present invention includes a conversational AI that provides a recipe based on user input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for analyzing the user's emotional state and adjusting the recipe based thereon, and a means for suggesting the next recipe based on the user's feedback and emotional state.

[1465] Program processing explanation

[1466] The server executes a program using Python. Specifically, it performs the following processes:

[1467] 1. Receive user input:

[1468] Users input information into the system using their smartphones, including cooking requests, emotional states, and feedback.

[1469] 2. Conversational AI dialogue:

[1470] Conversational AI uses natural language processing techniques to converse with users to suggest recipes and provide cooking instructions, while using generative AI models to analyze user input and generate appropriate responses.

[1471] 3. Gather feedback and learn:

[1472] The feedback provided by users after cooking is collected and stored in a database, which is then learned from to improve recipe suggestions for the next time.

[1473] 4. Emotional state analysis:

[1474] It uses an emotion engine to analyze the user's emotional state. If a user types, "I'm very tired today," the emotion engine analyzes that information and suggests easy recipes.

[1475] 5. Recipe suggestions for next time:

[1476] It takes into account the user's past feedback and emotional state to suggest the next recipe, providing the best recipe based on the user's preferences and emotional state.

[1477] Hardware and Software Used

[1478] Hardware: Smartphone

[1479] Software: Python, natural language processing technology, emotion engine, generative AI model

[1480] Examples

[1481] When a user uses their smartphone to input "I'm very tired today," the emotion engine analyzes that information and suggests easy-to-make recipes. For example, "Stir-fried chicken and vegetables" is suggested. If the user inputs feedback after cooking, such as "I wish it was less spicy," a less spicy recipe will be suggested next time.

[1482] Examples of prompt statements

[1483] User: I'm really tired today.

[1484] AI: Well, what about a simple recipe? For example, how about "chicken and vegetable stir fry"?

[1485] User: I'd like it less spicy.

[1486] AI: I'll suggest a less spicy recipe next time. Thank you for your feedback.

[1487] In this way, it is possible to suggest optimal recipes based on the user's individual preferences and emotional state, thereby increasing user satisfaction.

[1488] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1489] Step 1:

[1490] A user inputs information into the system using a smartphone. The input information can be a food request, an emotional state, feedback, etc. For example, a user might input, "I'm very tired today." This input data is sent to the server.

[1491] Step 2:

[1492] The server analyzes the received user input data. Specifically, it uses natural language processing technology to understand the user's input and generate an appropriate response. At this time, a generative AI model is used to analyze the user's input and identify the user's emotional state. For example, the emotion "tired" is analyzed.

[1493] Step 3:

[1494] The server proposes recipes suitable for the user based on the analysis results. Using an emotion engine, it selects recipes according to the user's emotional state. For example, it suggests "chicken and vegetable stir fry" as an "easy recipe to make." The proposal is displayed on the smartphone.

[1495] Step 4:

[1496] The user checks the suggested recipe and cooks the food. After cooking, the user inputs feedback via their smartphone. For example, they can input feedback such as "I would like it less spicy." This feedback data is sent to the server.

[1497] Step 5:

[1498] The server stores the received feedback in a database. The feedback data is then learned so that it can be reflected in the next recipe suggestions. Specifically, a machine learning algorithm is used to analyze the feedback data and learn the user's preferences.

[1499] Step 6:

[1500] The next time the user enters input into the system again, the server will suggest recipes taking into account the user's past feedback and emotional state. For example, it may suggest a "less spicy recipe" based on the user's previous feedback. The suggestions are displayed on the smartphone.

[1501] In this way, it is possible to suggest optimal recipes according to the individual preferences and emotional state of the user, thereby increasing user satisfaction.

[1502] Example 3

[1503] 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".

[1504] Conventional conversational AI systems do not adequately provide recipes that correspond to the user's preferences and emotional state. In addition, there is a lack of means to effectively utilize the user's past conversation history, making it difficult to provide personalized recipes based on the user's preferences. Furthermore, cooking procedures are not adjusted according to the user's emotional state, which makes it difficult to increase user satisfaction.

[1505] 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.

[1506] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing a recipe based thereon, and a means for analyzing the user's emotional state and adjusting the cooking steps based thereon. This makes it possible to provide personalized recipes according to the user's preferences and emotional state, thereby increasing user satisfaction.

[1507] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through a dialogue with a user and generates appropriate responses based on the user's input.

[1508] "User input" is data, such as text or voice, that a user provides to the system.

[1509] A "recipe" is information that lists the ingredients and steps for making a particular dish.

[1510] A "cooking procedure" is the specific steps or methods to follow when preparing a dish.

[1511] "Feedback" refers to reactions such as evaluations and opinions that users provide to a system.

[1512] "Means of learning" are ways that the system can improve itself based on user feedback and past data.

[1513] "Past conversation history" is a record of previous interactions between a user and the system.

[1514] "Emotional state" refers to a user's current psychological state or mood.

[1515] A "means for analyzing" is a method for analyzing the content of data input by a user and extracting specific information.

[1516] A "generative artificial intelligence model" is a machine learning model used to generate appropriate responses or information based on user input.

[1517] The present invention relates to a conversational artificial intelligence system that provides recipes based on user input. A specific embodiment of this system will be described below.

[1518] System Configuration

[1519] The server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, a means for referring to the user's past conversation history regarding meals and providing a recipe based thereon, and a means for analyzing the user's emotional state and adjusting the cooking steps based thereon.

[1520] Hardware and Software Used

[1521] The server uses the following hardware and software:

[1522] Hardware: A high-performance server machine (e.g. a cloud server equipped with CPU, memory, and storage)

[1523] software:

[1524] Database management systems (e.g. MySQL, PostgreSQL)

[1525] Generative AI models (e.g. GPT-4)

[1526] Sentiment analysis engine (e.g. Emotion API)

[1527] Data processing and calculation

[1528] The server processes and calculates data in the following procedure.

[1529] 1. Collecting user conversation history:

[1530] The server retrieves the user's past conversation history from a database, for example by using MySQL to run a SQL query to search the conversation history using the user ID as a key.

[1531] 2. Learning user preferences:

[1532] The server uses the acquired conversation history to train a generative AI model (GPT-4) to learn the user's preferences and tastes. For example, if a user mentions that they like spicy food, the model memorizes that information.

[1533] 3. Generate the recipe:

[1534] When a user requests a recipe, the server uses the learned information to generate a recipe that matches the user's preferences using a generative AI model. For example, if the user likes spicy food, the server will suggest spicy food recipes.

[1535] 4. Analysis of the user's emotional state:

[1536] The server analyzes the user's current emotional state using an emotion engine (Emotion API). If the server detects that the user is in a hurry, it provides simple and easy-to-understand instructions on how to cook the food.

[1537] 5. Adjust cooking procedure:

[1538] The server adjusts the cooking instructions based on the emotion engine's analysis results. For example, if the user is in a hurry, the instructions will be simplified.

[1539] 6. Recipe provided by:

[1540] The server transmits the adjusted recipe to the user's terminal, and the user receives the recipe.

[1541] Examples

[1542] If the user has previously mentioned that they like spicy food, the following prompt sentence can be input into the generative AI model:

[1543] Example prompt:

[1544] You know from the user's past conversation history that they like spicy food, so provide them with a simple and easy-to-follow recipe for spicy food since they are in a rush.

[1545] An example of a generated recipe:

[1546] Spicy Chicken Curry Recipe:

[1547] 1. Cut the chicken into bite-sized pieces.

