System

Abstract

To provide a system.SOLUTION: A system includes: means for performing machine learning on information on incoming ingredients on a daily basis; means for predicting demand for ingredients, using a machine learning model trained by the means for performing machine learning; means for acquiring biological information of a user, and analyzing the emotions of the user using an emotion recognition engine based on the acquired biological information; and means for generating purchase suggestions based on the demand prediction, information indicating the analyzed emotions of the user, and at least one of information on the user's preferences and health condition.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 an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

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

[0004] Currently, there are no systems that utilize food market trends and information on ingredients available at supermarkets to recommend recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance. There are also no systems that provide a seamless service from demand forecasting to purchasing and delivery. Furthermore, there are no systems that contribute to reducing food waste or provide shopping support for elderly households and households with children. A system that can solve these issues is needed. [Means for solving the problem]

[0005] This invention provides a means to use AI to learn daily information on food market trends and ingredients available at supermarkets, and to recommend recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. It also provides a means to provide a seamless service from demand forecasting to purchasing and delivery. It also provides a means to contribute to reducing food waste and to provide shopping support to elderly households and households with children. The above-mentioned issues can be solved by combining these methods. [Brief explanation 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. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 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. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

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

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

[0009] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

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

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

[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. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[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 WAN (Wide Area Network) and / or a LAN (Local Area Network).

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

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

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0023] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

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

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

[0027] "Example 1"

[0028] A first embodiment of the present invention is a system that uses AI to learn daily information about food market trends and supermarket stock information. This system collects market trend data and supermarket stock information, and the AI ​​learns based on this information. Specifically, market trend data includes, for example, food price fluctuations, consumer purchasing habits, and seasonal influences. Furthermore, supermarket stock information includes, for example, the arrival date, stock volume, and sales status of each product. Based on this information, the AI ​​recommends recipes that take into account the customer's preferences, health condition, seasonal ingredients, and nutritional balance.

[0029] "Example 2"

[0030] A second embodiment of the present invention is a system that provides a seamless service from demand forecasting to purchasing and delivery. This system predicts the demand for ingredients based on recipes recommended by AI, and enables customers to purchase them. Specifically, the AI ​​creates a list of ingredients included in the recommended recipes, and suggests purchasing ingredients based on that list. The purchased ingredients are then delivered to the customer using a delivery service.

[0031] "Example 3"

[0032] A third embodiment of the present invention is a system that contributes to reducing food waste and provides shopping support to elderly households and households with children. This system makes suggestions to optimize food consumption and reduce food waste. Specifically, it suggests purchasing necessary ingredients based on recipes recommended by AI, encouraging the consumption of surplus ingredients. In addition, for elderly households and households with children, it makes purchasing suggestions that take into account their health status and living situation.

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

[0034] "Example 1"

[0035] Step 1: Collect market trend data and supermarket inventory information. This information includes, for example, food price fluctuations, consumer purchasing habits, seasonal influences, arrival dates, quantities, and sales status of each product.

[0036] Step 2: The AI ​​learns from the collected information. Based on market trends and arrival information, the AI ​​learns a model to recommend recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[0037] Step 3: Using the AI-trained model, the system recommends recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[0038] "Example 2"

[0039] Step 1: Based on the recipe recommended by AI, the demand for ingredients is predicted. Specifically, a list of ingredients included in the recipe is created and purchasing suggestions are made based on that list. Step 2: The ingredients that the customer has decided to purchase are delivered to the customer using a delivery service.

[0040] "Example 3"

[0041] Step 1: Based on the recipes recommended by AI, suggestions are made to optimize ingredient consumption and reduce food waste. Specifically, a list of ingredients included in the recipe is created, and based on that, suggestions are made to encourage the consumption of surplus ingredients.

[0042] Step 2: Make purchasing suggestions to elderly households and households with children, taking into account their health and living conditions.

[0043] Example 1

[0044] 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 device 14 will be referred to as a "terminal."

[0045] In today's grocery market, consumers demand ingredient selection and recipe suggestions that accommodate diverse tastes, health conditions, and seasonal changes. However, providing appropriate recipes that meet these needs is not easy. Reducing food waste and providing shopping support for elderly households and families with children are also important issues. Furthermore, there is a need to build a comprehensive platform that delivers information based on customer attributes. To solve these issues, an efficient and effective system is required.

[0046] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0047] In this invention, the server includes: a means for performing machine learning on daily food market trends and retail store inventory information; a means for storing the collected data in a database; a means for preprocessing the stored data; a means for inputting the preprocessed data into a machine learning model; a means for generating recipes using the machine learning model that take into account customer preferences, health status, seasonal ingredients, and nutritional balance; a means for providing the generated recipes to a user terminal; and a means for collecting user feedback and using it to improve the system. This enables the provision of recipes that meet diverse consumer needs, reducing food waste and providing shopping support to elderly households and households with children. It also enables the creation of a comprehensive platform that distributes information based on customer attributes.

[0048] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[0049] "Retail store stock information" is information about product stock at retail stores, such as the arrival date, stock volume, and sales status of each product.

[0050] "Machine learning" is a technology that allows computers to learn patterns based on data and make predictions and classifications.

[0051] A "database" is a system for efficiently storing, managing, and retrieving data.

[0052] "Preprocessing" refers to the process of preparing data before inputting it into a machine learning model, such as cleaning and normalizing the data.

[0053] A "machine learning model" is an algorithm or mathematical model that learns from data and makes predictions or classifications.

[0054] A "recipe" is a set of instructions that shows how to prepare a dish and how to combine ingredients.

[0055] A "user terminal" is a device that allows a user to receive information and perform operations.

[0056] "Feedback" is information that is useful for improving the system, such as user ratings and comments.

[0057] "Food waste reduction" is an initiative aimed at reducing food waste and using it efficiently.

[0058] "Shopping support" is a service that provides shopping support to consumers with specific needs, such as elderly households and households with children.

[0059] "Customer attributes" refers to characteristics and information about customers, such as family structure, age, and preferences.

[0060] An "integrated platform" is a system that integrates multiple functions and services to provide users with a consistent experience.

[0061] This invention is a system that collects food market trends and retail store arrival information daily, and uses machine learning to generate recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance. Specific embodiments of this system are described below.

[0062] Server Processing

[0063] 1. Data Collection

[0064] The server collects food market trend data and retail store arrival information every day at 9:00 a.m. Market trend data includes food price fluctuations, consumer purchasing trends, seasonal influences, etc. This data is obtained using the API of the food market information service.

[0065] The retail store's stock information includes the arrival date, stock quantity, sales status, etc. of each product. This data is obtained from the retail store's inventory management system.

[0066] 2. Data storage

[0067] The server stores the collected data in a MySQL® database, which stores data such as food price fluctuation data, consumer purchasing trends, and retail store inventory information as tables.

[0068] 3. Data Preprocessing

[0069] The server preprocesses the stored data, including cleaning the data (filling in missing values ​​and removing outliers) and normalizing it. The server uses the Python (registered trademark) pandas library to clean and normalize the data.

[0070] 4. Inputting data into the AI ​​model

[0071] The server inputs the preprocessed data into an AI model built with TENSORFLOW®, which learns from food price fluctuations, consumer purchasing trends, and retail store inventory information.

[0072] 5. Recipe Generation

[0073] The server uses an AI model to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. The generated recipes are then stored in a database.

[0074] 6. Recipe provision

[0075] The server sends the generated recipe to the terminal via the API.

[0076] Terminal handling

[0077] 1. Receiving recipes

[0078] The terminal receives the recipe information sent from the server and stores the received recipe information in a local database of the terminal.

[0079] 2. Recipe display

[0080] The device then displays the received recipe information to the user. Front-end frameworks such as React and Vue.js are used for displaying the information, and detailed recipe information (ingredients, cooking instructions, nutritional information, etc.) is displayed in an easy-to-read format.

[0081] 3. Feedback Collection

[0082] The device collects feedback from users, including recipe ratings and comments, and stores the feedback in a local database on the device.

[0083] 4. Send Feedback

[0084] The terminal sends the collected feedback to the server.

[0085] User operations

[0086] 1. View recipes

[0087] The user browses the provided recipes through the device, checks the detailed information of the recipes, and selects the recipes they want to try.

[0088] 2. Recipe Trial

[0089] The user then tries out the selected recipe and cooks the dish according to the recipe.

[0090] 3. Feedback input

[0091] Users can input their ratings and comments on the recipes into the device, including feedback on the cooking results, taste, ease of use, etc.

[0092] 4. Send Feedback

[0093] The user sends the input feedback to the server through the terminal.

[0094] Specific examples

[0095] Example 1: Collecting market trend data

[0096] The server uses the API of a food market information service to obtain food price fluctuation data at 9:00 AM every day. For example, it sends the following API request:

[0097] GET / api / market_trends?date=2023-10-01

[0098] Example 2: Data input to an AI model

[0099] The server inputs the preprocessed data into an AI model built with TensorFlow, for example using a prompt like this:

[0100] Prompt: Based on food price fluctuation data and retail store stock information, predict food prices for the next week.

[0101] Example 3: Creating and serving recipes

[0102] The server uses the AI ​​model to generate recipes, which are then sent to the device. For example, the following recipes are provided:

[0103] Recipe name: Healthy salad made with seasonal vegetables

[0104] Ingredients: Tomato, cucumber, lettuce, olive oil, salt

[0105] Directions: 1. Thinly slice the tomato and cucumber. 2. Tear the lettuce into bite-sized pieces. 3. Place all ingredients in a bowl and toss with olive oil and salt.

[0106] This way, users receive fresh, healthy recipes every day.

[0107] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0108] Step 1: Data collection

[0109] The server collects food market trend data and retail store arrival information every day at 9:00 AM. Market trend data includes food price fluctuations, consumer purchasing trends, seasonal effects, etc. This data is obtained using the API of the food market information service. Retail store arrival information includes the arrival date, arrival quantity, sales status, etc. of each product. This data is obtained from the retail store's inventory management system. The input is the API request, and the output is the collected market trend data and arrival information.

[0110] Step 2: Save data

[0111] The server stores the collected data in a MySQL database. The database stores data such as food price fluctuation data, consumer purchasing trends, and retail store inventory information as tables. The input is the collected data, and the output is the data stored in the database.

[0112] Step 3: Data Preprocessing

[0113] The server preprocesses the stored data. Preprocessing includes data cleaning (filling in missing values ​​and removing outliers) and normalization. The Python pandas library is used to clean and normalize the data. The input is the raw data stored in the database, and the output is the preprocessed data.

[0114] Step 4: Input data into the AI ​​model

[0115] The server inputs the preprocessed data into an AI model built with TensorFlow. The AI ​​model learns based on food price fluctuations, consumer purchasing trends, and retail store inventory information. The input is the preprocessed data, and the output is the learning results.

[0116] Step 5: Recipe Generation

[0117] The server uses an AI model to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. The generated recipes are stored in a database. The input is the learning results of the AI ​​model, and the output is the generated recipe.

[0118] Step 6: Provide the recipe

[0119] The server sends the generated recipe to the terminal via API. The input is the generated recipe, and the output is the recipe information sent to the terminal.

[0120] Step 7: Receiving the recipe

[0121] The terminal receives recipe information sent from the server. The received recipe information is stored in the terminal's local database. The input is the recipe information sent from the server, and the output is the recipe information stored in the local database.

[0122] Step 8: View the recipe

[0123] The device displays the received recipe information to the user. Front-end frameworks such as React and Vue.js are used for display. Detailed information about the recipe (ingredients, cooking instructions, nutritional information, etc.) is displayed in an easy-to-read format. The input is the recipe information stored in the local database, and the output is the recipe information displayed to the user.

[0124] Step 9: Gather feedback

[0125] The device collects feedback from users. The feedback includes recipe ratings and comments. The collected feedback is stored in a local database on the device. The input is the feedback from the user, and the output is the feedback stored in the local database.

[0126] Step 10: Send feedback

[0127] The terminal sends the collected feedback to the server, where the input is the feedback stored in the local database and the output is the feedback sent to the server.

[0128] Step 11: Use feedback

[0129] The server receives feedback sent by users and uses it to improve the system. The input is the feedback sent by users, and the output is the feedback used to improve the system.

[0130] (Application example 1)

[0131] Next, a description will be given of Application Example 1 of Embodiment 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."

[0132] Modern consumers have diverse needs when it comes to purchasing groceries. In particular, they want recipe suggestions based on their health status and preferences, meal suggestions using seasonal ingredients, and efficient purchasing that takes into account market trends and supermarket stock information. However, there is no consistent system in place to meet these needs. Reducing food waste and providing shopping support for elderly households and households with children are also important issues.

[0133] 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.

[0134] In this invention, the server includes: means for daily AI learning of food market trends and supermarket inventory information; means for recommending recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance; means for providing an end-to-end process from demand forecasting and AI recommendations to purchase; means for recommending recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance through an application installed on a smartphone; and means for proposing recipes that suit the user's wishes using a generative AI model. This meets the diverse needs of consumers, enabling efficient grocery shopping and healthy eating.

[0135] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[0136] "Information on ingredients arriving at supermarkets" refers to information on the arrival of ingredients at supermarkets, such as the arrival date, amount of each product, and sales status.

[0137] "AI learning" is the learning process that enables artificial intelligence to recognize patterns based on large amounts of data and make predictions and judgments.

[0138] "Customer preferences" refers to information about the types of ingredients and dishes that a particular customer prefers.

[0139] "Health conditions" refers to information about the health of a particular customer, including, for example, allergies and restrictions on the intake of certain nutrients.

[0140] "Seasonal ingredients" are ingredients that are considered to be most delicious and nutritious during a particular season.

[0141] "Nutritional balance" refers to a diet in which the nutrients necessary to maintain health are properly distributed.

[0142] "Means for recommending recipes" refers to a method of suggesting optimal cooking recipes that take into consideration the customer's preferences, health condition, seasonal ingredients, and nutritional balance.

[0143] "Demand forecasting" is an analytical method for predicting future demand for a product.

[0144] "A means of providing a seamless service" is a method of consistently providing a series of processes from demand forecasting to recommendations and purchase.

[0145] An "application installed on a smartphone" is a software program that runs on a smartphone.

[0146] A "generative AI model" is a model that allows artificial intelligence to generate new information and suggestions based on data.

[0147] A "prompt" is an instruction given to a generative AI model, serving as a guideline for generating specific information or suggestions.

[0148] The following system configuration will be described as an embodiment of the present invention.

[0149] System Configuration

[0150] The system includes a server, the user's smartphone, and a generative AI model. The server collects daily information on food market trends and ingredients arriving at supermarkets, and uses this information for AI learning. A dedicated application is installed on the user's smartphone, which recommends recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance.

[0151] Program processing

[0152] The server processes the data using the following means:

[0153] 1. Data collection methods:

[0154] The server collects daily food market trend data (price fluctuations, consumer purchasing habits, seasonal influences, etc.) and supermarket inventory information (arrival date, amount, sales status, etc.). This data is obtained using APIs and web scraping technology.

[0155] 2. AI learning methods:

[0156] The server performs AI learning based on the collected data. Specifically, it uses Python machine learning libraries (e.g., TensorFlow, PyTorch) to build food demand forecasting models and recipe recommendation models.