[1548] 2. Heat oil in a frying pan and fry the chicken.

[1549] 3. Add the onion, garlic and ginger and continue to stir fry.

[1550] 4. Add curry powder, tomato paste, and chili powder.

[1551] 5. Add water and simmer.

[1552] 6. Season with salt.

[1553] In this way, the server can provide the optimal recipe according to the user's preferences and emotional state. The flow of the specification process in the third embodiment will be described with reference to FIG.

[1554] Step 1:

[1555] The server retrieves the user's conversation history from the database. It receives the user ID as input and executes a SQL query to search the conversation history. As output, it obtains the user's past conversation history data. Specifically, the server executes the following SQL query:

[1556] sql

[1557] SELECT FROM conversation_history WHERE user_id = '12345';

[1558] Step 2:

[1559] The server uses the acquired conversation history to train a generative AI model (GPT-4) and learn the user's preferences and tastes. It receives the conversation history data as input and inputs it into the generative AI model. As output, it obtains a model that reflects the user's preferences. In concrete terms, the server inputs the conversation history as text data into the generative AI model.

[1560] python

[1561] conversation_history = "User: I like spicy food."

[1562] model.train(conversation_history)

[1563] Step 3:

[1564] When a user requests a recipe, the server uses the learned information to generate a recipe that matches the user's preferences using a generative AI model. It receives the user's request and the learned model as input, and inputs a prompt sentence into the generative AI model. As output, it obtains a recipe that matches the user's preferences. In concrete terms, the server inputs the following prompt sentence into the generative AI model.

[1565] python

[1566] prompt = "Can you give me some spicy food recipes?"

[1567] recipe = model.generate(prompt)

[1568] Step 4:

[1569] The server analyzes the user's current emotional state using an emotion engine (Emotion API). It receives the user's text and voice data as input and inputs it into the emotion engine. It obtains the user's emotional state as output. In concrete terms, the server inputs the user's input data into the emotion engine.

[1570] python

[1571] emotion = emotion_api.analyze(user_input)

[1572] if emotion == "anxious":

[1573] instructions = simplify_instructions(recipe)

[1574] Step 5:

[1575] The server adjusts the cooking instructions based on the analysis results of the emotion engine. It receives the emotional state and the generated recipe as input and applies an algorithm to simplify the cooking instructions. It obtains the adjusted recipe as output. Specifically, the server simplifies the instructions as follows:

[1576] python

[1577] def simplify_instructions(recipe):

[1578] Process to simplify the procedure

[1579] simplified_recipe = ...

[1580] return simplified_recipe

[1581] Step 6:

[1582] The server sends the adjusted recipe to the user's terminal. The server receives the adjusted recipe as input and sends it to the user's terminal. The user receives the recipe as output. In concrete terms, the server sends the recipe as follows:

[1583] python

[1584] send_to_user(simplified_recipe)

[1585] (Application example 3)

[1586] 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".

[1587] Conventional food delivery systems have the problem that it is difficult to suggest optimal dishes according to the individual preferences and emotional state of the user. In addition, since a uniform service is provided without considering the user's past conversation history or emotional state, user satisfaction may decrease. Furthermore, when the user is in a hurry, the cooking procedure is not adjusted according to the emotional state, which may increase the user's stress.

[1588] 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.

[1589] In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing a recipe based thereon, a means for analyzing the user's emotional state and adjusting the cooking steps based thereon, and a means for suggesting optimal dishes based on the user's preferences and emotional state and supporting ordering. This makes it possible to suggest and order optimal dishes according to the user's individual preferences and emotional state.

[1590] "Conversational AI" is an artificial intelligence system that provides information through natural language dialogue with users and services according to their requests.

[1591] A "recipe" is a set of instructions that lists the ingredients and steps for making a particular dish.

[1592] "Cooking instructions" are instructions that show the specific operations and steps required to prepare a dish.

[1593] "Feedback" refers to information provided by users, such as opinions, ratings, and impressions after use.

[1594] "Past conversation history" refers to a record of past interactions between a user and the system.

[1595] "Emotional state" refers to a user's current psychological state or mood.

[1596] "Preferences" refers to the food and flavor tendencies that a user particularly likes.

[1597] "Dish suggestions" refers to the act of selecting and presenting appropriate dishes based on the user's preferences and emotional state.

[1598] "Means for supporting ordering" refers to a function that assists the user in the process of actually ordering the selected dish.

[1599] A system for implementing the present invention includes a conversational AI that provides a recipe based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing a recipe based thereon, a means for analyzing the user's emotional state and adjusting the cooking steps based thereon, and a means for suggesting optimal dishes based on the user's preferences and emotional state and supporting ordering.

[1600] System Configuration

[1601] This system is realized using the following hardware and software.

[1602] Hardware: Smartphone

[1603] Software: Python, OpenAI API, sentiment analysis library (e.g. sentiment_analysis), food delivery API (e.g. food_delivery_api)

[1604] Data processing and calculation

[1605] Get a user's conversation history

[1606] The server retrieves the user's past conversation history from the database, including what dishes the user has liked in the past and what feedback he or she has provided.

[1607] Learn your preferences

[1608] The server analyzes the conversation history and learns the user's preferences. For example, if the user mentions that they like spicy food, the server remembers that information and prioritizes spicy food suggestions.

[1609] Analyze your emotional state

[1610] The server analyzes the user's emotional state from the current conversation, using a sentiment analysis library to determine whether the user is "anxious" or "relaxed."

[1611] Propose the best dish

[1612] The server suggests the best recipes based on the user's preferences and emotional state. For example, if the user is in a hurry, it suggests recipes with simple cooking steps.

[1613] Order food

[1614] The server then orders the suggested dishes through a food delivery API, allowing users to receive their food without any hassle.

[1615] Examples

[1616] If a user says, "I'm busy and in a hurry today," the system will learn from the user's past conversation history that they like spicy food, suggest a simple spicy dish, and then order that dish through a food delivery API.

[1617] Examples of prompt statements

[1618] User ID: 12345

[1619] Current conversation: I'm busy and anxious today.

[1620] By inputting this prompt into a generative AI model, the system can suggest and order the most appropriate dish based on the user's emotional state.

[1621] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1622] Step 1:

[1623] The server receives user input, which includes the user ID and the current conversation. Based on this input, the server retrieves the user's past conversation history from the database. The output is the user's past conversation history.

[1624] Step 2:

[1625] The server analyzes the acquired conversation history and learns the user's preferences. Specifically, it extracts information about preferences, such as "I like spicy food," from the conversation history and stores it in a user preference database. The input is the user's past conversation history, and the output is the user's preference data.

[1626] Step 3:

[1627] The server analyzes the current conversation content using a sentiment analysis library (e.g., sentiment_analysis) to determine the user's emotional state. For example, from the input "I'm busy today and I'm anxious," the emotional state "anxious" is extracted. The input is the current conversation content, and the output is the user's emotional state.

[1628] Step 4:

[1629] The server proposes the best dish based on the user's preference data and emotional state. For example, if the user "likes spicy food" and is "in a hurry," the server proposes a simple spicy dish. The input is the user's preference data and emotional state, and the output is the proposed dish.

[1630] Step 5:

[1631] The server orders the suggested dishes through a food delivery API (e.g. food_delivery_api) by sending the suggested dishes to the API and completing the order process. The input is the suggested dishes and the output is the order confirmation.

[1632] Step 6:

[1633] The server receives feedback from users and learns from it. Specifically, it analyzes the feedback provided by users and incorporates it into the next suggestions. The input is the user feedback and the output is the updated preference data.

[1634] 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 for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1635] 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.

[1636] Another example of generative AI is Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1637] 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.

[1638] [Third embodiment]

[1639] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1640] 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.

[1641] 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).

[1642] 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.

[1643] 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.

[1644] 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 a typical healthy person).

[1645] 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.

[1646] 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.

[1647] 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.

[1648] 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.

[1649] 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.

[1650] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1651] "Example 1"

[1652] One embodiment of the present invention is a smartphone application equipped with a conversational AI. The user converses with the AI ​​through the application and provides information about the day's meals. For example, the user can tell the AI ​​their preferences, such as "I want to eat Japanese food today" or "Please tell me about some dishes that use chicken." Based on this information, the AI ​​provides recipes that match the user's preferences. The recipes are not just text, but also use images and videos to explain the cooking steps in detail. Furthermore, if the user has any questions or problems during the cooking process, they can directly contact the AI.

[1653] "Example 2"

[1654] Furthermore, another embodiment of the present invention is a function that allows the AI ​​to learn from feedback from users. After cooking, the user can tell the AI ​​their thoughts and suggestions for improvement, and the AI ​​will learn from this information. For example, if the AI ​​receives feedback such as "I would like it less spicy" or "I would like to use more vegetables," the AI ​​will provide recipes that reflect these requests from the next time. This makes it possible to provide recipes that are tailored to the user's preferences.

[1655] "Example 3"

[1656] In addition, as a further embodiment of the present invention, there is a function where the AI ​​refers to the user's past conversation history regarding food. The AI ​​learns the user's tastes and preferences from past conversations with the user. For example, if the user has previously mentioned that they "like spicy food," the AI ​​can remember that information and provide recipes for spicy dishes. This makes it possible to provide recipes that more closely correspond to the user's tastes.

[1657] The process flow of each embodiment will be described below.

[1658] "Example 1"

[1659] Step 1: The user opens the smartphone application and starts a conversation with the conversational AI.

[1660] Step 2: The user provides the AI ​​with information about the day's meals. For example, the user may tell the AI ​​a request such as, "I want to eat Japanese food today," or "Please tell me about dishes that use chicken."

[1661] Step 3: Based on the information provided by the user, the AI ​​will select an appropriate recipe and provide the recipe and cooking instructions to the user.

[1662] Step 4: If the user has any questions or problems during the cooking process, they can directly contact the AI, which will provide an appropriate answer to the question.

[1663] "Example 2"

[1664] Step 1: After cooking, the user tells the AI ​​what they think and what improvements they would like to make.

[1665] Step 2: The AI ​​learns from user feedback and incorporates that information into its next recipe selection.

[1666] "Example 3"

[1667] Step 1: The AI ​​learns the user's preferences and tastes from past conversations with them.

[1668] Step 2: When a user requests a new recipe, the AI ​​will refer to past conversation history to provide recipes that match the user's preferences.

[1669] Example 1

[1670] 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".

[1671] Conventional recipe provision systems have difficulty in providing recipes that meet the specific needs of users, and are unable to respond quickly and appropriately to follow-up questions during the cooking process. Furthermore, they do not adequately consider the user's preferences or past conversation history when providing recipes. This leads to a decrease in user satisfaction and a decrease in the frequency of system use.

[1672] 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.

[1673] In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for generating a recipe that meets the user's needs using a generative AI model, a means for providing the generated recipe to the user in text, image, or video format, and a means for generating an answer using the generative AI model again when the user asks an additional question during the cooking process. This makes it possible to provide a recipe according to the user's specific needs, and to quickly and appropriately respond to additional questions during the cooking process. It also makes it possible to provide a recipe that takes into account the user's preferences and past conversation history, thereby improving user satisfaction.

[1674] "User" means an individual or organization that uses the system to receive recipes.

[1675] "Input" refers to the provision of information, such as text or voice, by a user to a system.

[1676] A "recipe" is information that lists the ingredients and steps for making a particular dish.

[1677] "Conversational AI" is an artificial intelligence system that provides information to users through natural language dialogue.

[1678] "Feedback" refers to reactions such as evaluations and opinions that users provide to a system.

[1679] A "generative AI model" is an artificial intelligence model that generates new information based on user requests.

[1680] A "prompt" is an input sentence that prompts a generative AI model to generate specific information.

[1681] "Text" is data that includes character information.

[1682] An "image" is data that contains visual information.

[1683] "Video" is data that contains dynamic visual information.

[1684] A "cooking procedure" is the sequence of specific steps to prepare a dish.

[1685] A "server" is a computer that is the central part of a system and is a device that processes and stores data.

[1686] A "terminal" is a device that is directly operated by a user and is a device that connects to the system to exchange information.

[1687] "Analysis" is the process of understanding input information and extracting meaning from it.

[1688] An "HTTP request" is a protocol for sending and receiving data over the Internet.

[1689] This invention is a system that uses a smartphone application equipped with a conversational AI to provide recipes according to the user's request. A specific embodiment of this system will be described below.

[1690] First, a user launches a smartphone application. The device connects to a server over the Internet and loads a conversational AI model. This conversational AI model includes a natural language processing library (e.g., NLTK or spaCy) and a generative AI model (e.g., GPT-3).

[1691] The user inputs a request, such as "I want to eat Japanese food today" or "Tell me about some dishes that use chicken," into the text input field of the application. The device sends this input as text data to the server. An HTTP request is used for transmission.

[1692] The server analyzes the received text data using a natural language processing library to understand the user's intent. Based on the analysis results, it generates and sends an appropriate prompt to the generative AI model. For example, it generates the following prompt sentence:

[1693] User request: I want to eat Japanese food today

[1694] Prompt for generative AI model: Generate a recipe for a Japanese dish. Please provide detailed instructions for cooking and a list of ingredients needed to make it easy for users to follow. Also include links to images and videos that show the cooking steps.

[1695] The generative AI model receives the prompts and generates a recipe that matches the user's needs. The generated recipe includes cooking instructions, a list of ingredients needed, and links to related images and videos. The server sends the generated recipe information to the device, again using an HTTP request.

[1696] The device displays the received recipe information on the user's smartphone screen. The user can cook the dish while checking the text, images, and videos. For example, if the user enters "I want to eat Japanese food today," the server sends the following prompt to the generative AI model:

[1697] Prompt: The user says they want to eat Japanese food. Generate an easy-to-make Japanese recipe. Describe the cooking steps in detail and provide a list of ingredients needed. Also include links to images and videos that show the cooking steps.

[1698] The generative AI model generates a recipe for "Teriyaki Chicken" and the server sends it to the device, which displays it to the user as follows:

[1699] Recipe: Chicken Teriyaki

[1700] Ingredients: chicken thighs, soy sauce, mirin, sugar, sake

[1701] procedure:

[1702] 1. Cut the chicken thighs into bite-sized pieces.

[1703] 2. Heat oil in a frying pan and fry the chicken.

[1704] 3. Add the sauce made from soy sauce, mirin, sugar and sake and bring to a boil.

[1705] 4. When it's done, arrange it on a plate.