[0157] 3. Recipe recommendation methods:

[0158] The server recommends optimal recipes taking into consideration the user's preferences, health status, seasonal ingredients, and nutritional balance. Recipe information is provided through an application installed on the user's smartphone.

[0159] 4. Generative AI model means:

[0160] The server uses a generative AI model to suggest recipes that match the user's preferences, and the generative AI model generates new recipes based on the user's input (prompt sentence).

[0161] Hardware and software used

[0162] Hardware: Servers, smartphones

[0163] Software: Python, TensorFlow, PyTorch, dedicated applications

[0164] Specific examples

[0165] For example, if a user wants "low-carb recipes," they can input the following prompt into the generative AI model:

[0166] Example prompt sentence:

[0167] "The user wants low-carb recipes. Please recommend the best recipes based on market trend data and supermarket stock information."

[0168] By inputting this prompt into a generative AI model, the AI ​​will suggest recipes that match the user's preferences. The recipe information is provided through an application installed on the user's smartphone, and the user can purchase ingredients and cook based on the recipe.

[0169] This will meet the diverse needs of consumers, enable efficient grocery shopping, and encourage healthy eating habits.

[0170] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0171] Step 1:

[0172] The server collects food market trend data and supermarket stock information daily. Specifically, it uses API and web scraping technology to obtain data such as price fluctuations, consumer purchasing trends, seasonal influences, stock arrival dates, stock quantities, and sales status. The input is market trend data and stock arrival information, and the output is the collected data.

[0173] Step 2:

[0174] The server performs AI training based on the collected data. It uses Python machine learning libraries (e.g., TensorFlow, PyTorch) to build food demand forecasting models and recipe recommendation models. The input is the collected market trend data and stock information, and the output is a trained AI model.

[0175] Step 3:

[0176] Users input their preferences, health status, and desired recipe conditions through a dedicated application installed on their smartphone. The input is the user's preferences, health status, and desired recipe conditions, and the output is this information.

[0177] Step 4:

[0178] The server receives user input and uses an AI model to recommend optimal recipes. Specifically, it filters recipes and selects the most suitable one, taking into consideration the user's preferences, health status, seasonal ingredients, and nutritional balance. The input is the user's preferences, health status, and desired recipe conditions, and the output is the recommended recipe.

[0179] Step 5:

[0180] The server uses a generative AI model to generate a new recipe that matches the user's preferences. The prompt sentence is input into the generative AI model, which generates a recipe based on the user's preferences. The input is the prompt sentence, and the output is the generated recipe.

[0181] Step 6:

[0182] The server sends the recommended recipes and generated recipes to the user's smartphone. The user can review and select the suggested recipes through the smartphone application. The input is the recommended recipes and generated recipes, and the output is the recipe displayed on the user's smartphone.

[0183] Step 7:

[0184] The user purchases ingredients and cooks based on the selected recipe. A shopping list and cooking instructions are provided through a smartphone application. The input is the selected recipe, and the output is the shopping list and cooking instructions.

[0185] This will meet the diverse needs of consumers, enable efficient grocery shopping, and encourage healthy eating habits.

[0186] Example 2

[0187] Next, a description will be given of 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."

[0188] Conventional food purchasing systems have problems such as difficulty in proposing recipes based on customer preferences and health status, and the entire process from demand forecasting and purchasing to delivery is not carried out efficiently.Furthermore, there are issues such as the reduction of food waste and the lack of shopping support for elderly households and households with children.

[0189] 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.

[0190] In this invention, the server includes a means for recommending recipes using a generative AI model, a means for creating a list of ingredients based on the recommended recipe, a means for forecasting demand based on the created list of ingredients, a means for making purchase suggestions to customers based on the results of the demand forecast, and a means for arranging delivery of ingredients that customers have decided to purchase. This allows for recipe suggestions based on customers' preferences and health status, and efficiently executes the entire process from demand forecasting to purchase and delivery. It also helps reduce food waste and provides shopping support for elderly households and households with children.

[0191] A "generative AI model" is an algorithm that uses artificial intelligence to generate optimal answers or suggestions for a specific task.

[0192] "Recipe recommendation means" is a feature that uses a generative AI model to suggest cooking recipes based on the user's preferences and health status.

[0193] The "means for creating an ingredient list" is a function for creating a list of ingredients required for a recommended recipe.

[0194] "Means for demand forecasting" is a function that predicts future demand for food ingredients based on past data and current market trends.

[0195] The "means for making purchasing suggestions" is a function that suggests appropriate food ingredients to customers to purchase based on the results of demand forecasting.

[0196] The "means for arranging delivery" is a function for carrying out procedures for delivering ingredients that a customer has decided to purchase to a specified location.

[0197] "Food waste contribution" is an initiative that aims to reduce food waste and consume it efficiently.

[0198] "Shopping support for elderly households and households with children" is a service that helps elderly people and households with children to make shopping easier.

[0199] "Customer attributes" refers to personal information such as a customer's family structure, age, and preferences.

[0200] "Means of creating a comprehensive platform" refers to the ability to integrate multiple services and functions and provide them on a single platform.

[0201] This invention is a system that uses a generative AI model to recommend recipes and provides a seamless service from demand forecasting to purchasing and delivery. Specific embodiments of this system are described below.

[0202] Server Processing

[0203] The server uses a generative AI model to recommend recipes. Specifically, it uses Python to send prompts to a generative AI model (e.g., GPT-4 (registered trademark)) to generate recipes. The generated recipes are returned to the server in JSON format.

[0204] The server then creates an ingredient list based on the recommended recipe by running an SQL query to retrieve information about the ingredients in the recipe from a database and generating the ingredient list.

[0205] The server then performs demand forecasting based on the created ingredient list. It uses a machine learning model (e.g., a time series forecasting model) to predict the demand for ingredients and stores the prediction results in a database.

[0206] Furthermore, the server makes purchase proposals to customers based on the results of the demand forecast, generating the contents of the purchase proposals and sending them to the terminal via the REST API.

[0207] Finally, the server arranges for the delivery of the ingredients that the customer has decided to purchase by calling the API of the delivery service, arranging the delivery, and sending the delivery information to the terminal.

[0208] Terminal handling

[0209] The device receives the recommended recipes and ingredient lists sent from the server. When the user decides to purchase something through the device, the device sends that information to the server. The device uses an iOS or ANDROID (registered trademark) application to display the recommended recipes and ingredient lists to the user. The device also learns the user's purchase history and preferences, collecting data to reflect in future recommendations.

[0210] User operations

[0211] The user checks the recommended recipes through the device and purchases the necessary ingredients. Once the user decides to purchase, the device sends the information to the server, and the ingredients are delivered to the user using a delivery service. The user can manage their purchase history and preferences through the application and reflect them in the next recommendation.

[0212] Specific examples

[0213] An example of a prompt to be input to the generative AI model is, "Please suggest recipes for healthy meals. Preferably with ingredients that are easy to obtain and easy to prepare." Based on this prompt, the generative AI model recommends recipes such as "grilled chicken salad" and "soup with lots of vegetables." The server then creates a list of ingredients needed based on these recipes and performs demand forecasting. Users can check these recipes through their devices and purchase the necessary ingredients. The purchased ingredients are then delivered to the user using a delivery service.

[0214] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0215] Step 1:

[0216] The server inputs a prompt into the generative AI model to recommend recipes. Specifically, it runs a Python script and sends the prompt, "Please suggest recipes for healthy meals. Preferably ones that use easily available ingredients and are easy to prepare," to the generative AI model. The generative AI model generates a recipe based on this prompt and returns it to the server in JSON format.

[0217] Input: prompt statement

[0218] Output: Recommended recipes (JSON format)

[0219] Specific operations: Execute Python script, send prompts to generative AI model, receive recipes

[0220] Step 2:

[0221] The server creates a list of ingredients based on the recommended recipe. The server executes an SQL query to retrieve information about the ingredients in the recipe from a database and generate the list of ingredients.

[0222] Input: Recommended recipes (JSON format)

[0223] Output: Ingredient list (JSON format)

[0224] Specific operations: Execute SQL queries, retrieve ingredient information from the database, and generate an ingredient list.

[0225] Step 3:

[0226] The server performs demand forecasting based on the created ingredient list. It uses a machine learning model (e.g., a time series forecasting model) to predict the demand for ingredients and stores the prediction results in a database.

[0227] Input: Ingredient list (JSON format)

[0228] Output: Demand forecast results (saved in database)

[0229] Specific operations: Execution of machine learning model, demand forecast, saving of forecast results to database

[0230] Step 4:

[0231] The server makes purchase proposals to customers based on the results of the demand forecast, generates the contents of the purchase proposals, and sends them to the terminal via the REST API.

[0232] Input: Demand forecast results

[0233] Output: Purchase offer (sent to device)

[0234] Specific operation: Generates purchase proposals and sends them to the device via REST API

[0235] Step 5:

[0236] The user checks the purchase suggestions through the terminal, selects the ingredients they need, and decides to purchase them. The terminal then sends this information to the server.

[0237] Input: Purchase Proposal

[0238] Output: Purchase decision information (sent to server)

[0239] Specific operations: Confirm purchase offer on device, tap purchase button, send purchase information to server

[0240] Step 6:

[0241] The server receives the user's purchase information and arranges for delivery by calling the delivery service's API, arranging delivery, and sending the delivery information to the terminal.

[0242] Input: Purchase decision information

[0243] Output: Delivery information (sent to terminal)

[0244] Specific operations: Calling the delivery service API, arranging delivery, sending delivery information to the terminal

[0245] (Application example 2)

[0246] 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."

[0247] Modern consumers spend a lot of time in their busy daily lives purchasing ingredients and selecting recipes. Food waste and food loss have become social issues, placing a heavy burden on elderly households and those with children. Furthermore, it is difficult to provide meals tailored to individual tastes and health conditions. To solve these problems, an efficient and personalized food purchasing and delivery system is needed.

[0248] 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.

[0249] In this invention, the server includes a means for performing machine learning on a daily basis to learn about food market trends and ingredients available at retailers, a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing a seamless service from demand forecasting and machine learning recommendations to purchasing, and a means for creating recipe-based ingredient lists and seamless purchasing and delivery via an application installed on a smartphone. This allows consumers to efficiently purchase ingredients and easily prepare meals tailored to their individual needs. It also reduces food waste and provides shopping support for elderly households and those with children.

[0250] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[0251] "Retail store" means a store that sells goods directly to consumers.

[0252] "Information on newly arrived ingredients" refers to information on the types and quantities of ingredients newly arrived at retail stores and markets.

[0253] "Daily" means something that happens every day.

[0254] "Machine learning" refers to a technology that allows computers to analyze data and learn patterns and rules.

[0255] "Customer preferences" refers to the types of ingredients and dishes that customers prefer.

[0256] "Health Status" refers to the state of a Customer's physical health.

[0257] "Seasonal ingredients" refer to ingredients that are at their most delicious and nutritious during that season.

[0258] "Nutritional balance" refers to the balance of nutrients contained in a meal.

[0259] A "recipe" refers to how to prepare a dish or how to combine ingredients.

[0260] "Recommendation" refers to the recommendation of a particular option over other options.

[0261] "Demand forecasting" refers to predicting future demand.

[0262] "Through" means to do something consistently from start to finish.

[0263] A "smartphone" refers to a mobile phone that can connect to the Internet and use applications.

[0264] "Application" refers to software that performs a specific function.

[0265] "Ingredient list" refers to a list of ingredients needed for a dish.

[0266] "Purchase" refers to the purchase of a product.

[0267] "Delivery" means the delivery of the goods to the designated location.

[0268] "Food waste" refers to ingredients and food that are thrown away even though they are still edible.

[0269] An "elderly household" refers to a household in which an elderly person primarily lives.

[0270] "Child-rearing households" refers to households raising children.

[0271] "Shopping assistance" refers to assistance with shopping.

[0272] "Customer attributes" refer to characteristics such as a customer's family structure, age, and preferences.

[0273] A "comprehensive platform" refers to a foundation that provides multiple functions and services in a single system.

[0274] A system for implementing this invention includes a server, a smartphone, and an internet connection. The server has a means for performing machine learning on a daily basis to learn food market trends and information on ingredients received at retail stores, etc. This allows the system to constantly monitor the latest market information and use it to forecast demand.

[0275] The server also has the means to recommend recipes that take into account the customer's preferences, health condition, seasonal ingredients, and nutritional balance, allowing it to provide the best recipes for each individual customer. Based on the recommended recipe, it creates a list of necessary ingredients and presents it to the customer through an application installed on their smartphone.

[0276] Customers can use a smartphone application to purchase ingredients based on the recommended recipes. The application has a means to complete the purchasing and delivery procedures in one go, allowing customers to purchase the ingredients they need without any hassle and have them delivered to their home.

[0277] For example, if a user searches for "pasta recipes" on a smartphone app, the server uses AI to recommend "pasta with tomato sauce." This recipe requires "tomatoes, pasta, olive oil, and garlic." The application lists these ingredients and automatically places an order for delivery to the user's address.

[0278] An example of a prompt for the generative AI model is as follows:

[0279] User ID: 12345

[0280] Recipe search: Pasta

[0281] When this prompt sentence is input into the AI ​​recommendation model, the server will return the following response:

[0282] "Pasta with tomato sauce" requires "tomatoes, pasta, olive oil, and garlic." Based on this, the application handles the purchasing and delivery of ingredients in a single step.

[0283] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0284] Step 1:

[0285] The server collects daily information on food market trends and incoming ingredients from retailers and other sources, and inputs this information into a machine learning model. This allows for demand forecasts based on the latest market information. The inputs are market trend data and incoming ingredient information, and the output is demand forecast data. Specifically, the system obtains market data via an API and analyzes it using a machine learning algorithm.

[0286] Step 2:

[0287] The server inputs recipes based on the customer's preferences, health status, seasonal ingredients, and nutritional balance into a generative AI model to generate recommended recipes. The input is customer profile data and seasonal ingredient information, and the output is recommended recipes. Specifically, the server retrieves information from the customer database and inputs prompt statements into the generative AI model.

[0288] Step 3:

[0289] The server creates a list of ingredients based on the recommended recipe and sends it to the smartphone application. The input is the recommended recipe and the output is the ingredient list. Specifically, the server extracts ingredient information from the recipe data and converts it into a list format.

[0290] Step 4:

[0291] The user uses a smartphone application to check the ingredients based on the recommended recipe and then complete the purchase procedure. The input is the ingredient list, and the output is the purchase request. Specifically, the user displays the ingredient list through the application interface and presses the purchase button.

[0292] Step 5:

[0293] The server receives the user's purchase request and starts the delivery process. The input is the purchase request and the output is delivery instructions. Specifically, it calls the delivery service's API and sends delivery instructions.

[0294] Step 6:

[0295] The delivery service delivers ingredients to a specified address. The input is a delivery instruction, and the output is a delivery completion notification. Specifically, the service plans a delivery route and delivers the ingredients to the user's address.

[0296] Step 7:

[0297] The user receives the delivered ingredients and cooks the dish according to the recommended recipe. The input is the delivered ingredients, and the output is the completed dish. The specific operation is to cook the ingredients according to the recipe.