[1706] Image link: [Chicken Teriyaki Image]

[1707] Video link: [Chicken Teriyaki Cooking Video]

[1708] If the user has any questions or difficulties during the cooking process, they again enter their question into the application's text input field, and the server sends the following prompt to the generative AI model:

[1709] Prompt: The user says they don't know the amount of sauce. Please explain in detail the amount of sauce in the chicken teriyaki.

[1710] The generative AI model generates the answer "2 tablespoons soy sauce, 2 tablespoons mirin, 1 tablespoon sugar, 1 tablespoon sake" and the server sends it to the device. The device displays it to the user as follows:

[1711] Answer: The amount of sauce for chicken teriyaki is 2 tablespoons of soy sauce, 2 tablespoons of mirin, 1 tablespoon of sugar, and 1 tablespoon of sake.

[1712] In this way, users can easily get cooking recipes through the application and also receive support during the cooking process.

[1713] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1714] Step 1:

[1715] The user launches an application.

[1716] Specific operation: The user taps the application icon on the smartphone's home screen to launch it.

[1717] Input: User actions

[1718] Output: Application launch

[1719] Step 2:

[1720] The device connects to the server and loads the conversational AI model.

[1721] Specific operation: The device connects to a server via the internet and loads a conversational AI model (e.g. GPT-3).

[1722] Input: Start application

[1723] Output: Conversational AI model loaded

[1724] Step 3:

[1725] The user inputs a request.

[1726] Specific operation: The user inputs a request such as "I want to eat Japanese food today" or "Please tell me some recipes that use chicken" into a text input field of the application.

[1727] Input: User's request text

[1728] Output: The requested text

[1729] Step 4:

[1730] The terminal transmits the input request to the server.

[1731] Specific operation: The terminal sends the user's input as text data to the server using an HTTP request.

[1732] Input: The requested text

[1733] Output: HTTP request to the server

[1734] Step 5:

[1735] The server analyzes the request and sends prompts to the generative AI model.

[1736] Specific operation: The server analyzes the received text data using a natural language processing library (e.g., spaCy) to understand the user's intent. Based on the analysis results, it generates and sends an appropriate prompt to the generative AI model.

[1737] Input: Request text received in HTTP request

[1738] Output: prompts to the generative AI model

[1739] Step 6:

[1740] A generative AI model generates the recipe.

[1741] What it does: A generative AI model (e.g. GPT-3) takes the prompt and generates a recipe tailored to the user's needs, including cooking instructions, a list of needed ingredients, and links to related images and videos.

[1742] Input: A prompt for the generative AI model

[1743] Output: Generated recipe information

[1744] Step 7:

[1745] The server sends the generated recipe to the terminal.

[1746] Specific operation: The server sends the generated recipe information to the terminal, again using an HTTP request.

[1747] Input: Generated recipe information

[1748] Output: HTTP request to the terminal

[1749] Step 8:

[1750] The terminal displays the recipe to the user.

[1751] Specific operation: The device displays the received recipe information on the user's smartphone screen. The user can then prepare the dish while checking the text, images, and videos.

[1752] Input: Recipe information received via HTTP request

[1753] Output: The recipe displayed on the user's smartphone screen

[1754] Step 9:

[1755] If the user has any questions or problems while cooking, they can contact the AI ​​again.

[1756] Specific operation: If the user has any questions or problems while cooking, they enter their questions into the application's text input field again.

[1757] Input: User's additional question text

[1758] Output: The additional question text entered

[1759] Step 10:

[1760] The server generates answers to the additional questions and sends them to the terminal.

[1761] Specific operation: The server receives the user's additional question, again uses the generative AI model to generate an appropriate answer, and sends the generated answer to the terminal, which displays it to the user.

[1762] Input: Additional question text received in HTTP request

[1763] Output: Generated answer information and HTTP request to the terminal

[1764] (Application example 1)

[1765] 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".

[1766] Conventional recipe provision systems using conversational AI can provide recipes that match the user's preferences, but they do not provide sufficient support when the user actually cooks the recipe. In addition, there is no way for the user to immediately enjoy the suggested dishes. Furthermore, there is a lack of a way for the user to respond in real time when they have questions or problems about cooking. This reduces user convenience and leads to issues with satisfaction.

[1767] 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.

[1768] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for suggesting dishes that suit the user's preferences and ordering them for delivery, a means for providing detailed information on the cooking process and ingredients of the suggested dishes in the form of images and videos, and a means for the user to inquire of the AI ​​if they have any questions or problems regarding cooking. This allows the user to not only receive suggestions for dishes that suit their preferences, but also to immediately order the dishes for delivery, and furthermore, to visually check the cooking process and detailed information on the ingredients, thereby enhancing cooking support. In addition, since questions can be answered in real time, the convenience and satisfaction of the user is improved.

[1769] "Conversational AI" is an artificial intelligence system that provides information through natural language dialogue with users and generates responses based on the user's requests.

[1770] A "recipe" is information that describes the ingredients and cooking steps for making a particular dish.

[1771] "Cooking instructions" are instructions that show the specific operations and steps required to prepare a dish.

[1772] "Feedback" refers to information provided by users, such as opinions, ratings, and impressions after use.

[1773] A "delivery order" is an ordering procedure for delivering the food selected by the user to a specified location.

[1774] A "cooking process" is a series of cooking operations or steps performed to complete a dish.

[1775] "Detailed information about ingredients" refers to specific information such as the type, amount, nutritional value, and origin of the ingredients used in the dish.

[1776] An "image" is a still image used to visually convey information.

[1777] A "video" is a moving image that conveys information visually.

[1778] An "inquiry" is a question or inquiry made by a user to resolve a concern or problem.

[1779] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, a means for suggesting dishes that suit the user's preferences and placing a delivery order, a means for providing images and videos of the cooking process and detailed information about ingredients of the suggested dishes, and a means for the user to contact the AI ​​if they have any questions or problems regarding cooking.

[1780] Program processing explanation

[1781] Hardware used

[1782] Smartphone

[1783] server

[1784] Software used

[1785] Python

[1786] OpenAI API

[1787] Data processing and calculation

[1788] The server receives user input and generates an appropriate recipe using a generative AI model. Specifically, when a user inputs "I want to eat Japanese food today" through their smartphone, the information is sent to the server. The server uses the OpenAI API to generate a recipe based on the user's request. This generated recipe is provided to the user in the form of text, images, and videos.

[1789] When a user places a delivery order for the suggested food, the server sends the order information to the delivery service, allowing the user to enjoy the suggested food immediately. If the user has any problems during the cooking process, they can contact the AI ​​via their smartphone. The AI ​​will respond to questions in real time and support the user.

[1790] Examples

[1791] If a user types "I want to eat Japanese food today," the server uses the OpenAI API to generate a recipe for "Teriyaki Chicken." This recipe is provided to the user in the form of text, images, and videos. If the user answers "Yes," the server sends the order information to a delivery service, and the food is delivered to the user's location.

[1792] Examples of prompt statements

[1793] What recipes would you suggest when a user says, "I want to eat Japanese food today"?

[1794] In this way, users can not only receive suggestions for dishes that suit their tastes, but also order those dishes for delivery immediately, and can visually check detailed information about the cooking process and ingredients, providing more support for cooking.In addition, users can have their questions answered in real time, improving convenience and satisfaction for users.

[1795] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1796] Step 1:

[1797] A user inputs a request for a meal using a smartphone terminal. For example, the user inputs, "I want to eat Japanese food today." This input data is text data including the user's preferences and requests.