[0298] Example 3

[0299] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0300] Food waste and food loss are major issues in modern society. Elderly households and those with children face the challenge of finding and purchasing the right ingredients for their health and lifestyle. Furthermore, there is a need for a system that efficiently utilizes ingredients owned by users and provides nutritionally balanced meals.

[0301] The identification process by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes a means for performing machine learning on a daily basis of food market trends and information on ingredients received from retailers, etc.; a means for recommending recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance; a means for providing an integrated service from demand forecasting and machine learning recommendations to purchasing; a means for inputting ingredient information owned by the user; a means for saving the input ingredient information in a database; a means for generating recipes using a generative AI model based on the saved ingredient information; and a means for transmitting the generated recipes and necessary ingredient lists to the user. This enables users to consume ingredients efficiently and reduce food waste. Furthermore, suggestions tailored to the special needs of elderly households and households with children are provided, supporting healthy eating habits.

[0302] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[0303] "Retailer" refers to a store or supermarket that sells food products to consumers.

[0304] "Information on newly arrived ingredients" refers to information such as the type, quantity, and arrival date of ingredients newly received at the store.

[0305] "Machine learning" refers to a technology in which a computer automatically learns from data and makes predictions and classifications.

[0306] "User" refers to an individual or household who uses this system.

[0307] "Preferences" refers to the types of ingredients and dishes that a user prefers.

[0308] "Health status" refers to the physical health and nutritional status of the user.

[0309] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[0310] "Nutritional balance" refers to a state in which necessary nutrients are contained in the appropriate proportions.

[0311] "Cooking method" refers to the method or recipe for making a dish using ingredients.

[0312] "Demand forecasting" refers to predicting future demand for food ingredients.

[0313] "Providing consistently" refers to providing multiple processes continuously and without interruption.

[0314] A "database" refers to a system for systematically storing and managing data.

[0315] A "generative AI model" refers to a model that uses artificial intelligence to generate new information or results from data.

[0316] "Ingredient list" refers to a list of ingredients required for a recipe.

[0317] "Food waste" refers to ingredients and food that are discarded without being consumed.

[0318] An "elderly household" refers to a household where elderly people primarily live.

[0319] "Child-rearing households" refers to households raising children.

[0320] "Special needs" refers to the unique wants and needs of particular users.

[0321] A "comprehensive platform" refers to a foundation that integrates and provides multiple functions and services.

[0322] This invention is a system for optimizing food consumption and reducing food waste. Specifically, users input information about ingredients they own, and a generative AI model is used to generate recipes based on that information and suggest the necessary ingredients to purchase. For elderly households and those with children, purchasing suggestions are also made taking into account their health and living conditions.

[0323] Hardware and software used

[0324] Device: A device such as a smartphone, computer, or tablet is used by the user to input ingredient information.

[0325] Server: A server is used to store data, process data, and run generative AI models. Servers are often built on cloud services (e.g., AWS (registered trademark), Google (registered trademark) Cloud).

[0326] Database: Uses a relational database such as MySQL or PostgreSQL, which efficiently stores and manages the ingredient information entered by users.

[0327] Generative AI model: Uses a generative AI model such as GPT-4 to generate recipes based on the input ingredients.

[0328] Explanation of program processing

[0329] 1. The user enters ingredient information

[0330] Users use devices such as smartphones or computers to input information about ingredients they currently own. For example, they might enter information such as "Tomatoes, 3 pieces, 2023-10-15" or "Chicken, 500g, 2023-10-20" into the application's input form.

[0331] 2. The device sends the ingredient information to the server

[0332] The device sends the ingredient information entered by the user to the server as an HTTP request. At this time, the data is often sent in JSON format. For example, it is sent in the format {"Ingredients": "Tomato", "Quantity": "3", "Best before": "2023-10-15"}.

[0333] 3. The server saves the ingredient information in a database

[0334] The server analyzes the received ingredient information and stores it in a database. A relational database such as MySQL or PostgreSQL is used as the database. For example, an SQL query is used to store the information in the following format: INSERT INTO ingredient_table (ingredient, quantity, expiration date) VALUES ('tomatoes', '3', '2023-10-15').

[0335] 4. The server generates a recipe using the AI ​​model.

[0336] The server generates recipes using a generative AI model (e.g., GPT-4) based on the stored ingredient information. Specifically, the ingredient information is input into the AI ​​model as a prompt to generate a recipe. For example, the prompt might be, "Please generate a simple, nutritionally balanced recipe using the following ingredients: 3 tomatoes, 500g chicken."

[0337] 5. The server sends the recipe and ingredients list to the user.

[0338] The server sends the created recipe and a list of ingredients required for the recipe to the user's device. For example, it sends the HTTP response in the format {"Recipe": "Tomato and Chicken Stew", "Required Ingredients": ["Tomato", "Chicken"]}.

[0339] 6. Users purchase ingredients based on recipes

[0340] The user purchases the necessary ingredients based on the recipe and ingredient list sent from the server. For example, the user goes to a supermarket and purchases "3 tomatoes" and "500g chicken."

[0341] 7. The server makes special suggestions for elderly households and households with children.

[0342] The server makes special purchasing suggestions based on information about the user's health and living situation. For example, it might suggest "nutritiously balanced meals" to elderly households, and "recipes containing ingredients that children like" to households with children. Specifically, it generates recipes by inputting special prompts into the AI ​​model based on the user's profile information.

[0343] Specific examples

[0344] Example of ingredient information entered by the user

[0345] "Tomatoes, 3, 2023-10-15"

[0346] "Chicken, 500g, 2023-10-20"

[0347] Example prompts to be input to the generative AI model

[0348] "Make a simple, nutritionally balanced recipe using the following ingredients: 3 tomatoes, 500g chicken."

[0349] Example of a generated recipe

[0350] "Tomato and Chicken Stew: Finely chop the tomatoes and simmer them in a pot with the chicken. Add salt, pepper, and olive oil to taste."

[0351] In this way, users can consume ingredients efficiently and reduce food waste. Additionally, for elderly households and households with children, suggestions are made according to their special needs, which supports healthy eating habits. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[0352] Step 1:

[0353] The user inputs the ingredient information.

[0354] Users use devices such as smartphones or PCs to input information about ingredients they currently own. For example, they might input information such as "Tomatoes, 3 pieces, 2023-10-15" or "Chicken, 500g, 2023-10-20" into the application's input form. The input information includes the type of ingredient, quantity, expiration date, etc.

[0355] Step 2:

[0356] The terminal transmits the ingredient information to the server.

[0357] The terminal sends the ingredient information entered by the user to the server as an HTTP request. At this time, the data is often sent in JSON format. For example, it is sent in the format {"ingredient": "tomato", "quantity": "3", "best before": "2023-10-15"}. The input is the ingredient information entered by the user, and the output is the JSON format data sent to the server.

[0358] Step 3:

[0359] The server stores the ingredient information in a database.

[0360] The server analyzes the received ingredient information and stores it in a database. A relational database such as MySQL or PostgreSQL is used as the database. For example, an SQL query is used to store the information in the form of INSERT INTO ingredient table (ingredient, quantity, expiration date) VALUES ('tomatoes', '3', '2023-10-15'). The input is JSON-formatted data sent to the server, and the output is the ingredient information stored in the database.

[0361] Step 4:

[0362] The server generates recipes using AI models.

[0363] The server generates a recipe using a generative AI model (e.g., GPT-4) based on the stored ingredient information. Specifically, the ingredient information is input into the AI ​​model as a prompt to generate a recipe. For example, the prompt might be "Please generate a simple, nutritionally balanced recipe using the following ingredients: 3 tomatoes, 500g chicken." The input is the ingredient information stored in the database, and the output is the generated recipe.

[0364] Step 5:

[0365] The server sends the recipe and a list of ingredients to the user.

[0366] The server sends the generated recipe and the list of ingredients required for the recipe to the user's device. For example, it sends the HTTP response in the format {"Recipe": "Tomato and Chicken Stew", "Required Ingredients": ["Tomato", "Chicken"]}. The input is the generated recipe and the list of ingredients, and the output is the data sent to the user's device.

[0367] Step 6:

[0368] A user purchases ingredients based on a recipe.

[0369] The user purchases the necessary ingredients based on the recipe and ingredient list sent from the server. For example, you go to a supermarket and purchase "3 tomatoes" and "500g chicken." The input is the recipe and ingredient list sent from the server, and the output is the purchased ingredients.

[0370] Step 7:

[0371] The server makes special suggestions for elderly households and households with children.

[0372] The server makes special purchasing suggestions based on information about the user's health and lifestyle. For example, it suggests "nutritiously balanced meals" to elderly households, and "recipes containing ingredients that children like" to households with children. Specifically, it generates recipes by inputting special prompts into the AI ​​model based on the user's profile information. The input is information about the user's health and lifestyle, and the output is special suggestions.

[0373] (Application example 3)

[0374] 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."

[0375] In modern society, the problem of food waste is serious, and it is particularly difficult for elderly households and those with children to manage and purchase ingredients. Furthermore, there is a lack of suggestions for appropriate ingredients and recipes based on health status and living situations, making it difficult to eat a nutritionally balanced diet. Furthermore, there is little progress in making effective use of surplus ingredients, resulting in waste.

[0376] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes a means for performing machine learning on a daily basis of food market trends and information on ingredients received at retailers, etc., a means for recommending recipes that take into account the customer's preferences, health condition, seasonal ingredients, and nutritional balance, a means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchasing, a means for generating recipes to encourage the consumption of surplus ingredients, and a means for suggesting ingredients and recipes based on the customer's health condition and lifestyle. This makes it possible to reduce food waste and support shopping for elderly households and households with children.

[0377] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[0378] "Retail store" means a store that sells goods directly to consumers.

[0379] "Information on newly arrived ingredients" refers to information on ingredients that have newly arrived at retail stores or markets.

[0380] "Machine learning" refers to a technology in which computers analyze data, learn patterns, and make predictions and classifications.

[0381] "Customer preferences" refers to the types of ingredients and dishes that customers prefer, as well as their taste preferences.

[0382] "Health Status" refers to information about a customer's current health, such as blood pressure, blood sugar levels, allergies, etc.

[0383] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[0384] "Nutritional balance" refers to a state in which the nutrients necessary to maintain health are appropriately consumed.

[0385] A "recipe" is a set of instructions that describes how to prepare a dish.

[0386] "Recommendation" refers to suggesting the best option based on specific criteria.

[0387] "Demand forecasting" refers to predicting future demand.

[0388] "Through" means to do something consistently from start to finish.

[0389] "Excess ingredients" refers to ingredients that have been purchased but left unused.

[0390] "Encouraging consumption" refers to encouraging people to take a particular action.

[0391] "Lifestyle" refers to information about the customer's daily life, such as family composition and lifestyle habits.

[0392] The system for carrying out the present invention operates in cooperation with three entities: a server, a terminal, and a user. A specific embodiment of the system will be described below.

[0393] Server Processing

[0394] The server uses machine learning to learn daily information about food market trends and ingredients arriving at retail stores. This allows it to constantly grasp the latest market information and forecast demand. The server uses a generative AI model to recommend recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. It also generates recipes to encourage the consumption of surplus ingredients. This information is sent to the user's device.

[0395] Terminal handling

[0396] The device accepts user input and displays recipe and ingredient information sent from the server. Users can input their preferences, health status, and surplus ingredients into the device. The device then sends this information to the server and receives suggestions for optimal recipes and ingredients. Based on the recipe selected by the user, the device generates a shopping list of the necessary ingredients and works with the delivery service to support purchasing.

[0397] User operations

[0398] Users input their preferences, health status, and excess ingredients through their device. This allows them to receive optimal recipe and ingredient suggestions from the server. The user then checks the suggested recipes and creates a list for purchasing the ingredients they need. They can also purchase ingredients using a delivery service.

[0399] Hardware and software used

[0400] Hardware: Smartphone

[0401] Software: Python, OpenAI® API, requests library

[0402] Specific examples

[0403] For example, if a user prefers "Japanese food, low calorie" and has a health condition of "high blood pressure," the following prompt sentences are used to generate recipes:

[0404] Example prompt sentence:

[0405] User preferences: Japanese food, low calorie

[0406] Health condition: High blood pressure

[0407] Based on this, please suggest the best recipe.

[0408] By inputting this prompt into a generative AI model, the system will suggest the best recipe for the user. The user can then review the suggested recipe, generate a shopping list for the necessary ingredients, and purchase the ingredients using a delivery service.

[0409] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0410] Step 1:

[0411] The server collects daily information on food market trends and incoming ingredients from retailers and other sources, and inputs this information into a machine learning model. This allows the system to learn the latest market information and perform demand forecasts. The input data is market trends and incoming ingredient information, and the output is the demand forecast results.

[0412] Step 2:

[0413] The user inputs their preferences, health status, and excess ingredients through the terminal. The input data is the user's preferences, health status, and excess ingredient information, and this information is sent to the server.

[0414] Step 3:

[0415] The server receives information on preferences, health status, and surplus ingredients sent by the user and generates the optimal recipe using a generative AI model. The input data are the user's preferences, health status, and surplus ingredient information, and the output is the optimal recipe.

[0416] Step 4:

[0417] The server sends the generated recipe to the terminal. The input data is the generated recipe, and the output is the recipe sent to the terminal.

[0418] Step 5:

[0419] The terminal displays the recipe sent from the server. The user checks the suggested recipe and generates a list of ingredients to purchase. The input data is the recipe sent from the server, and the output is the ingredient purchasing list.

[0420] Step 6:

[0421] The terminal then sends the generated food shopping list to the delivery service. The input data is the food shopping list, and the output is order information for the delivery service.

[0422] Step 7:

[0423] The delivery service delivers ingredients to users based on order information sent from the terminal. The input data is the order information, and the output is the delivery of ingredients to the user.

[0424] Step 8:

[0425] The user receives the delivered ingredients and prepares a dish based on the proposed recipe. The input data are the delivered ingredients and the recipe, and the output is the completed dish.

[0426] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0427] "Example 1"

[0428] One embodiment of the present invention is a system that incorporates an emotion engine. This system recognizes a user's emotions and recommends recipes based on those emotions. Specifically, when a user is feeling happy, the system recommends recipes that will amplify that joy, and when a user is feeling sad, the system recommends recipes that will alleviate that sadness. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, and the content of social media posts.

[0429] "Example 2"

[0430] Another embodiment of the present invention is a system that incorporates an emotion engine. This system recognizes the user's emotions and makes food purchase suggestions based on those emotions. Specifically, when the user is feeling stressed, the system suggests foods that will relieve stress, and when the user is relaxed, the system suggests foods that will help maintain that relaxed state. The emotion engine recognizes emotions from the user's biometric information (heart rate, blood pressure, etc.).

[0431] "Example 3"

[0432] Furthermore, another embodiment of the present invention is a system that incorporates an emotion engine. This system recognizes a user's emotions and provides delivery services according to those emotions. Specifically, when a user is feeling angry, a service that alleviates the anger (e.g., expedited delivery) is provided, and when a user is feeling happy, a service that amplifies the joy (e.g., including a surprise gift) is provided. The emotion engine recognizes emotions from the user's behavioral patterns and history data.

[0433] The processing flow of each embodiment will be described below.

[0434] "Example 1"

[0435] Step 1: Recognize emotions from the user's facial expressions, tone of voice, and social media posts.

[0436] Step 2: Depending on the emotion you recognize, decide whether to increase joy or reduce sadness.

[0437] Step 3: Based on the judgment, recommend an appropriate recipe.

[0438] "Example 2"

[0439] Step 1: Recognize emotions from the user's biometric information (heart rate, blood pressure, etc.).

[0440] Step 2: Depending on the emotion you identify, we'll suggest foods that will help you relieve stress or keep you relaxed.

[0441] Step 3: Based on your judgment, make appropriate food purchasing recommendations.

[0442] "Example 3"

[0443] Step 1: Recognize emotions from user behavior patterns and historical data.

[0444] Step 2: Depending on the emotion you identify, decide whether to offer a service that reduces anger or increases joy.

[0445] Step 3: Use your judgment to provide the appropriate delivery service.

[0446] Example 1

[0447] 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 device 14 will be referred to as a "terminal."

[0448] In today's grocery market, there is a demand for recipes that meet diverse consumer needs, reducing food waste, and providing shopping support for elderly households and those with children. However, to meet these needs, a system is needed that can efficiently collect information on food market trends and retail store stock, and provide recipes that match consumers' preferences, health status, and emotions. Conventional systems have difficulty meeting all of these requirements in a comprehensive manner, and this has led to the issue of not being able to increase consumer satisfaction.

[0449] 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.

[0450] In this invention, the server includes a means for collecting daily information on food market trends and ingredients arriving at retail stores, a means for training an AI model based on the collected data, a means for recognizing user emotions and recommending recipes based on those emotions, a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, and a means for providing a seamless service from demand forecasting and AI recommendations to purchasing. This makes it possible to provide recipes that meet the diverse needs of consumers, reduce food waste, and provide shopping support for elderly households and households with children.

[0451] "Food market trends" refers to information such as fluctuations in food prices in the market, consumer purchasing trends, and seasonal influences.

[0452] "Information on ingredients arriving at retail stores" refers to information such as the arrival date, quantity, and sales status of each product at a retail store.

[0453] An "AI model" refers to an artificial intelligence model built using a machine learning framework that learns from specific data and makes predictions and classifications.

[0454] "User emotions" refers to the emotional state recognized from the user's facial expressions, tone of voice, and the content of their social media posts.

[0455] "Means for recommending recipes" refers to a function that suggests recipes suitable for the user based on collected data and the learning results of the AI ​​model.

[0456] "Demand forecasting" refers to predicting future demand for a product based on collected data.

[0457] "A means of providing a seamless service" refers to a function that consistently provides a series of processes from data collection to recipe recommendations and purchase.

[0458] "Food waste contribution" refers to reducing food waste and promoting sustainable consumption.

[0459] "Shopping support for elderly households and households with children" refers to a function that helps elderly people and households with children to shop more conveniently.

[0460] "Customer attributes" refers to individual information about customers, such as family composition, age, and preferences.

[0461] A "comprehensive platform" refers to a system that integrates multiple functions and services and provides them to users in a centralized manner.

[0462] MODE FOR CARRYING OUT THE INVENTION

[0463] This invention is a system that collects daily information on food market trends and ingredients arriving at retail stores, and uses an AI model to recommend recipes based on customer preferences, health status, and emotions. Specific embodiments of this system are described below.

[0464] Server Roles

[0465] The server collects daily information on food market trends and food ingredients arriving at retail stores. Specifically, the server obtains food price fluctuation data, consumer purchasing trend data, and seasonal influence data via API. It also downloads CSV files from the FTP server to collect arrival information from each retail store. This data is stored in a database such as MySQL or PostgreSQL.

[0466] Next, the server preprocesses the collected data. Specifically, it complements missing values, normalizes the data, and detects and corrects outliers. Based on the preprocessed data, an AI model is trained using a machine learning framework such as TensorFlow or PyTorch. Once trained, the AI ​​model generates recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[0467] Device Role

[0468] The device uses a camera and microphone to recognize the user's emotions. Specifically, the camera captures the user's facial expressions and the microphone records the tone of voice. With the user's permission, the device also accesses their social media accounts and analyzes their posts. This data is sent to the emotion engine, which uses TensorFlow to analyze the facial and voice data and identify the user's emotions.

[0469] User Roles

[0470] The user receives recipe recommendations through their device. For example, if the user is feeling happy, the system will recommend recipes that will amplify that joy. Specifically, if the user is smiling, the system will display a recipe for a "fruit parfait." Conversely, if the user is feeling sad, the system will display a recipe for a "heartwarming chicken soup."

[0471] Specific examples

[0472] Examples of data collection

[0473] The server retrieves food price fluctuation data from "https: / / api.marketdata.com / prices", authenticates using an API key, and receives the data in JSON format.

[0474] Retail store arrival information is obtained by downloading a CSV file from the FTP server.

[0475] Specific examples of data preprocessing

[0476] The server imputes missing values ​​with the mean and normalizes the data, e.g., scaling price data to a range of 0 to 1.

[0477] To detect outliers, we use Z scores, with data exceeding 3 standard deviations being treated as outliers.

[0478] Specific examples of emotion recognition

[0479] The device captures the user's facial expressions using OpenCV and sends them to the emotion engine, which uses TensorFlow to analyze the facial expression data and identify the user's emotions.

[0480] Example of recipe recommendation

[0481] If the user is feeling happy, the system displays a recipe for a "fruit parfait," detailing the ingredients needed and how to make it.

[0482] Prompt Sentence Examples

[0483] "Build an AI model that can recommend recipes based on market trend data and retail store stock information, taking into account customer preferences and health conditions. Also, combine it with a system that recognizes user emotions and recommends recipes based on those emotions."

[0484] In this way, by clarifying the roles of the server, terminal, and user, and explaining the specific processing content and the hardware and software used, it becomes easier to understand the overall picture of the system.

[0485] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0486] Step 1:

[0487] The server collects food market trend data and information on ingredients arriving at retail stores. Specifically, it obtains food price fluctuation data, consumer purchasing trends, and seasonal influence data via API. It also downloads CSV files from the FTP server to collect arrival information from each retail store. This data is stored in a database such as MySQL or PostgreSQL.

[0488] Input: Market trend data obtained from API, stock information downloaded from FTP server

[0489] Output: Market trend data and stock information stored in a database

[0490] Step 2:

[0491] The server preprocesses the collected data by imputing missing values, normalizing the data, and detecting and correcting outliers. For example, it imputes missing values ​​with the mean and scales price data to a range of 0 to 1. For outlier detection, it uses Z-scores and treats data that is greater than 3 standard deviations as an anomaly.

[0492] Input: Market trend data and stock information stored in the database

[0493] Output: Preprocessed data

[0494] Step 3:

[0495] The server trains an AI model based on the preprocessed data. TensorFlow and PyTorch are used as machine learning frameworks. Once trained, the AI ​​model generates recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[0496] Input: Preprocessed data

[0497] Output: Trained AI model

[0498] Step 4:

[0499] The device uses a camera and microphone to recognize the user's emotions. Specifically, the camera captures the user's facial expressions and the microphone records the user's tone of voice. With the user's permission, the device also accesses their social media accounts and analyzes their posts. This data is then sent to the emotion engine.

[0500] Input: facial expression data captured by the camera, audio data recorded by the microphone, and social media posts

[0501] Output: Data sent to the emotion engine

[0502] Step 5:

[0503] The emotion engine analyzes the collected data and identifies the user's emotions. Specifically, it uses TensorFlow to analyze facial expression and voice data, recognizing "joy" when the user is smiling and "sadness" when the user is crying.

[0504] Input: Data sent to the emotion engine

[0505] Output: Identified user sentiment

[0506] Step 6:

[0507] The user receives recipe recommendations through their device. For example, if the user is feeling happy, the system will recommend recipes that will amplify that joy. Specifically, if the user is smiling, the system will display a recipe for a "fruit parfait." Conversely, if the user is feeling sad, the system will display a recipe for a "heartwarming chicken soup."

[0508] Input: Identified user sentiment

[0509] Output: Recommended recipes

[0510] (Application example 1)

[0511] Next, a description will be given of Application Example 1 of Embodiment 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."

[0512] Modern consumers have diverse needs when it comes to grocery shopping. In particular, they want to be able to select ingredients based on their health-conscious or personal preferences, receive recipe suggestions using seasonal ingredients, and even have meal suggestions tailored to their emotions. However, meeting these needs requires a system that can process vast amounts of information in real time and make appropriate recommendations. There is also a demand for smartphone applications to improve the shopping experience in brick-and-mortar stores. The lack of a comprehensive system to solve these challenges presents a problem.

[0513] 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.

[0514] In this invention, the server includes a means for performing machine learning on a daily basis on food market trends and information on ingredients in stock at retailers, a means for recommending recipes that take into account a customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchase, a means for recognizing a user's emotions and recommending recipes according to those emotions, and a means for providing a smartphone application for improving the shopping experience in physical stores. This enables recommendations that meet the diverse needs of consumers and also improves the shopping experience in physical stores.

[0515] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[0516] A "retail store" is a store that sells goods directly to consumers.

[0517] "Incoming food information" is information about food ingredients arriving at the store, such as the arrival date, amount of each product, and sales status.

[0518] "Machine learning" is a technology that allows computers to automatically learn from data and make predictions and classifications.

[0519] "Customer preferences" refers to information about the preferences of individual customers, such as the ingredients and types of dishes that customers prefer, and their taste preferences.

[0520] "Health Status" refers to information about a customer's current health, including, for example, allergies or specific nutritional needs.

[0521] "Seasonal ingredients" are ingredients that are considered to be at their most delicious and nutritious during that season.

[0522] "Nutritional balance" refers to the appropriate balance of nutrients in a diet.

[0523] A "recipe" is a list of steps and ingredients for making a particular dish.

[0524] "Recommendation" means suggesting the best option based on specific criteria.

[0525] "Demand forecasting" is the process of predicting future demand for a product.

[0526] "Through" means to do something consistently from start to finish.

[0527] "User emotion" refers to the emotional state that the user is feeling, such as joy, sadness, or anger.

[0528] A "brick and mortar store" is a store that exists in a physical location and where consumers visit in person to purchase products.

[0529] A "smartphone application" is a software program that runs on a smartphone.

[0530] A system for implementing this invention is configured as follows: A server collects daily information on food market trends and food ingredients arriving at retail stores, and inputs this data into a machine learning model for learning. Specifically, the server collects market trend data such as food price fluctuations, consumer purchasing habits, and seasonal influences, and collects retail store arrival information such as the arrival date, arrival volume, and sales status of each product.

[0531] The server uses this data to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. Furthermore, to recognize the user's emotions, the server receives input from the user's facial expressions, tone of voice, and social media posts, and uses an emotion engine to recognize their emotions. Based on this emotional information, the server recommends recipes that match the user's emotions.

[0532] The server then provides the generated recipes to users through a smartphone application. This application is designed to enhance the shopping experience in brick-and-mortar stores, allowing users to receive real-time recipe recommendations in-store. It also helps users efficiently purchase the ingredients they need based on demand forecasts.

[0533] The hardware used includes a server and a smartphone. The software used includes Python, a requests library, and an emotion recognition library. The server uses this software to collect data, train machine learning models, recognize emotions, and generate recipes.

[0534] As a concrete example, the following prompt sentences can be used to recognize the user's emotions and recommend appropriate recipes:

[0535] Example prompt sentence:

[0536] "The user's facial expression is smiling, their voice tone is cheerful, and their social media post states, 'Today was such a fun day!'"

[0537] Based on this prompt, the server can recognize that the user is feeling joy and recommend recipes (e.g., party dishes or desserts) that will enhance that joy.

[0538] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0539] Step 1:

[0540] The server collects food market trend data. Specifically, it obtains market trend data (price fluctuations, consumer purchasing trends, seasonal influences, etc.) through API. The input is market trend data from the API, and the output is the collected market trend data.

[0541] Step 2:

[0542] The server collects information on incoming ingredients from retail stores. Specifically, it obtains data such as the arrival date, quantity, and sales status of each product through the API. The input is the incoming ingredient information from the API, and the output is the collected incoming ingredient information.

[0543] Step 3:

[0544] The server inputs the collected market trend data and information on incoming ingredients into a machine learning model for learning. Specifically, it preprocesses the data using Python and machine learning libraries and inputs it into the model. The input is market trend data and information on incoming ingredients, and the output is the trained model.

[0545] Step 4:

[0546] The server generates recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance. Specifically, it uses a trained model to generate recipes based on the user's profile information. The input is the user's profile information, and the output is the generated recipe.

[0547] Step 5:

[0548] The server recognizes the user's emotions. Specifically, it receives the user's facial expressions, tone of voice, and social media posts as input, and uses an emotion engine to recognize emotions. The input is the user's emotional data, and the output is the recognized emotional information.

[0549] Step 6:

[0550] The server recommends recipes based on the recognized emotional information. Specifically, the emotional information is input into a trained model to generate an appropriate recipe. The input is the recognized emotional information, and the output is a recipe based on the emotion.

[0551] Step 7:

[0552] The server provides the generated recipe to the user through a smartphone application. Specifically, it sends recipe information to the application so that the user can check the recipe in real time. The input is the generated recipe, and the output is the recipe information displayed on the user's smartphone.

[0553] Step 8:

[0554] Users use a smartphone application to enhance their shopping experience in a brick-and-mortar store by purchasing ingredients based on recipes recommended through the application. The input is recipe information from the smartphone application, and the output is the purchased ingredients.

[0555] Example 2

[0556] Next, a description will be given of 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."

[0557] In today's modern diet, it is difficult to select and purchase the right ingredients based on customers' preferences and health status. It is also difficult to grasp food market trends and retailer stock information in real time and forecast demand. Furthermore, there is a lack of systems that can suggest ingredients based on the customer's emotional state and efficiently carry out the entire process from purchase to delivery. To solve these issues, a comprehensive system is needed.

[0558] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0559] In this invention, the server includes a means for performing machine learning on a daily basis on food market trends and information on ingredients received at retailers, a means for recommending recipes that take into account a customer's preferences, health condition, seasonal ingredients, and nutritional balance, a means for providing an end-to-end process from demand forecasting and machine learning recommendations to purchasing, a means for analyzing recipes and generating a list of necessary ingredients, a means for collecting biometric information and analyzing emotions, a means for suggesting ingredients to purchase based on the customer's emotions, and a means for arranging delivery after the purchase is confirmed. This enables the selection and purchase of appropriate ingredients based on the customer's preferences, health condition, and emotional state, and allows the entire process from purchase to delivery to be carried out efficiently.

[0560] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[0561] "Retail store" refers to a store or business that sells products directly to consumers.

[0562] "Machine learning" refers to the technology that allows computers to learn patterns from data and make predictions and classifications.

[0563] "Preference" refers to an individual's tendency to prefer certain ingredients or dishes.

[0564] "Health status" refers to the state of an individual's physical and mental health.

[0565] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[0566] "Nutritional balance" refers to a state in which necessary nutrients are consumed in appropriate proportions.

[0567] A "recipe" is a list of steps or ingredients for making a particular dish.

[0568] A "recommendation" is a recommendation for a particular choice or action.

[0569] "Demand forecasting" refers to the process of predicting future demand.

[0570] "Through" means to do something consistently from start to finish.