[1798] Step 2:

[1799] The device sends the user's input data to the server. The server receives this input data and inputs it as a prompt sentence into the generative AI model. An example of a prompt sentence is, "Please tell me the recipe of the dish that you would suggest when the user says, 'I want to eat Japanese food today.'"

[1800] Step 3:

[1801] The server uses a generative AI model (OpenAI API) to generate a recipe based on the prompt text. This generated recipe is output as text data. For example, a recipe for "Teriyaki Chicken" is generated.

[1802] Step 4:

[1803] The server retrieves relevant media data from a database to complement the generated recipe in the form of images and videos, thus providing visual information to the user.

[1804] Step 5:

[1805] The server sends the generated recipe and related images and videos to the terminal. The terminal displays this data to the user. The user checks the proposed recipe and selects whether or not to place a delivery order.

[1806] Step 6:

[1807] When a user places a delivery order, the terminal sends the order information to the server. The server then sends the order information to the delivery service and arranges for the food to be delivered. The order information includes the user's address and the order details.

[1808] Step 7:

[1809] If a user has any problems during the cooking process, they can ask the AI ​​through their device. The server receives the user's question and uses the generative AI model to generate an appropriate response. This response is output as text data and sent to the device.

[1810] Step 8:

[1811] The device displays the response received from the server to the user, who can then continue cooking while referring to the instructions and advice from the AI.

[1812] In this way, users can not only receive suggestions for dishes that suit their tastes, but also order those dishes for delivery immediately, and can visually check detailed information about the cooking process and ingredients, providing more support for cooking. In addition, users can have their questions answered in real time, improving convenience and satisfaction for users.

[1813] Example 2

[1814] 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".

[1815] Conventional recipe provision systems have difficulty in fully reflecting individual user preferences and feedback, and have been unable to increase user satisfaction. In addition, there was a lack of a mechanism to effectively learn from user feedback and reflect it in the next recipe, making it difficult to provide recipes that match the user's preferences.

[1816] 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.

[1817] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning it, a means for analyzing the user's feedback using a generative AI model and reflecting it in the next recipe, a terminal means for transmitting the user's feedback to the server, a means for the server to input the feedback to the generative AI model and generate a customized recipe, and a means for transmitting the recipe generated by the server to the terminal. This makes it possible to effectively learn the individual preferences and feedback of the user and reflect them in the next recipe.

[1818] "User input" refers to information such as text or voice provided by a user through a terminal.

[1819] A "recipe" is information that includes instructions for making a dish and a list of ingredients.

[1820] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through dialogue with a user and responds according to the user's requests.

[1821] The "means for instructing cooking steps" is a function for explaining and instructing the user on cooking steps.

[1822] "Feedback" refers to impressions and suggestions for improvement provided by users after cooking.

[1823] A "generative AI model" is an artificial intelligence model that analyzes user feedback and reflects it in the next recipe.

[1824] The "terminal means" is a device that allows a user to input feedback and transmit it to the server.

[1825] The "server" is a computer system that receives user feedback, generates customized recipes using a generative AI model, and sends them to the terminal.

[1826] A "customized recipe" is a cooking procedure or list of ingredients that has been adjusted based on user feedback and preferences.

[1827] The present invention is a system that provides recipes based on user input, learns feedback from the user, and reflects the feedback in the next recipe. A specific embodiment of this system will be described below.

[1828] First, a conversational AI means is used to provide a recipe for the user to cook. The AI ​​means instructs the user on the cooking steps of the recipe through dialogue with the user. For example, when the user inputs "I want to make spaghetti," the AI ​​means provides a recipe for spaghetti and instructs the user on the cooking steps step by step.

[1829] Next, after the user has finished cooking, they input their feedback using a device (such as a smartphone or tablet). The feedback includes specific impressions and suggestions for improvement, such as "I wish it was less spicy" or "I wish it had more vegetables." The device then sends this feedback to the server.

[1830] The server inputs the received feedback into a generative AI model, such as OpenAI's GPT-4. The model analyzes the feedback and learns the user's preferences and requests. The next time the user requests a recipe, the server generates a customized recipe based on the learned information and sends it to the device.

[1831] For example, if a user provides feedback such as "I would like it less spicy," the device sends this feedback to the server. The server inputs the feedback into a generative AI model, which then generates a prompt such as "Please make the recipe less spicy next time." The server generates a recipe for spaghetti with less spiciness based on this prompt and sends it to the device. The user can then use this customized recipe to cook next time.

[1832] In this way, the system can effectively learn from users' individual preferences and feedback and incorporate them into future recipes, thereby increasing user satisfaction.

[1833] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1834] Step 1:

[1835] The user cooks

[1836] A user cooks a dish using a provided recipe. For example, to make spaghetti, the user prepares the ingredients and cooks the dish according to the recipe. The input is the provided recipe and the output is the finished dish.

[1837] Step 2:

[1838] User enters feedback

[1839] After cooking is finished, the user uses the terminal to input feedback. For example, they input specific impressions and suggestions for improvement, such as "I'd like it to be less spicy" or "I'd like to use more vegetables." The input is the user's impressions and suggestions for improvement, and the output is the feedback data.

[1840] Step 3:

[1841] The device sends feedback to the server

[1842] The terminal transmits the feedback input by the user to the server. At this time, the feedback is transmitted as text data. The input is the feedback data, and the output is the feedback transmitted to the server.

[1843] Step 4:

[1844] The server inputs the feedback into the generative AI model

[1845] The server inputs the received feedback into a generative AI model. For example, when using OpenAI's GPT-4, the feedback is passed to the model as a prompt sentence. The input is the feedback data, and the output is the prompt sentence.

[1846] Step 5:

[1847] Generative AI models learn from feedback

[1848] The generative AI model analyzes the feedback and learns the user's preferences and requests. For example, if it receives feedback such as "Make it less spicy," it will learn to make the recipe less spicy the next time it generates a recipe. The input is the prompt, and the output is the learning result.

[1849] Step 6:

[1850] The server generates a customized recipe

[1851] The server uses the information from the generative AI model to generate a customized recipe based on the user's preferences. For example, it creates a recipe for spaghetti with less spice. The input is the learning result, and the output is the customized recipe.

[1852] Step 7:

[1853] Send the recipe generated by the server to the terminal

[1854] The server sends the generated customized recipe to the terminal. The recipe is sent as text data. The input is the customized recipe, and the output is the recipe sent to the terminal.

[1855] Step 8:

[1856] User receives customized recipe

[1857] The user receives the customized recipe on the terminal and uses it for the next cooking. For example, the user can check the recipe for spaghetti with less spice to help with the next cooking. The input is the recipe sent to the terminal, and the output is the customized recipe received by the user.

[1858] (Application example 2)

[1859] 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".

[1860] Conventional food delivery systems have a problem in that it is difficult to suggest recipes that fully reflect the user's preferences and feedback. In particular, even if a user provides specific feedback on a dish, there is a lack of a mechanism to reflect that feedback in the next order, making it difficult to improve user satisfaction.

[1861] 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. In this invention, the server includes a conversational AI means for providing a recipe based on a user's input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, and a means for proposing the next recipe based on the user's feedback. This makes it possible to propose recipes that reflect the user's preferences and feedback.

[1862] "User input" means information or instructions provided by a user to a system.

[1863] A "recipe" is a list of ingredients and steps for making a particular dish.

[1864] "Conversational AI" is an artificial intelligence system that uses natural language processing technology to interact with users and provide them with information.

[1865] A "cooking procedure" is the order or method of specific steps required to prepare a dish.