[0571] "Recipe analysis" refers to the process of analyzing the contents of a recipe and extracting the necessary information.

[0572] "Ingredient list" refers to the list of ingredients needed to make a particular dish.

[0573] "Biometric information" refers to an individual's physical data, such as heart rate and blood pressure.

[0574] "Emotion analysis" refers to the process of analyzing an individual's emotional state based on biometric and other data.

[0575] "Purchase Offer" means an offer to purchase a particular product or service.

[0576] "Confirmation of purchase" refers to a customer's final decision to purchase a product or service.

[0577] "Delivery arrangements" refers to the procedures for delivering purchased products to customers.

[0578] MODE FOR CARRYING OUT THE INVENTION

[0579] This invention is a system that uses machine learning to learn daily information on food market trends and ingredients in stock at retail stores, recommends recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance, and provides a seamless service from demand forecasting to purchasing and delivery. Specific embodiments of this system are described below.

[0580] Hardware and software used

[0581] 1. Server

[0582] The server collects, analyzes, and stores data, and runs machine learning models, using the following software and frameworks:

[0583] Machine learning frameworks: TensorFlow, PyTorch

[0584] Database: MySQL, PostgreSQL

[0585] Natural language processing libraries: NLTK, spaCy

[0586] Sentiment analysis API: IBM Watson (registered trademark), Microsoft (registered trademark) Azure (registered trademark)

[0587] 2. Terminal

[0588] The terminal provides an interface with the user and inputs and displays data. Specifically, the following devices and software are used:

[0589] Devices: Smartphones, tablets, computers

[0590] Applications: Web applications, mobile applications

[0591] 3. Users

[0592] Users use the system to purchase ingredients. They access the system through their terminal, check the suggested recipes and ingredients, and then decide on the purchase.

[0593] Program processing

[0594] 1. Recipe analysis

[0595] The server receives the recipe recommendations using the generative AI model and analyzes them using natural language processing techniques, which generates a list of the ingredients needed.

[0596] 2. Demand forecast

[0597] The server then performs demand forecasting based on the generated ingredient list, taking into account past purchase data, seasonality, and trends. A time series analysis model is used for time series analysis.

[0598] 3. Purchase proposal

[0599] The server then makes food purchasing suggestions to customers based on predicted demand, which are displayed on their devices via a web or mobile application.

[0600] 4. Biometric data collection and emotion analysis

[0601] The device collects the user's biometric information (heart rate, blood pressure, etc.) through sensors and sends it to the server. The server then uses an emotion engine to analyze the user's emotions and makes food purchase suggestions based on the analysis results.

[0602] 5. Purchase confirmation and delivery arrangements

[0603] The user checks the suggested ingredients and decides to purchase them. The purchase information is sent to the server, and the order is confirmed. The server then sends the order information to a delivery service, and the ingredients are delivered to the user.

[0604] Specific examples

[0605] For example, if the AI ​​recommends a recipe for "chicken curry," the server generates a list of ingredients, such as chicken, curry powder, onions, and tomatoes. The server uses past data to predict demand for these ingredients and suggests them to the user. Once the user decides to purchase them, the ingredients are delivered to the user via a delivery service.

[0606] Prompt Sentence Examples

[0607] "Generate a list of suggested foods for when the user is feeling stressed."

[0608] "Create a list of ingredients needed for the following recipe and forecast demand: Chicken curry."

[0609] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0610] Step 1: Recipe Analysis

[0611] The server receives recipes recommended using a generative AI model. As input, it receives the recipe's text data. The server analyzes the recipe content using natural language processing techniques (e.g., NLTK or spaCy) and generates a list of required ingredients. As output, it obtains a list of ingredients. This list includes the name and required quantity of each ingredient. Specifically, it tokenizes the recipe text and extracts the ingredient names and their quantities.

[0612] Step 2: Demand forecast

[0613] The server performs demand forecasting based on the ingredient list generated in step 1. As input, it receives the ingredient list, past purchase data, seasonal data, and trend data. The server uses a time series analysis model to predict the demand for each ingredient. As output, it obtains the predicted demand amount. Specifically, it inputs past data into the time series model to predict future demand.

[0614] Step 3: Purchase proposal

[0615] The server makes food purchase suggestions to customers based on the demand predicted in step 2. The inputs are the predicted demand, customer preference data, and health status data. The server integrates these data and generates optimal purchase suggestions. The output is a list of purchase suggestions. This list is displayed on the device via a web or mobile application. Specifically, the suggestion list is filtered based on the customer's preferences and health status to select the most suitable ingredients.

[0616] Step 4: Biometric data collection and emotion analysis

[0617] The device collects the user's biometric information (heart rate, blood pressure, etc.) through sensors. The biometric data is given as input. The device sends this data to a server. The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure emotion recognition API) to analyze the user's emotions. The output is the user's emotional state. Specifically, the biometric information is input into the emotion engine, and the emotional state is analyzed.

[0618] Step 5: Emotion-based purchase recommendations

[0619] The server makes food purchase suggestions based on the emotional state analyzed in step 4. As input, the emotional state data and a list of ingredients are given. The server selects ingredients according to the emotional state and generates a list of purchase suggestions. As output, a list of purchase suggestions based on emotions is obtained. This list is displayed on the terminal. Specifically, the server filters ingredients according to the emotional state and makes the most suitable suggestions.

[0620] Step 6: Confirm your purchase and arrange delivery

[0621] The user checks the suggested ingredients and decides to purchase them. Purchase confirmation information is provided as input. The terminal sends this information to the server. The server receives the purchase information and confirms the order. Confirmed order information is obtained as output. The server sends this information to the delivery service and arranges for delivery. Specifically, the order information is entered into the delivery service's system and the delivery procedure begins.

[0622] (Application example 2)

[0623] 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."

[0624] Conventional food purchasing systems have problems such as difficulty in suggesting ingredients that match the customer's preferences, health condition, and emotions, and the entire process from purchase to delivery is not efficient.Furthermore, there are issues such as insufficient efforts to reduce food waste and insufficient shopping support for elderly households and households with children.

[0625] 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.

[0626] In this invention, the server includes a means for performing machine learning on a daily basis to acquire information on food market trends and ingredients available at retailers, a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing a seamless service from demand forecasting and machine learning recommendations to purchasing, a means for acquiring biometric information and analyzing it with an emotion recognition engine, a means for suggesting ingredients and recipes based on emotions, and a means for purchasing the suggested ingredients and delivering them to the customer using a delivery service. This allows for the recommendation of ingredients based on the customer's preferences, health status, and emotions, and efficiently carries out the entire process from purchase to delivery. It also enables the reduction of food waste and shopping support for elderly households and households with children.

[0627] "Food market trends" refers to fluctuations and trends in supply and demand in the food market.

[0628] "Retail store" means a store that sells goods directly to consumers.

[0629] "Information on newly arrived ingredients" refers to information on ingredients that have newly arrived at retail stores or markets.

[0630] "Daily" means something that happens every day.

[0631] "Machine learning" refers to a technology in which computers analyze data and learn patterns to automatically improve their performance.

[0632] "Customer preferences" refers to the types of ingredients and dishes that customers prefer.

[0633] "Health Status" refers to the state of a Customer's physical health.

[0634] "Seasonal ingredients" refer to ingredients that are at their most delicious and nutritious during that season.

[0635] "Nutritional balance" refers to the balance of nutrients contained in a meal.

[0636] A "recipe" refers to how to prepare a dish or how to combine ingredients.

[0637] "Recommendation" refers to suggesting the best option based on specific criteria.

[0638] "Demand forecasting" refers to predicting future demand.

[0639] "Through" means to do something consistently from start to finish.

[0640] "Biometric information" refers to physical data such as heart rate and blood pressure.

[0641] An "emotion recognition engine" refers to technology that analyzes biometric information to recognize a user's emotions.

[0642] "Recommendation" refers to showing the best options based on specific criteria.

[0643] "Delivery service" refers to a service that delivers purchased products to customers.

[0644] "Food waste reduction" refers to reducing food waste.

[0645] An "elderly household" refers to a household in which an elderly person primarily lives.

[0646] "Child-rearing households" refers to households raising children.

[0647] "Shopping assistance" refers to providing support when shopping.

[0648] "Customer attributes" refer to characteristics such as a customer's family structure, age, and preferences.

[0649] A "comprehensive platform" refers to a foundation that provides multiple functions and services in a single system.

[0650] A system for carrying out this invention is configured as follows: The server includes means for performing machine learning on a daily basis information on food market trends and ingredients received at retailers, etc., means for recommending recipes that take into consideration the customer's preferences and health status, seasonal ingredients, and nutritional balance, means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchasing, means for acquiring biometric information and analyzing it with an emotion recognition engine, means for suggesting ingredients and recipes according to emotions, and means for purchasing the suggested ingredients and delivering them to the customer using a delivery service.

[0651] Hardware and software used

[0652] Hardware: Smartphones, wearable devices (heart rate, blood pressure measurement)

[0653] Software: EmotionEngine (emotion recognition engine), RecipeAI (recipe recommendation engine), DeliveryAPI (delivery service API)

[0654] Data processing and calculation

[0655] 1. Acquisition of biometric information: The device (smartphone or wearable device) acquires biometric information such as the user's heart rate and blood pressure.

[0656] 2. Emotion recognition: The biometric information acquired by the server is analyzed using EmotionEngine to recognize the user's emotions.

[0657] 3. Recipe recommendation: The server uses RecipeAI to recommend appropriate recipes based on the emotions recognized by the server.

[0658] 4. Purchase and delivery: The server purchases the necessary ingredients based on the recommended recipe using the Delivery API and delivers them to the user using a delivery service.

[0659] Specific examples

[0660] For example, if a user is feeling stressed, EmotionEngine will recognize this and RecipeAI will recommend recipes containing ingredients that have stress-relieving properties, such as chamomile tea and dark chocolate.

[0661] Prompt Sentence Examples

[0662] "Suggest recipes that are suitable for when the user's heart rate is high and they are feeling stressed."

[0663] In this way, a system can be realized that provides food delivery services according to the user's preferences, health condition, and emotions.

[0664] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0665] Step 1:

[0666] The terminal acquires the user's biometric information.

[0667] Input: User's vital signs such as heart rate and blood pressure

[0668] Data processing: The device uses sensors to measure biometric information and acquires it as digital data.

[0669] Output: Acquired biometric data

[0670] Step 2:

[0671] The server sends biometric information to the Emotion Engine to recognize emotions.

[0672] Input: Acquired biometric data

[0673] Data calculation: EmotionEngine analyzes biometric information and recognizes the user's emotions

[0674] Output: Recognized emotion data (e.g., stress, relaxation, etc.)

[0675] Step 3:

[0676] Based on the recognized emotional data, the server sends prompts to RecipeAI and recommends appropriate recipes.

[0677] Input: Recognized emotion data

[0678] Data processing: The server generates prompts based on the emotion data and sends them to RecipeAI.

[0679] Output: Recommended recipe data

[0680] Step 4:

[0681] The server generates a list of necessary ingredients based on the recommended recipe data.

[0682] Input: Recommended recipe data

[0683] Data processing: The server analyzes the recipe data and generates a list of required ingredients.

[0684] Output: List of ingredients needed

[0685] Step 5:

[0686] The server purchases ingredients using the Delivery API based on the list of ingredients needed.

[0687] Input: List of ingredients needed

[0688] Data calculation: The server sends the ingredients list to the Delivery API and completes the purchase process.

[0689] Output: Purchase confirmation data

[0690] Step 6:

[0691] Based on the purchase confirmation data, the server uses a delivery service to deliver the ingredients to the user.

[0692] Input: Purchase confirmation data

[0693] Data processing: The server sends the necessary information to the delivery service and processes the delivery.

[0694] Output: Delivery confirmation data

[0695] In this way, a series of processes is realized, including recognizing the user's emotions based on their biometric information, recommending appropriate recipes, and purchasing and delivering the necessary ingredients.

[0696] Example 3

[0697] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0698] Food waste and food loss are major issues in modern society. Elderly households and those with children face the added challenge of having to make the most of shopping, making it difficult for them to select the right ingredients for their health and living situation. Furthermore, there is a demand for services that respond to the emotions of users, but the current system does not adequately address these needs.

[0699] 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.

[0700] In this invention, the server includes: a means for performing machine learning on food market trends and information on ingredients available at retail stores on a daily basis; a means for recommending recipes that take into account a user's preferences, health status, seasonal ingredients, and nutritional balance; a means for providing a comprehensive service from demand forecasting and machine learning recommendations to purchasing; a means for inputting a user's ingredient information; a means for analyzing the input ingredient information and generating optimal recipes; a means for recognizing emotions based on the user's behavioral patterns and historical data; and a means for providing delivery services based on the results of emotion recognition. This makes it possible to reduce food waste and food loss. It also makes it possible to provide elderly households and households with children with support in selecting and purchasing ingredients appropriate for their health status and lifestyle. Furthermore, it is possible to provide services tailored to the user's emotions, which is expected to improve user satisfaction.

[0701] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[0702] "Information on ingredients arriving at retail stores" refers to information on the types and quantities of ingredients newly arrived at retail stores.

[0703] "Machine learning" refers to a technology in which a computer automatically learns from data and makes predictions and classifications.

[0704] "User preferences" refers to personal preferences such as the types of ingredients and dishes preferred by the user, and taste preferences.

[0705] "Health Condition" refers to a User's physical health status or specific health condition (e.g., high blood pressure, diabetes, etc.).

[0706] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[0707] "Nutritional balance" refers to a state in which a diet contains an appropriate amount of necessary nutrients.

[0708] A "recipe" is a list of steps or ingredients for making a particular dish.

[0709] "Demand forecasting" refers to the analysis and calculations used to predict future demand.

[0710] "Ingredient information" refers to information regarding the types and quantities of ingredients owned by the user.

[0711] "Behavioral patterns" refer to a user's past actions and habits.

[0712] "Historical data" refers to a user's past purchase history and behavioral history.

[0713] "Emotion recognition" refers to technology that analyzes and recognizes the user's emotions.

[0714] "Delivery Service" refers to a service that delivers purchased products to a location designated by the user.

[0715] MODE FOR CARRYING OUT THE INVENTION

[0716] This invention is a system for optimizing food consumption and reducing food waste. This system inputs the user's ingredient information, generates optimal recipes, and suggests purchasing the necessary ingredients. It also makes purchasing suggestions for elderly households and households with children, taking into account their health status and living situation. Furthermore, by combining it with an emotion engine, it provides a delivery service that responds to the user's emotions.

[0717] Hardware and software used

[0718] Server: Database server, AI model server

[0719] Device: User's smartphone or tablet

[0720] Software: AI recommendation engine, emotion recognition engine, delivery management system

[0721] Specific operation of the system

[0722] 1. Enter the user's ingredient information

[0723] Users use their smartphones or tablets to enter current food inventory information into the app.

[0724] Specifically, the user enters information such as "3 tomatoes, 1 pack of cheese, 1 bunch of basil" into the app's input form.

[0725] 2. Sending food information

[0726] The terminal transmits the input ingredient information to the server.

[0727] Specifically, the terminal converts the input data into JSON format and sends it to the server using the HTTPS protocol.

[0728] 3. Optimizing food consumption

[0729] The server passes the received ingredient information to the AI ​​recommendation engine.