[1866] "Feedback" refers to thoughts and suggestions for improvement provided by users regarding the system.

[1867] "Means of learning" refers to the techniques and methods by which the system analyzes user feedback and incorporates it into its next suggestions.

[1868] "Recipe suggestions" refers to suggestions of cooking ingredients and steps provided to the user by the system.

[1869] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing the cooking steps of the recipe through conversation with the user, a means for receiving and learning from feedback from the user, and a means for suggesting the next recipe based on the user's feedback.

[1870] System Configuration

[1871] The system uses the following hardware and software:

[1872] Hardware: Smartphone

[1873] Software: Python, scikit-learn

[1874] Program Processing

[1875] The server receives input from the user and generates a recipe based on it. It instructs the user on the cooking steps through dialogue. After the user finishes cooking, they can provide feedback. This feedback is collected by the server and saved as learning data. The next time a recipe is proposed, a recipe tailored to the user's preferences will be suggested based on this feedback.

[1876] Examples

[1877] For example, if a user provides feedback such as "I want it less spicy," the next recipe will be suggested with less spiciness. This makes it possible to suggest recipes that are tailored to the user's preferences.

[1878] Examples of prompt statements

[1879] Below are examples of prompts to be input to a generative AI model.

[1880] User ID: user123

[1881] Feedback: I would like it to be less spicy

[1882] Next suggestion: A less spicy recipe

[1883] In this way, recipe suggestions that reflect user feedback are realized.

[1884] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1885] Step 1:

[1886] A user accesses the system using a smartphone and requests a recipe. The user's input (request content) is sent to the server. The server receives this input and starts processing it as initial data for generating a recipe.

[1887] Step 2:

[1888] The server uses conversational AI to generate specific recipes through dialogue with the user. The conversational AI uses natural language processing technology to analyze the user's request and select an appropriate recipe. In this process, the user's preferences and past history are also taken into account. The input is the user's request, and the output is the generated recipe.

[1889] Step 3:

[1890] The server provides the generated recipe to the user and instructs the cooking procedure. The user cooks while checking the cooking procedure through their smartphone. The input is the generated recipe, and the output is the cooking procedure instruction to the user.

[1891] Step 4:

[1892] After the user finishes cooking, he / she provides feedback. The feedback is sent to the server via the smartphone. The input is the user's feedback, and the output is the feedback data stored in the server.

[1893] Step 5:

[1894] The server analyzes the received feedback and saves it as training data. Specifically, the feedback content is divided into features and evaluation, and training is performed using a linear regression model. The input is the feedback data, and the output is the updated training model.

[1895] Step 6:

[1896] The next time the user requests a recipe, the server uses the learned model to suggest recipes that match the user's preferences. The server takes into account the previous feedback and generates a recipe optimized for the user's preferences. The input is the user's new request and the learned model, and the output is the optimized recipe.

[1897] In this way, recipe suggestions that reflect user feedback are realized.

[1898] Example 3

[1899] 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".

[1900] Conventional conversational AI systems have had difficulty providing recipes that fully reflect the user's tastes and preferences. In addition, they lacked the means to effectively utilize the user's past conversation history, and the recipes they provide based on the user's preferences were limited. Furthermore, they lacked the ability to effectively learn from user feedback and evolve recipes.

[1901] 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.

[1902] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for storing the user's past conversation history about meals in a database, a means for analyzing the stored conversation history and learning the user's preferences, a means for using a generative AI model for generating recipes based on the user's preferences, and a means for providing the generated recipes to the user. This makes it possible to provide recipes that deeply correspond to the user's preferences and tastes.

[1903] A "conversational artificial intelligence means" is a technology that provides information through natural language dialogue with a user and generates responses according to the user's requests.

[1904] A "means for instructing cooking procedures" is a technology that provides users with step-by-step instructions on each step of a recipe, supporting them in the progress of cooking.

[1905] "Means of receiving feedback and learning from it" refers to technology that collects opinions and evaluations provided by users and uses them to improve the system's performance and content.

[1906] "Means for storing past conversation history in a database" refers to a technology that records the content of past conversations with a user and stores it in a database for future reference.

[1907] "Means for analyzing conversation history and learning user preferences" refers to technology that analyzes stored conversation history to identify user preferences and tastes and reflects them in the system.

[1908] "Means of using generative AI models" refers to technologies that utilize machine learning models to generate appropriate information and content based on user preferences and requests.

[1909] A "means for generating recipes" is a technology that automatically creates cooking recipes that meet a user's requests and preferences.

[1910] A "means for providing recipes" is a technology that presents the generated recipes to users and enables them to view and use them.

[1911] This invention relates to a conversational AI system that provides recipes based on user input. The system has the function of instructing the cooking steps of a recipe through conversation with the user, receiving feedback from the user, and learning from it. The system also stores the user's past conversation history about meals in a database, and analyzes the stored conversation history to learn the user's preferences. Furthermore, the system uses a generative AI model that generates recipes based on the user's preferences, and provides the generated recipes to the user.

[1912] Hardware and Software Used

[1913] Hardware:

[1914] Server: Manages the database, runs the generative AI models, and provides the interface with the user.

[1915] Terminal: A device such as a smartphone or PC used by a user.

[1916] software:

[1917] Database management system: Stores user conversation history and preference information.

[1918] Generative AI models: For example, using natural language processing models such as GPT-4.

[1919] Conversational artificial intelligence system: manages dialogue with the user, providing recipes and cooking instructions.

[1920] Program Processing

[1921] The server executes a conversational artificial intelligence means that provides recipes based on user input. When the user says "I like spicy food" through the terminal, the terminal transmits the statement to the server. The server stores the statement in a database and analyzes it using a generative AI model. As a result of the analysis, it is determined that the user likes spicy food.

[1922] Later, when the user requests, "What recipes do you recommend for dinner tonight?", the server inputs the following prompt into the generative AI model:

[1923] Example prompt:

[1924] You know from the user's past conversation history that they like spicy food, so provide them with some recommended spicy food recipes.

[1925] The generative AI model generates a recipe that matches the user's preferences based on this prompt. For example, a recipe for spicy chicken curry is generated. The server sends the generated recipe to the terminal, and the terminal displays to the user, "Here is the recipe for spicy chicken curry."

[1926] In this way, a system is realized in which the server, the terminal, and the user work together to provide recipes that match the user's preferences. The flow of the specification process in the third embodiment will be described with reference to FIG.

[1927] Step 1:

[1928] A user starts a dialogue with the conversational AI system through a terminal. For example, the user says, "I like spicy food."

[1929] Input: What the user says

[1930] Output: User utterance data

[1931] Step 2:

[1932] The terminal transmits the user's utterance data to the server, which then stores the received utterance data in a database.

[1933] Input: User utterance data

[1934] Output: Speech data stored in a database

[1935] Step 3:

[1936] The server periodically analyzes the conversation history in the database and uses a generative AI model to extract keywords and phrases from users' utterances using natural language processing techniques.

[1937] Input: Conversation history in a database

[1938] Output: Extracted keywords and phrases

[1939] Step 4:

[1940] The server learns the user's preferences based on the extracted keywords and phrases. For example, based on the information "I like spicy food," the server recognizes that the user likes spicy food. This information is stored in a database as a user profile.

[1941] Input: Extracted keywords or phrases

[1942] Output: Updated user profile

[1943] Step 5:

[1944] The user requests a new recipe through the terminal, for example, by saying, "Tell me a recommended recipe for tonight's dinner."