[0730] The server uses an AI recommendation engine to analyze the input ingredient information and generate the optimal recipe.

[0731] Specifically, the server generates a recipe for "Caprese salad" based on "tomato, cheese, and basil" and suggests it to the user.

[0732] 4. Generate purchase offers

[0733] The server generates a shopping list of necessary ingredients based on the generated recipe.

[0734] The server makes purchasing suggestions to elderly households and households with children, taking into account their health and living conditions.

[0735] Specifically, the server adds the olive oil, salt, and pepper needed for a Caprese salad to the shopping list, and suggests low-sodium salt for elderly households.

[0736] 5. Emotion recognition

[0737] The server uses an emotion recognition engine to analyze the user's emotions based on the user's behavioral patterns and historical data.

[0738] Specifically, the server analyzes the user's past purchase history and reviews, and if the user is dissatisfied with "slow delivery," it recognizes that emotion as "anger."

[0739] 6. Providing delivery services

[0740] The server provides appropriate delivery services based on the results of emotion recognition.

[0741] The server offers rush delivery if the user is angry, or a surprise gift package if the user is happy.

[0742] Specifically, the server will offer "free delivery on the next order" and, if the user is happy, will include a "surprise gift."

[0743] Examples and prompts

[0744] Specific examples

[0745] The user inputs "ingredients in the refrigerator" on their smartphone.

[0746] The terminal transmits the input data to the server.

[0747] The server generates a recipe for "Caprese salad" based on "tomato, cheese, and basil" and suggests it to the user.

[0748] Servers add the olive oil, salt, and pepper needed for a Caprese salad to the shopping list, and suggest low-sodium salt for elderly households.

[0749] The server analyzes the user's past purchase history and reviews, and if the user is dissatisfied with "slow delivery," it recognizes that emotion as "anger."

[0750] The server offers "free shipping on your next delivery" and, if the customer is happy, includes a "surprise gift."

[0751] Prompt Sentence Examples

[0752] Please enter the ingredients you have in your refrigerator.

[0753] "We recommend purchasing the following ingredients based on your health status."

[0754] "Based on current sentiment, we'll offer your next delivery free."

[0755] The above is a specific embodiment of the system of the present invention. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0756] Step 1:

[0757] Users use their smartphones or tablets to enter current food inventory information into the app.

[0758] Input: The type and quantity of ingredients the user owns (e.g., "3 tomatoes, 1 pack of cheese, 1 bunch of basil")

[0759] Output: The entered ingredient information is saved on the device.

[0760] Specifically, the user enters ingredient information into the application's input form and presses the submit button.

[0761] Step 2:

[0762] The terminal transmits the input ingredient information to the server.

[0763] Input: Ingredient information entered by the user

[0764] Output: Ingredient information sent to the server

[0765] Specifically, the terminal converts the input data into JSON format and sends it to the server using the HTTPS protocol.

[0766] Step 3:

[0767] The server passes the received ingredient information to the AI ​​recommendation engine.

[0768] Input: Ingredient information sent from the device

[0769] Output: Ingredient information passed to the AI ​​recommendation engine

[0770] Specifically, the server analyzes the received data and passes it on to the AI ​​recommendation engine.

[0771] Step 4:

[0772] The server uses an AI recommendation engine to analyze the input ingredient information and generate the optimal recipe.

[0773] Input: Ingredient information passed to the AI ​​recommendation engine

[0774] Output: The generated recipe (e.g. "Caprese Salad")

[0775] Specifically, the server generates a recipe for "Caprese salad" based on "tomato, cheese, and basil" and suggests it to the user.

[0776] Step 5:

[0777] The server generates a shopping list of necessary ingredients based on the generated recipe.

[0778] Input: Generated recipe

[0779] Output: Shopping list (e.g. "olive oil, salt, pepper")

[0780] Specifically, the server lists the ingredients needed for a recipe and suggests "low-sodium salt" for elderly households.

[0781] Step 6:

[0782] The server uses an emotion recognition engine to analyze the user's emotions based on the user's behavioral patterns and historical data.

[0783] Input: User behavior patterns and historical data

[0784] Output: Recognized emotion (e.g., "anger")

[0785] Specifically, the server analyzes the user's past purchase history and reviews, and if the user is dissatisfied with "slow delivery," it recognizes that emotion as "anger."

[0786] Step 7:

[0787] The server provides appropriate delivery services based on the results of emotion recognition.

[0788] Input: Recognized emotion

[0789] Output: Delivery service offered (e.g. "Next delivery is free")

[0790] Specifically, the server will offer "free delivery on the next order" and, if the user is happy, will include a "surprise gift."

[0791] (Application example 3)

[0792] 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."

[0793] In modern society, food waste is an issue, and the need for shopping support for elderly households and those with children is growing. There is also a demand for services that respond to users' emotions, but no effective system exists to achieve this. Another issue is the lack of a system that can make recommendations based on customer preferences and health status, and provide a seamless service from demand forecasting to purchasing.

[0794] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes a means for performing machine learning on a daily basis of food market trends and information on ingredients received at retailers, etc., a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchasing, and a means for recognizing user emotions and providing delivery services in accordance with those emotions. This makes it possible to reduce food waste, provide shopping support to elderly households and households with children, and further realize the provision of services in accordance with the user's emotions.

[0795] "Food market trends" refers to fluctuations and trends in food supply and demand.

[0796] "Retail store" means a store that sells goods directly to consumers.

[0797] "Information on newly arrived ingredients" refers to information on the types and quantities of ingredients newly arrived at retail stores and markets.

[0798] "Machine learning" refers to a technology in which computers analyze data, recognize patterns, and make predictions and judgments.

[0799] "Customer preferences" refer to the types and characteristics of products and services that customers prefer.

[0800] "Health status" refers to the state of an individual's physical and mental health.

[0801] "Seasonal ingredients" refer to ingredients that are at their most delicious and nutritious during that season.

[0802] "Nutritional balance" refers to the proper distribution of nutrients necessary to maintain health.

[0803] A "recipe" is a set of instructions that describes how to prepare a dish.

[0804] "Recommendation" refers to presenting the best options based on specific criteria.

[0805] "Demand forecasting" refers to predicting future demand.

[0806] "Through" means to do something consistently from start to finish.

[0807] "User emotion" refers to the emotions or moods that a user is feeling.

[0808] "Delivery service" refers to a service that delivers products to a location designated by the customer.

[0809] A system for implementing the present invention includes the following major components:

[0810] 1. Server:

[0811] The server uses machine learning to learn daily information on food market trends and incoming ingredients from retailers, etc. This allows it to constantly monitor the latest market trends and incoming information and store it in a database. The software used is a machine learning algorithm (e.g., TensorFlow, PyTorch).

[0812] 2. Terminal:

[0813] The device (smartphone, tablet, etc.) recommends recipes that take into consideration the customer's preferences, health status, seasonal ingredients, and nutritional balance. The device receives data sent from the server and displays the optimal recipe for the user. Recipe recommendations use a recommendation engine (e.g., Collaborative Filtering, Content-Based Filtering).

[0814] 3. User:

[0815] Users can check recipes and purchase the necessary ingredients through the device. The device is equipped with emotion recognition software to recognize the user's emotions. The device analyzes emotions from text and voice data entered by the user and provides delivery services according to those emotions.

[0816] 4. Delivery Service:

[0817] The server provides a delivery service according to the user's emotions. For example, if the user is feeling angry, the server will send a rush delivery, and if the user is feeling happy, the server will include a surprise gift. A delivery management system is used to manage the delivery service.

[0818] Examples:

[0819] User profile: 70-year-old elderly person with diabetes, prefers a low-carb, high-fiber diet

[0820] User input: "I'm feeling very stressed today."

[0821] Example prompt for a generative AI model:

[0822] User profile: Age: 70, Household type: Seniors, Health condition: Diabetic, Preferences: Low sugar, High fiber

[0823] User input: "I'm feeling very stressed today."

[0824] Design an application to recommend appropriate recipes to this user and provide delivery services based on their emotions.

[0825] In this way, we can realize a system that reduces food waste and provides shopping support to elderly households and families with children. Furthermore, it will be possible to provide services that respond to the user's emotions.

[0826] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0827] Step 1:

[0828] The server collects daily information on food market trends and incoming ingredients from retailers, etc., and analyzes it using machine learning algorithms. Input includes arrival information from each retailer and market trend data. This data is analyzed and the latest market trends and arrival status are stored in a database. The output is the analysis results, which are saved in the database.

[0829] Step 2:

[0830] The server recommends recipes based on the customer's preferences, health status, seasonal ingredients, and nutritional balance. The input includes customer profile information (age, health status, preferences, etc.). A recommendation engine is used to analyze this information and generate optimal recipes. The output is the recommended recipe sent to the device.

[0831] Step 3:

[0832] The terminal displays the recipe sent from the server to the user. The input includes the recipe data sent from the server. The terminal visually displays this to the user, allowing the user to confirm the recipe. The output is a screen where the user can confirm the recipe.

[0833] Step 4:

[0834] The user checks the recipe through the terminal and purchases the necessary ingredients. The input includes the recipe information displayed on the terminal. The user selects the necessary ingredients based on the recipe and completes the purchase procedure. The output is a shopping list that is generated and sent to the server.

[0835] Step 5:

[0836] The device uses emotion recognition software to recognize the user's emotions. Inputs include the user's text and voice data. The emotion recognition software analyzes this data and identifies the user's emotions. The output is recognized emotion data, which is sent to the server.

[0837] Step 6:

[0838] The server provides delivery services according to the user's emotions. The input includes the recognized emotion data. The server selects an appropriate delivery service based on the emotion data and issues instructions to the delivery management system. For example, if the user is feeling angry, the server will expedite the delivery, and if the user is feeling happy, the server will include a surprise gift. The output is a delivery instruction that is generated and sent to the delivery management system.

[0839] Step 7:

[0840] The delivery management system executes deliveries based on instructions from the server. The input includes delivery instructions from the server. The delivery management system follows the instructions to properly package the product and deliver it to the user using the specified delivery method. The output is the product being delivered to the user.

[0841] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0842] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0843] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0844] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0845] [Second embodiment]

[0846] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0847] 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.

[0848] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0849] 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.

[0850] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0851] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0852] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0853] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0854] 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.

[0855] 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.

[0856] 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.

[0857] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0858] "Example 1"

[0859] A first embodiment of the present invention is a system that uses AI to learn daily information about food market trends and supermarket stock information. This system collects market trend data and supermarket stock information, and the AI ​​learns based on this information. Specifically, market trend data includes, for example, food price fluctuations, consumer purchasing habits, and seasonal influences. Furthermore, supermarket stock information includes, for example, the arrival date, stock volume, and sales status of each product. Based on this information, the AI ​​recommends recipes that take into account the customer's preferences, health condition, seasonal ingredients, and nutritional balance.

[0860] "Example 2"

[0861] A second embodiment of the present invention is a system that provides a seamless service from demand forecasting to purchasing and delivery. This system predicts the demand for ingredients based on recipes recommended by AI, and enables customers to purchase them. Specifically, the AI ​​creates a list of ingredients included in the recommended recipes, and suggests purchasing ingredients based on that list. The purchased ingredients are then delivered to the customer using a delivery service.

[0862] "Example 3"

[0863] A third embodiment of the present invention is a system that contributes to reducing food waste and provides shopping support to elderly households and households with children. This system makes suggestions to optimize food consumption and reduce food waste. Specifically, it suggests purchasing necessary ingredients based on recipes recommended by AI, encouraging the consumption of surplus ingredients. In addition, for elderly households and households with children, it makes purchasing suggestions that take into account their health status and living situation.

[0864] The processing flow of each embodiment will be described below.

[0865] "Example 1"

[0866] Step 1: Collect market trend data and supermarket inventory information. This information includes, for example, food price fluctuations, consumer purchasing habits, seasonal influences, arrival dates, quantities, and sales status of each product.

[0867] Step 2: The AI ​​learns from the collected information. Based on market trends and arrival information, the AI ​​learns a model to recommend recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[0868] Step 3: Using the AI-trained model, the system recommends recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[0869] "Example 2"

[0870] Step 1: Based on the recipes recommended by AI, predict the demand for ingredients. Specifically, create a list of ingredients included in the recipes and suggest purchasing ingredients based on that list.

[0871] Step 2: The ingredients that the customer has decided to purchase are delivered to the customer using a delivery service.

[0872] "Example 3"

[0873] Step 1: Based on the recipes recommended by AI, suggestions are made to optimize ingredient consumption and reduce food waste. Specifically, a list of ingredients included in the recipe is created, and based on that, suggestions are made to encourage the consumption of surplus ingredients.

[0874] Step 2: Make purchasing suggestions to elderly households and households with children, taking into account their health and living conditions.

[0875] Example 1

[0876] 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."

[0877] In today's grocery market, consumers demand ingredient selection and recipe suggestions that accommodate diverse tastes, health conditions, and seasonal changes. However, providing appropriate recipes that meet these needs is not easy. Reducing food waste and providing shopping support for elderly households and families with children are also important issues. Furthermore, there is a need to build a comprehensive platform that delivers information based on customer attributes. To solve these issues, an efficient and effective system is required.

[0878] 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.

[0879] In this invention, the server includes: a means for performing machine learning on daily food market trends and retail store inventory information; a means for storing the collected data in a database; a means for preprocessing the stored data; a means for inputting the preprocessed data into a machine learning model; a means for generating recipes using the machine learning model that take into account customer preferences, health status, seasonal ingredients, and nutritional balance; a means for providing the generated recipes to a user terminal; and a means for collecting user feedback and using it to improve the system. This enables the provision of recipes that meet diverse consumer needs, reducing food waste and providing shopping support to elderly households and households with children. It also enables the creation of a comprehensive platform that distributes information based on customer attributes.

[0880] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[0881] "Retail store stock information" is information about product stock at retail stores, such as the arrival date, stock volume, and sales status of each product.

[0882] "Machine learning" is a technology that allows computers to learn patterns based on data and make predictions and classifications.

[0883] A "database" is a system for efficiently storing, managing, and retrieving data.

[0884] "Preprocessing" refers to the process of preparing data before inputting it into a machine learning model, such as cleaning and normalizing the data.

[0885] A "machine learning model" is an algorithm or mathematical model that learns from data and makes predictions or classifications.

[0886] A "recipe" is a set of instructions that shows how to prepare a dish and how to combine ingredients.

[0887] A "user terminal" is a device that allows a user to receive information and perform operations.

[0888] "Feedback" is information that is useful for improving the system, such as user ratings and comments.

[0889] "Food waste reduction" is an initiative aimed at reducing food waste and using it efficiently.

[0890] "Shopping support" is a service that provides shopping support to consumers with specific needs, such as elderly households and households with children.

[0891] "Customer attributes" refers to characteristics and information about customers, such as family structure, age, and preferences.

[0892] An "integrated platform" is a system that integrates multiple functions and services to provide users with a consistent experience.

[0893] This invention is a system that collects food market trends and retail store arrival information daily, and uses machine learning to generate recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance. Specific embodiments of this system are described below.

[0894] Server Processing

[0895] 1. Data Collection

[0896] The server collects food market trend data and retail store arrival information every day at 9:00 a.m. Market trend data includes food price fluctuations, consumer purchasing trends, seasonal influences, etc. This data is obtained using the API of the food market information service.