[1945] Input: User request

[1946] Output: User request data

[1947] Step 6:

[1948] The server receives the user's request data and inputs a prompt sentence into the generative AI model. The prompt sentence includes the user's preference information. For example, the prompt sentence might say, "We know from the user's past conversation history that the user likes spicy food. Please provide some recommended spicy food recipes for the user."

[1949] Input: User request data, user preferences

[1950] Output: Prompt statement

[1951] Step 7:

[1952] The generative AI model uses the prompt text to generate recipes that match the user's preferences, for example a recipe for spicy chicken curry.

[1953] Input: prompt statement

[1954] Output: The generated recipe

[1955] Step 8:

[1956] The server transmits the generated recipe to the terminal. The terminal displays the recipe to the user. For example, the recipe is provided in the form of "Here is the recipe for spicy chicken curry."

[1957] Input: The generated recipe

[1958] Output: The recipe displayed to the user

[1959] In this way, a system is realized in which the server, terminal, and user work together to provide recipes that suit the user's preferences.

[1960] (Application example 3)

[1961] 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".

[1962] Conventional recipe provision systems have difficulty in providing recipes that fully reflect the user's tastes and preferences, and therefore have been unable to increase user satisfaction. In addition, they lacked the functionality to suggest restaurants and menus that the user would like, and therefore were unable to comprehensively support the user's dining experience.

[1963] 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.

[1964] In this invention, the server includes a conversational AI means for providing recipes based on user input, a means for instructing cooking steps of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referring to the user's past conversation history about meals and providing recipes based on the history, and a means for suggesting restaurants based on the user's preferences. This makes it possible to provide recipes and suggest restaurants that reflect the user's preferences and tastes.

[1965] "User input" means information or instructions provided by a user to a system.

[1966] A "conversational AI that provides recipes" is an artificial intelligence that suggests cooking recipes through dialogue with the user.

[1967] A "means for instructing the cooking steps of a recipe" refers to a function that provides specific guidance to the user on the cooking steps.

[1968] "Means of receiving and learning from user feedback" refers to the ability of the system to collect opinions and evaluations provided by users and make improvements based on them.

[1969] "A means of referencing the user's past conversation history about food and providing recipes based on that" refers to a function that uses records of the user's past conversations to suggest recipes that suit the user's preferences.

[1970] "A means of suggesting restaurants based on user preferences" refers to a function that recommends suitable restaurants based on the user's preferences and past conversation history.

[1971] A "generative AI model" is a model that uses artificial intelligence to generate new information and patterns from data.

[1972] A "prompt sentence" is an instruction sentence entered into an artificial intelligence to elicit a specific response.

[1973] A system for implementing this invention includes a conversational AI that provides recipes based on user input, a means for instructing cooking steps for the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, a means for referencing the user's past conversation history about meals and providing recipes based thereon, and a means for suggesting restaurants based on the user's preferences.

[1974] System Configuration

[1975] The server uses the following hardware and software:

[1976] Hardware: powerful processor, sufficient memory, database server

[1977] Software: conversational AI engine (e.g. OpenAI's GPT-3), database management system (e.g. SQLite), user interface (e.g. smartphone app)

[1978] Data processing and calculation

[1979] The server processes the data in the following steps:

[1980] 1. Receive user input: Receive information entered by the user through the smartphone app.

[1981] 2. Recipe provision by conversational AI: A conversational AI engine generates appropriate recipes based on user input.

[1982] 3. Cooking Instructions: Provide the user with step-by-step instructions for cooking a recipe.

[1983] 4. Collect feedback and learn: Collect feedback from users and let the AI ​​engine learn from that information.

[1984] 5. Referencing past conversation history: Retrieve the user’s past conversation history from the database and provide recipes based on the user’s preferences.

[1985] 6. Restaurant Suggestions: Suggest suitable restaurants based on user preferences.

[1986] Examples

[1987] For example, if a user has previously mentioned that they like spicy food, the system will prioritize spicy recipes based on that information and also suggest restaurants that serve spicy food.

[1988] Example prompt:

[1989] User conversation history: "I like spicy food. I especially like curry and mapo tofu."

[1990] Extract user preferences.

[1991] Using this prompt, a generative AI model extracts the user's preferences and uses the results to provide recipes and restaurant suggestions.

[1992] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1993] Step 1:

[1994] Users input meal information through a smartphone app.

[1995] Input: User's dietary information (e.g. favorite dishes, allergies, etc.)

[1996] Output: The user's input information is sent to the server.

[1997] Specific behavior: A user enters information into an input form in the app and presses the submit button.

[1998] Step 2:

[1999] The server receives the user's input information and sends it to the conversational AI engine.

[2000] Input: User input information

[2001] Output: User information sent to the conversational AI engine

[2002] Specific operation: The server analyzes the user information it receives, converts it into an appropriate format, and sends it to the AI ​​engine.

[2003] Step 3:

[2004] A conversational AI engine generates recipes based on user input.

[2005] Input: User input information

[2006] Output: The generated recipe

[2007] What it does: The AI ​​engine generates an appropriate recipe taking into account the user's preferences and constraints.

[2008] Step 4:

[2009] The server sends the generated recipe to the user.

[2010] Input: The generated recipe

[2011] Output: A recipe displayed on the user's smartphone app

[2012] Specific operation: The server sends the generated recipe to the user's smartphone app and displays it.

[2013] Step 5:

[2014] The user follows the recipe, cooks the dish and then provides feedback in the app.

[2015] Input: User feedback (e.g. recipe rating, suggestions for improvement, etc.)

[2016] Output: The feedback information is sent to the server.

[2017] Specific action: After cooking, the user enters information into the app's feedback form and presses the submit button.

[2018] Step 6:

[2019] The server receives user feedback and sends it to the AI ​​engine for learning.

[2020] Input: User feedback information

[2021] Output: The trained AI model

[2022] Specific operation: The server analyzes the received feedback information and sends it to the AI ​​engine to update the model.

[2023] Step 7:

[2024] The server retrieves the user's past conversation history from the database and sends it to the AI ​​engine.

[2025] Input: User's past conversation history

[2026] Output: Conversation history sent to the AI ​​engine

[2027] Specific operation: The server retrieves the user's past conversation history from the database and sends it to the AI ​​engine.

[2028] Step 8:

[2029] The AI ​​engine extracts the user's preferences based on past conversation history and provides recipes.

[2030] Input: User's past conversation history

[2031] Output: A recipe based on the user's preferences.

[2032] How it works: The AI ​​engine analyzes conversation history, extracts user preferences, and generates appropriate recipes.

[2033] Step 9:

[2034] The server suggests restaurants based on the user's preferences.

[2035] Input: User preferences

[2036] Output: Suggested restaurant information

[2037] Specific operation: The server searches the database for suitable restaurants based on the user's preferences and makes suggestions.

[2038] 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.

[2039] "Example 1"

[2040] One embodiment of the present invention is a system that provides a conversational AI that includes an emotion engine that recognizes the emotions of a user. The system provides a recipe based on the user's input and instructs the user on how to cook the recipe through a conversation. The system also receives feedback from the user and learns from it. Specifically, if the user inputs, "I'm feeling sad today, so I want to eat something that will cheer me up," the emotion engine recognizes the user's "sad" mood and the AI ​​provides a recipe that is said to be uplifting.