[0897] The retail store's stock information includes the arrival date, stock quantity, sales status, etc. of each product. This data is obtained from the retail store's inventory management system.

[0898] 2. Data storage

[0899] The server stores the collected data in a MySQL database, which stores data such as food price fluctuation data, consumer purchasing trends, and retail store inventory information as tables.

[0900] 3. Data Preprocessing

[0901] The server preprocesses the stored data, including cleaning the data (filling in missing values ​​and removing outliers) and normalizing it. The server uses the Python pandas library to clean and normalize the data.

[0902] 4. Inputting data into the AI ​​model

[0903] The server then inputs the preprocessed data into an AI model built with TensorFlow, which learns from food price fluctuations, consumer purchasing trends, and retailer inventory information.

[0904] 5. Recipe Generation

[0905] The server uses an AI model to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. The generated recipes are then stored in a database.

[0906] 6. Recipe provision

[0907] The server sends the generated recipe to the terminal via the API.

[0908] Terminal handling

[0909] 1. Receiving recipes

[0910] The terminal receives the recipe information sent from the server and stores the received recipe information in a local database of the terminal.

[0911] 2. Recipe display

[0912] The device then displays the received recipe information to the user. Front-end frameworks such as React and Vue.js are used for displaying the information, and detailed recipe information (ingredients, cooking instructions, nutritional information, etc.) is displayed in an easy-to-read format.

[0913] 3. Feedback Collection

[0914] The device collects feedback from users, including recipe ratings and comments, and stores the feedback in a local database on the device.

[0915] 4. Send Feedback

[0916] The terminal sends the collected feedback to the server.

[0917] User operations

[0918] 1. View recipes

[0919] The user browses the provided recipes through the device, checks the detailed information of the recipes, and selects the recipes they want to try.

[0920] 2. Recipe Trial

[0921] The user then tries out the selected recipe and cooks the dish according to the recipe.

[0922] 3. Feedback input

[0923] Users can input their ratings and comments on the recipes into the device, including feedback on the cooking results, taste, ease of use, etc.

[0924] 4. Send Feedback

[0925] The user sends the input feedback to the server through the terminal.

[0926] Specific examples

[0927] Example 1: Collecting market trend data

[0928] The server uses the API of a food market information service to obtain food price fluctuation data at 9:00 AM every day. For example, it sends the following API request:

[0929] GET / api / market_trends?date=2023-10-01

[0930] Example 2: Data input to an AI model

[0931] The server inputs the preprocessed data into an AI model built with TensorFlow, for example using a prompt like this:

[0932] Prompt: Based on food price fluctuation data and retail store stock information, predict food prices for the next week.

[0933] Example 3: Creating and serving recipes

[0934] The server uses the AI ​​model to generate recipes, which are then sent to the device. For example, the following recipes are provided:

[0935] Recipe name: Healthy salad made with seasonal vegetables

[0936] Ingredients: Tomato, cucumber, lettuce, olive oil, salt

[0937] Directions: 1. Thinly slice the tomato and cucumber. 2. Tear the lettuce into bite-sized pieces. 3. Place all ingredients in a bowl and toss with olive oil and salt.

[0938] This way, users receive fresh, healthy recipes every day.

[0939] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0940] Step 1: Data collection

[0941] The server collects food market trend data and retail store arrival information every day at 9:00 AM. Market trend data includes food price fluctuations, consumer purchasing trends, seasonal effects, etc. This data is obtained using the API of the food market information service. Retail store arrival information includes the arrival date, arrival quantity, sales status, etc. of each product. This data is obtained from the retail store's inventory management system. The input is the API request, and the output is the collected market trend data and arrival information.

[0942] Step 2: Save data

[0943] The server stores the collected data in a MySQL database. The database stores data such as food price fluctuation data, consumer purchasing trends, and retail store inventory information as tables. The input is the collected data, and the output is the data stored in the database.

[0944] Step 3: Data Preprocessing

[0945] The server preprocesses the stored data. Preprocessing includes data cleaning (filling in missing values ​​and removing outliers) and normalization. The Python pandas library is used to clean and normalize the data. The input is the raw data stored in the database, and the output is the preprocessed data.

[0946] Step 4: Input data into the AI ​​model

[0947] The server inputs the preprocessed data into an AI model built with TensorFlow. The AI ​​model learns based on food price fluctuations, consumer purchasing trends, and retail store inventory information. The input is the preprocessed data, and the output is the learning results.

[0948] Step 5: Recipe Generation

[0949] The server uses an AI model to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. The generated recipes are stored in a database. The input is the learning results of the AI ​​model, and the output is the generated recipe.

[0950] Step 6: Provide the recipe

[0951] The server sends the generated recipe to the terminal via API. The input is the generated recipe, and the output is the recipe information sent to the terminal.

[0952] Step 7: Receiving the recipe

[0953] The terminal receives recipe information sent from the server. The received recipe information is stored in the terminal's local database. The input is the recipe information sent from the server, and the output is the recipe information stored in the local database.

[0954] Step 8: View the recipe

[0955] The device displays the received recipe information to the user. Front-end frameworks such as React and Vue.js are used for display. Detailed information about the recipe (ingredients, cooking instructions, nutritional information, etc.) is displayed in an easy-to-read format. The input is the recipe information stored in the local database, and the output is the recipe information displayed to the user.

[0956] Step 9: Gather feedback

[0957] The device collects feedback from users. The feedback includes recipe ratings and comments. The collected feedback is stored in a local database on the device. The input is the feedback from the user, and the output is the feedback stored in the local database.

[0958] Step 10: Send feedback

[0959] The terminal sends the collected feedback to the server, where the input is the feedback stored in the local database and the output is the feedback sent to the server.

[0960] Step 11: Use feedback

[0961] The server receives feedback sent by users and uses it to improve the system. The input is the feedback sent by users, and the output is the feedback used to improve the system.

[0962] (Application example 1)

[0963] 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."

[0964] Modern consumers have diverse needs when it comes to purchasing groceries. In particular, they want recipe suggestions based on their health status and preferences, meal suggestions using seasonal ingredients, and efficient purchasing that takes into account market trends and supermarket stock information. However, there is no consistent system in place to meet these needs. Reducing food waste and providing shopping support for elderly households and households with children are also important issues.

[0965] 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.

[0966] In this invention, the server includes: means for daily AI learning of food market trends and supermarket inventory information; means for recommending recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance; means for providing an end-to-end process from demand forecasting and AI recommendations to purchase; means for recommending recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance through an application installed on a smartphone; and means for proposing recipes that suit the user's wishes using a generative AI model. This meets the diverse needs of consumers, enabling efficient grocery shopping and healthy eating.

[0967] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[0968] "Information on ingredients arriving at supermarkets" refers to information on the arrival of ingredients at supermarkets, such as the arrival date, amount of each product, and sales status.

[0969] "AI learning" is the learning process that enables artificial intelligence to recognize patterns based on large amounts of data and make predictions and judgments.

[0970] "Customer preferences" refers to information about the types of ingredients and dishes that a particular customer prefers.

[0971] "Health conditions" refers to information about the health of a particular customer, including, for example, allergies and restrictions on the intake of certain nutrients.

[0972] "Seasonal ingredients" are ingredients that are considered to be most delicious and nutritious during a particular season.

[0973] "Nutritional balance" refers to a diet in which the nutrients necessary to maintain health are properly distributed.

[0974] "Means for recommending recipes" refers to a method of suggesting optimal cooking recipes that take into consideration the customer's preferences, health condition, seasonal ingredients, and nutritional balance.

[0975] "Demand forecasting" is an analytical method for predicting future demand for a product.

[0976] "A means of providing a seamless service" is a method of consistently providing a series of processes from demand forecasting to recommendations and purchase.

[0977] An "application installed on a smartphone" is a software program that runs on a smartphone.

[0978] A "generative AI model" is a model that allows artificial intelligence to generate new information and suggestions based on data.

[0979] A "prompt" is an instruction given to a generative AI model, serving as a guideline for generating specific information or suggestions.

[0980] The following system configuration will be described as an embodiment of the present invention.

[0981] System Configuration

[0982] The system includes a server, the user's smartphone, and a generative AI model. The server collects daily information on food market trends and ingredients arriving at supermarkets, and uses this information for AI learning. A dedicated application is installed on the user's smartphone, which recommends recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance.

[0983] Program processing

[0984] The server processes the data using the following means:

[0985] 1. Data collection methods:

[0986] The server collects daily food market trend data (price fluctuations, consumer purchasing habits, seasonal influences, etc.) and supermarket inventory information (arrival date, amount, sales status, etc.). This data is obtained using APIs and web scraping technology.

[0987] 2. AI learning methods:

[0988] The server performs AI learning based on the collected data. Specifically, it uses Python machine learning libraries (e.g., TensorFlow, PyTorch) to build food demand forecasting models and recipe recommendation models.

[0989] 3. Recipe recommendation methods:

[0990] The server recommends optimal recipes taking into consideration the user's preferences, health status, seasonal ingredients, and nutritional balance. Recipe information is provided through an application installed on the user's smartphone.

[0991] 4. Generative AI model means:

[0992] The server uses a generative AI model to suggest recipes that match the user's preferences, and the generative AI model generates new recipes based on the user's input (prompt sentence).

[0993] Hardware and software used

[0994] Hardware: Servers, smartphones

[0995] Software: Python, TensorFlow, PyTorch, dedicated applications

[0996] Specific examples

[0997] For example, if a user wants "low-carb recipes," they can input the following prompt into the generative AI model:

[0998] Example prompt sentence:

[0999] "The user wants low-carb recipes. Please recommend the best recipes based on market trend data and supermarket stock information."

[1000] By inputting this prompt into a generative AI model, the AI ​​will suggest recipes that match the user's preferences. The recipe information is provided through an application installed on the user's smartphone, and the user can purchase ingredients and cook based on the recipe.

[1001] This will meet the diverse needs of consumers, enable efficient grocery shopping, and encourage healthy eating habits.

[1002] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1003] Step 1:

[1004] The server collects food market trend data and supermarket stock information daily. Specifically, it uses API and web scraping technology to obtain data such as price fluctuations, consumer purchasing trends, seasonal influences, stock arrival dates, stock quantities, and sales status. The input is market trend data and stock arrival information, and the output is the collected data.

[1005] Step 2:

[1006] The server performs AI training based on the collected data. It uses Python machine learning libraries (e.g., TensorFlow, PyTorch) to build food demand forecasting models and recipe recommendation models. The input is the collected market trend data and stock information, and the output is a trained AI model.

[1007] Step 3:

[1008] Users input their preferences, health status, and desired recipe conditions through a dedicated application installed on their smartphone. The input is the user's preferences, health status, and desired recipe conditions, and the output is this information.

[1009] Step 4:

[1010] The server receives user input and uses an AI model to recommend optimal recipes. Specifically, it filters recipes and selects the most suitable one, taking into consideration the user's preferences, health status, seasonal ingredients, and nutritional balance. The input is the user's preferences, health status, and desired recipe conditions, and the output is the recommended recipe.

[1011] Step 5:

[1012] The server uses a generative AI model to generate a new recipe that matches the user's preferences. The prompt sentence is input into the generative AI model, which generates a recipe based on the user's preferences. The input is the prompt sentence, and the output is the generated recipe.

[1013] Step 6:

[1014] The server sends the recommended recipes and generated recipes to the user's smartphone. The user can review and select the suggested recipes through the smartphone application. The input is the recommended recipes and generated recipes, and the output is the recipe displayed on the user's smartphone.

[1015] Step 7:

[1016] The user purchases ingredients and cooks based on the selected recipe. A shopping list and cooking instructions are provided through a smartphone application. The input is the selected recipe, and the output is the shopping list and cooking instructions.

[1017] This will meet the diverse needs of consumers, enable efficient grocery shopping, and encourage healthy eating habits.

[1018] Example 2

[1019] Next, a description will be given of 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."

[1020] Conventional food purchasing systems have problems such as difficulty in proposing recipes based on customer preferences and health status, and the entire process from demand forecasting and purchasing to delivery is not carried out efficiently.Furthermore, there are issues such as the reduction of food waste and the lack of shopping support for elderly households and households with children.

[1021] 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.

[1022] In this invention, the server includes a means for recommending recipes using a generative AI model, a means for creating a list of ingredients based on the recommended recipe, a means for forecasting demand based on the created list of ingredients, a means for making purchase suggestions to customers based on the results of the demand forecast, and a means for arranging delivery of ingredients that customers have decided to purchase. This allows for recipe suggestions based on customers' preferences and health status, and efficiently executes the entire process from demand forecasting to purchase and delivery. It also helps reduce food waste and provides shopping support for elderly households and households with children.

[1023] A "generative AI model" is an algorithm that uses artificial intelligence to generate optimal answers or suggestions for a specific task.

[1024] "Recipe recommendation means" is a feature that uses a generative AI model to suggest cooking recipes based on the user's preferences and health status.

[1025] The "means for creating an ingredient list" is a function for creating a list of ingredients required for a recommended recipe.

[1026] "Means for demand forecasting" is a function that predicts future demand for food ingredients based on past data and current market trends.

[1027] The "means for making purchasing suggestions" is a function that suggests appropriate food ingredients to customers to purchase based on the results of demand forecasting.

[1028] The "means for arranging delivery" is a function for carrying out procedures for delivering ingredients that a customer has decided to purchase to a specified location.

[1029] "Food waste contribution" is an initiative that aims to reduce food waste and consume it efficiently.

[1030] "Shopping support for elderly households and households with children" is a service that helps elderly people and households with children to make shopping easier.

[1031] "Customer attributes" refers to personal information such as a customer's family structure, age, and preferences.

[1032] "Means of creating a comprehensive platform" refers to the ability to integrate multiple services and functions and provide them on a single platform.

[1033] This invention is a system that uses a generative AI model to recommend recipes and provides a seamless service from demand forecasting to purchasing and delivery. Specific embodiments of this system are described below.

[1034] Server Processing

[1035] The server uses a generative AI model to recommend recipes. Specifically, it uses Python to send prompts to a generative AI model (e.g., GPT-4) to generate recipes. The generated recipes are returned to the server in JSON format.

[1036] The server then creates an ingredient list based on the recommended recipe by running an SQL query to retrieve information about the ingredients in the recipe from a database and generating the ingredient list.

[1037] The server then performs demand forecasting based on the created ingredient list. It uses a machine learning model (e.g., a time series forecasting model) to predict ingredient demand and stores the forecast results in a database.

[1038] Furthermore, the server makes purchase proposals to customers based on the results of the demand forecast, generating the contents of the purchase proposals and sending them to the terminal via the REST API.

[1039] Finally, the server arranges for the delivery of the ingredients that the customer has decided to purchase by calling the API of the delivery service, arranging the delivery, and sending the delivery information to the terminal.

[1040] Terminal handling

[1041] The device receives the recommended recipes and ingredient lists sent from the server. When the user decides to purchase something through the device, the device sends that information to the server. The device then displays the recommended recipes and ingredient lists to the user using an iOS or Android application. The device also learns the user's purchase history and preferences, collecting data to reflect in future recommendations.