[2041] "Example 2"

[2042] In addition, the emotion engine adjusts the recipe offerings based on the user's emotions. For example, if a user feels "very tired today," the emotion engine will recognize that information and offer recipes that are easy to make even for tired users. This makes it possible to offer optimal recipes according to the user's emotional state.

[2043] "Example 3"

[2044] The emotion engine also adjusts recipe cooking instructions based on the user's emotions. For example, if the emotion engine recognizes that the user is in a hurry, the AI ​​will provide concise and easy-to-understand cooking instructions. This makes it possible to provide optimal cooking instructions according to the user's emotional state.

[2045] The process flow of each embodiment will be described below.

[2046] "Example 1"

[2047] Step 1: The user gives input to the conversational AI. For example, the user might say, "I'm feeling sad today, so I want to eat something that will cheer me up."

[2048] Step 2: The emotion engine recognizes emotions from the user's input. In this example, the "sad" mood is recognized.

[2049] Step 3: The AI ​​provides a recipe based on the output of the emotion engine. In this example, it provides a recipe that is said to be uplifting.

[2050] "Example 2"

[2051] Step 1: A user gives input to the conversational AI. For example, the user might say, "I'm very tired today."

[2052] Step 2: The emotion engine recognizes the emotion from the user's input. In this example, the state "tired" is recognized.

[2053] Step 3: The AI ​​provides a recipe based on the output of the emotion engine. In this example, it provides a recipe that is easy to make even for tired users.

[2054] "Example 3"

[2055] Step 1: The user gives input to the conversational AI. For example, the user might say, "I'm anxious."

[2056] Step 2: The emotion engine recognizes emotions from the user's input. In this example, the state of "anxious" is recognized.

[2057] Step 3: The AI ​​gives instructions for cooking based on the output of the emotion engine. In this example, the instructions are given succinctly and clearly.

[2058] Example 1

[2059] 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".

[2060] Conventional conversational AI systems were able to provide recipes based on user input, but they lacked the ability to suggest recipes that took the user's emotions into account, or to learn from the user's feedback and reflect it in future suggestions. This made it difficult to provide optimal recipes based on the user's mood and preferences. In addition, they did not provide sufficient real-time support to users during the cooking process.

[2061] 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.

[2062] In this invention, the server includes a conversational artificial intelligence means for providing a recipe based on a user's input, a means for instructing the cooking procedure of the recipe through conversation with the user, a means for receiving feedback from the user and learning from it, an emotion recognition means for recognizing the emotion of the user, and a means for providing a recipe based on the emotion of the user. This makes it possible to provide an optimal recipe according to the user's mood and preferences.

[2063] A "conversational artificial intelligence means" is an artificial intelligence system that provides information through a dialogue with a user and generates appropriate responses based on the user's input.

[2064] The "means for instructing cooking steps" is a function that explains the specific cooking steps of a recipe to the user and sequentially instructs the user on the necessary steps.

[2065] "Means of learning feedback" refers to a function that collects evaluations and opinions from users and uses them to improve the system's performance and the content of its suggestions.

[2066] An "emotion recognition means" is a technology for analyzing emotions from a user's input or dialogue and recognizing those emotions.

[2067] A "means for providing recipes" is a function that generates appropriate cooking recipes based on the user's requests and feelings and provides them to the user.

[2068] This invention is a system that uses a smartphone application equipped with conversational artificial intelligence to provide optimal recipes based on the user's meal-related desires and emotions. The system analyzes the user's input, recognizes emotions, and generates recipes using a generative AI model. It also learns feedback from the user and reflects it in future suggestions.

[2069] Hardware and Software Used

[2070] Hardware

[2071] Device: Mobile devices such as smartphones and tablets

[2072] Server: Cloud or on-premise server

[2073] software

[2074] Natural Language Processing (NLP) technology: Google Cloud Natural Language API, IBM Watson Natural Language Understanding

[2075] Emotion recognition engine: Microsoft Azure Emotion API

[2076] Generative AI model: OpenAI's GPT-4

[2077] Machine learning algorithms: TensorFlow, PyTorch

[2078] Program Processing

[2079] User Input

[2080] The user starts the smartphone application and inputs their meal-related request into a text box, for example, "I'd like to eat Japanese food today" or "Tell me about some dishes that use chicken."

[2081] Analyzing data

[2082] The terminal sends the user's input to the server, which then analyzes the received text data using natural language processing technology. This analysis allows the server to understand the user's request.

[2083] Emotion recognition

[2084] Based on the analysis results, the server uses an emotion recognition engine to recognize the user's emotions. For example, if the user enters "I'm feeling sad today, so I want to eat something that cheers me up," the server recognizes the emotion "sad."

[2085] Recipe Suggestions

[2086] The server uses a generative AI model to generate recipes that match the user's preferences based on the analysis results and emotion recognition results. The generated recipes are sent to the device in the form of text, images, and videos.

[2087] User interaction

[2088] If a user has a question or difficulty while cooking, they can ask the AI ​​through the application. The device sends the question to the server, which then generates a response in real time and returns it to the device.

[2089] Learning Feedback

[2090] The user provides feedback on the provided recipe, and the device sends this feedback to the server, which then learns from the feedback using machine learning algorithms.

[2091] Examples

[2092] If a user types, "I want to eat Japanese food today," the device will send this request to the server. The server will analyze the request and suggest Japanese recipes using a generative AI model. For example, a recipe for "teriyaki chicken" will be provided in text, images, and video. If a user types, "I'm feeling sad today, so I want to eat something that will cheer me up," the server will recognize the emotion of "sadness" and suggest a recipe for "curry rice," which is said to be uplifting.

[2093] Examples of prompt statements

[2094] "A user says they want to eat Japanese food. Please suggest some Japanese recipes."

[2095] "A user says they're feeling sad. Can you suggest some cheer-up recipes?"

[2096] In this way, the system provides the optimal recipe based on the user's desires and feelings.

[2097] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2098] Step 1:

[2099] The user starts the smartphone application and inputs their meal-related requests into a text box. For example, they may make requests such as, "I'd like to eat Japanese food today," or "Tell me about some dishes that use chicken." The input data is stored in text format on the device.

[2100] Step 2:

[2101] The device sends the user's input data to the server. The server analyzes the received text data using natural la...

Claims

1. A conversational AI that provides recipes based on user input; A means for instructing the cooking procedure of the recipe through conversation with the user; means for receiving feedback from said user and learning said feedback; A means for suggesting dishes that match the user's preferences and placing a delivery order; A means for providing detailed information on the cooking process and ingredients of the proposed dish in the form of images and videos; A means for inquiring of a question from the user about the dish to the conversational AI; means for providing the user with the recipe updated by the conversational AI as an answer to the question; A system including:

2. A conversational AI that provides recipes based on user input; A means for instructing the cooking procedure of the recipe through conversation with the user; means for receiving feedback from said user and learning said feedback; an emotion engine that recognizes an emotional state of the user; means for providing the user with the recipe adjusted by the conversational AI in response to the emotional state of the user; A means for ordering the suggested dishes for delivery; A system including:

3. A conversational AI that provides recipes based on user input; A means for instructing the cooking procedure of the recipe through conversation with the user; means for receiving feedback from said user and learning said feedback; A means for referencing a past conversation history of the user regarding meals and providing a recipe based on the conversation history; an emotion engine that recognizes an emotional state of the user; means for suggesting restaurants based on the user's preferences and the emotional state; means for assisting the user in ordering food from the restaurant; A system including:

Images