[1042] User operations

[1043] The user checks the recommended recipes through the device and purchases the necessary ingredients. Once the user decides to purchase, the device sends the information to the server, and the ingredients are delivered to the user using a delivery service. The user can manage their purchase history and preferences through the application and reflect them in the next recommendation.

[1044] Specific examples

[1045] An example of a prompt to be input to the generative AI model is, "Please suggest recipes for healthy meals. Preferably with ingredients that are easy to obtain and easy to prepare." Based on this prompt, the generative AI model recommends recipes such as "grilled chicken salad" and "soup with lots of vegetables." The server then creates a list of ingredients needed based on these recipes and performs demand forecasting. Users can check these recipes through their devices and purchase the necessary ingredients. The purchased ingredients are then delivered to the user using a delivery service.

[1046] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1047] Step 1:

[1048] The server inputs a prompt into the generative AI model to recommend recipes. Specifically, it runs a Python script and sends the prompt, "Please suggest recipes for healthy meals. Preferably ones that use easily available ingredients and are easy to prepare," to the generative AI model. The generative AI model generates a recipe based on this prompt and returns it to the server in JSON format.

[1049] Input: prompt statement

[1050] Output: Recommended recipes (JSON format)

[1051] Specific operations: Execute Python script, send prompts to generative AI model, receive recipes

[1052] Step 2:

[1053] The server creates a list of ingredients based on the recommended recipe. The server executes an SQL query to retrieve information about the ingredients in the recipe from a database and generate the list of ingredients.

[1054] Input: Recommended recipes (JSON format)

[1055] Output: Ingredient list (JSON format)

[1056] Specific operations: Execute SQL queries, retrieve ingredient information from the database, and generate an ingredient list.

[1057] Step 3:

[1058] The server performs demand forecasting based on the created ingredient list. It uses a machine learning model (e.g., a time series forecasting model) to predict the demand for ingredients and stores the prediction results in a database.

[1059] Input: Ingredient list (JSON format)

[1060] Output: Demand forecast results (saved in database)

[1061] Specific operations: Execution of machine learning model, demand forecast, saving of forecast results to database

[1062] Step 4:

[1063] The server makes purchase proposals to customers based on the results of the demand forecast, generates the contents of the purchase proposals, and sends them to the terminal via the REST API.

[1064] Input: Demand forecast results

[1065] Output: Purchase offer (sent to device)

[1066] Specific operation: Generates purchase proposals and sends them to the device via REST API

[1067] Step 5:

[1068] The user checks the purchase suggestions through the terminal, selects the ingredients they need, and decides to purchase them. The terminal then sends this information to the server.

[1069] Input: Purchase Proposal

[1070] Output: Purchase decision information (sent to server)

[1071] Specific operations: Confirm purchase offer on device, tap purchase button, send purchase information to server

[1072] Step 6:

[1073] The server receives the user's purchase information and arranges for delivery by calling the delivery service's API, arranging delivery, and sending the delivery information to the terminal.

[1074] Input: Purchase decision information

[1075] Output: Delivery information (sent to terminal)

[1076] Specific operations: Calling the delivery service API, arranging delivery, sending delivery information to the terminal

[1077] (Application example 2)

[1078] 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."

[1079] Modern consumers spend a lot of time in their busy daily lives purchasing ingredients and selecting recipes. Food waste and food loss have become social issues, placing a heavy burden on elderly households and those with children. Furthermore, it is difficult to provide meals tailored to individual tastes and health conditions. To solve these problems, an efficient and personalized food purchasing and delivery system is needed.

[1080] 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.

[1081] In this invention, the server includes a means for performing machine learning on a daily basis to learn about food market trends and ingredients available at retailers, a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing a seamless service from demand forecasting and machine learning recommendations to purchasing, and a means for creating recipe-based ingredient lists and seamless purchasing and delivery via an application installed on a smartphone. This allows consumers to efficiently purchase ingredients and easily prepare meals tailored to their individual needs. It also reduces food waste and provides shopping support for elderly households and those with children.

[1082] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[1083] "Retail store" means a store that sells goods directly to consumers.

[1084] "Information on newly arrived ingredients" refers to information on the types and quantities of ingredients newly arrived at retail stores and markets.

[1085] "Daily" means something that happens every day.

[1086] "Machine learning" refers to a technology that allows computers to analyze data and learn patterns and rules.

[1087] "Customer preferences" refers to the types of ingredients and dishes that customers prefer.

[1088] "Health Status" refers to the state of a Customer's physical health.

[1089] "Seasonal ingredients" refer to ingredients that are at their most delicious and nutritious during that season.

[1090] "Nutritional balance" refers to the balance of nutrients contained in a meal.

[1091] A "recipe" refers to how to prepare a dish or how to combine ingredients.

[1092] "Recommendation" refers to the recommendation of a particular option over other options.

[1093] "Demand forecasting" refers to predicting future demand.

[1094] "Through" means to do something consistently from start to finish.

[1095] A "smartphone" refers to a mobile phone that can connect to the Internet and use applications.

[1096] "Application" refers to software that performs a specific function.

[1097] "Ingredient list" refers to a list of ingredients needed for a dish.

[1098] "Purchase" refers to the purchase of a product.

[1099] "Delivery" means the delivery of the goods to the designated location.

[1100] "Food waste" refers to ingredients and food that are thrown away even though they are still edible.

[1101] An "elderly household" refers to a household in which an elderly person primarily lives.

[1102] "Child-rearing households" refers to households raising children.

[1103] "Shopping assistance" refers to assistance with shopping.

[1104] "Customer attributes" refer to characteristics such as a customer's family structure, age, and preferences.

[1105] A "comprehensive platform" refers to a foundation that provides multiple functions and services in a single system.

[1106] A system for implementing this invention includes a server, a smartphone, and an internet connection. The server has a means for performing machine learning on a daily basis to acquire information on food market trends and ingredients received at retail stores, etc. This allows the system to constantly grasp the latest market information and use it to forecast demand.

[1107] The server also has the means to recommend recipes that take into account the customer's preferences, health condition, seasonal ingredients, and nutritional balance, allowing it to provide the best recipes for each individual customer. Based on the recommended recipe, it creates a list of necessary ingredients and presents it to the customer through an application installed on their smartphone.

[1108] Customers can use a smartphone application to purchase ingredients based on the recommended recipes. The application has a means to complete the purchasing and delivery procedures in one go, allowing customers to purchase the ingredients they need without any hassle and have them delivered to their home.

[1109] For example, if a user searches for "pasta recipes" on a smartphone app, the server uses AI to recommend "pasta with tomato sauce." This recipe requires "tomatoes, pasta, olive oil, and garlic." The application lists these ingredients and automatically places an order for delivery to the user's address.

[1110] An example of a prompt for the generative AI model is as follows:

[1111] User ID: 12345

[1112] Recipe search: Pasta

[1113] When this prompt sentence is input into the AI ​​recommendation model, the server will return the following response:

[1114] "Pasta with tomato sauce" requires "tomatoes, pasta, olive oil, and garlic." Based on this, the application handles the purchasing and delivery of ingredients in a single step.

[1115] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1116] Step 1:

[1117] The server collects daily information on food market trends and incoming ingredients from retailers and other sources, and inputs this information into a machine learning model. This allows for demand forecasts based on the latest market information. The inputs are market trend data and incoming ingredient information, and the output is demand forecast data. Specifically, the system obtains market data via an API and analyzes it using a machine learning algorithm.

[1118] Step 2:

[1119] The server inputs recipes based on the customer's preferences, health status, seasonal ingredients, and nutritional balance into a generative AI model to generate recommended recipes. The input is customer profile data and seasonal ingredient information, and the output is recommended recipes. Specifically, the server retrieves information from the customer database and inputs prompt statements into the generative AI model.

[1120] Step 3:

[1121] The server creates a list of ingredients based on the recommended recipe and sends it to the smartphone application. The input is the recommended recipe and the output is the ingredient list. Specifically, the server extracts ingredient information from the recipe data and converts it into a list format.

[1122] Step 4:

[1123] The user uses a smartphone application to check the ingredients based on the recommended recipe and then complete the purchase procedure. The input is the ingredient list, and the output is the purchase request. Specifically, the user displays the ingredient list through the application interface and presses the purchase button.

[1124] Step 5:

[1125] The server receives the user's purchase request and starts the delivery process. The input is the purchase request and the output is delivery instructions. Specifically, it calls the delivery service's API and sends delivery instructions.

[1126] Step 6:

[1127] The delivery service delivers ingredients to a specified address. The input is a delivery instruction, and the output is a delivery completion notification. Specifically, the service plans a delivery route and delivers the ingredients to the user's address.

[1128] Step 7:

[1129] The user receives the delivered ingredients and cooks the dish according to the recommended recipe. The input is the delivered ingredients, and the output is the completed dish. The specific operation is to cook the ingredients according to the recipe.

[1130] Example 3

[1131] Next, a description will be given of 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."

[1132] Food waste and food loss are major issues in modern society. Elderly households and those with children face the challenge of finding and purchasing the right ingredients for their health and lifestyle. Furthermore, there is a need for a system that efficiently utilizes ingredients owned by users and provides nutritionally balanced meals.

[1133] The identification process by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes a means for performing machine learning on a daily basis of food market trends and information on ingredients received from retailers, etc.; a means for recommending recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance; a means for providing an integrated service from demand forecasting and machine learning recommendations to purchasing; a means for inputting ingredient information owned by the user; a means for saving the input ingredient information in a database; a means for generating recipes using a generative AI model based on the saved ingredient information; and a means for transmitting the generated recipes and necessary ingredient lists to the user. This enables users to consume ingredients efficiently and reduce food waste. Furthermore, suggestions tailored to the special needs of elderly households and households with children are provided, supporting healthy eating habits.

[1134] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[1135] "Retailer" refers to a store or supermarket that sells food products to consumers.

[1136] "Information on newly arrived ingredients" refers to information such as the type, quantity, and arrival date of ingredients newly received at the store.

[1137] "Machine learning" refers to a technology in which a computer automatically learns from data and makes predictions and classifications.

[1138] "User" refers to an individual or household who uses this system.

[1139] "Preferences" refers to the types of ingredients and dishes that a user prefers.

[1140] "Health status" refers to the physical health and nutritional status of the user.

[1141] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[1142] "Nutritional balance" refers to a state in which necessary nutrients are contained in the appropriate proportions.

[1143] "Cooking method" refers to the method or recipe for making a dish using ingredients.

[1144] "Demand forecasting" refers to predicting future demand for food ingredients.

[1145] "Providing consistently" refers to providing multiple processes continuously and without interruption.

[1146] A "database" refers to a system for systematically storing and managing data.

[1147] A "generative AI model" refers to a model that uses artificial intelligence to generate new information or results from data.

[1148] "Ingredient list" refers to a list of ingredients required for a recipe.

[1149] "Food waste" refers to ingredients and food that are discarded without being consumed.

[1150] An "elderly household" refers to a household where elderly people primarily live.

[1151] "Child-rearing households" refers to households raising children.

[1152] "Special needs" refers to the unique wants and needs of particular users.

[1153] A "comprehensive platform" refers to a foundation that integrates and provides multiple functions and services.

[1154] This invention is a system for optimizing food consumption and reducing food waste. Specifically, users input information about ingredients they own, and a generative AI model is used to generate recipes based on that information and suggest the necessary ingredients to purchase. For elderly households and those with children, purchasing suggestions are also made taking into account their health and living conditions.

[1155] Hardware and software used

[1156] Device: A device such as a smartphone, computer, or tablet is used by the user to input ingredient information.

[1157] Server: A server is used to store data, process it, and run generative AI models. Servers are often built on cloud services (e.g., AWS, Google Cloud).

[1158] Database: Uses a relational database such as MySQL or PostgreSQL, which efficiently stores and manages the ingredient information entered by users.

[1159] Generative AI model: Uses a generative AI model such as GPT-4 to generate recipes based on the input ingredients.

[1160] Explanation of program processing

[1161] 1. The user enters ingredient information

[1162] Users use devices such as smartphones or computers to input information about ingredients they currently own. For example, they might enter information such as "Tomatoes, 3 pieces, 2023-10-15" or "Chicken, 500g, 2023-10-20" into the application's input form.

[1163] 2. The device sends the ingredient information to the server

[1164] The device sends the ingredient information entered by the user to the server as an HTTP request. At this time, the data is often sent in JSON format. For example, it is sent in the format {"Ingredients": "Tomato", "Quantity": "3", "Best before": "2023-10-15"}.

[1165] 3. The server saves the ingredient information in a database

[1166] The server analyzes the received ingredient information and stores it in a database. A relational database such as MySQL or PostgreSQL is used as the database. For example, an SQL query is used to store the information in the following format: INSERT INTO ingredient_table (ingredient, quantity, expiration date) VALUES ('tomatoes', '3', '2023-10-15').

[1167] 4. The server generates a recipe using the AI ​​model.

[1168] The server generates recipes using a generative AI model (e.g., GPT-4) based on the stored ingredient information. Specifically, the ingredient information is input into the AI ​​model as a prompt to generate a recipe. For example, the prompt might be, "Please generate a simple, nutritionally balanced recipe using the following ingredients: 3 tomatoes, 500g chicken."

[1169] 5. The server sends the recipe and ingredients list to the user.

[1170] The server sends the created recipe and a list of ingredients required for the recipe to the user's device. For example, it sends the HTTP response in the format {"Recipe": "Tomato and Chicken Stew", "Required Ingredients": ["Tomato", "Chicken"]}.

[1171] 6. Users purchase ingredients based on recipes

[1172] The user purchases the necessary ingredients based on the recipe and ingredient list sent from the server. For example, the user goes to a supermarket and purchases "3 tomatoes" and "500g chicken."

[1173] 7. The server makes special suggestions for elderly households and households with children.

[1174] The server makes special purchasing suggestions based on information about the user's health and living situation. For example, it might suggest "nutritiously balanced meals" to elderly households, and "recipes containing ingredients that children like" to households with children. Specifically, it generates recipes by inputting special prompts into the AI ​​model based on the user's profile information.

[1175] Specific examples

[1176] Example of ingredient information entered by the user

[1177] "Tomatoes, 3, 2023-10-15"

[1178] "Chicken, 500g, 2023-10-20"

[1179] Example prompts to be input to the generative AI model

[1180] "Make a simple, nutritionally balanced recipe using the following ingredients: 3 tomatoes, 500g chicken."

[1181] Example of a generated recipe

[1182] "Tomato and Chicken Stew: Finely chop the tomatoes and simmer them in a pot with the chicken. Add salt, pepper, and olive oil to taste."

[1183] In this way, users can consume ingredients efficiently and reduce food waste. Additionally, for elderly households and households with children, suggestions are made according to their special needs, which supports healthy eating habits. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[1184] Step 1:

[1185] The user inputs the ingredient information.

[1186] Users use devices such as smartphones or PCs to input information about ingredients they currently own. For example, they might input information such as "Tomatoes, 3 pieces, 2023-10-15" or "Chicken, 500g, 2023-10-20" into the application's input form. The input information includes the type of ingredient, quantity, expiration date, etc.

[1187] Step 2:

[1188] The terminal transmits the ingredient information to the server.

[1189] The terminal sends the ingredient information entered by the user to the server as an HTTP request. At this time, the data is often sent in JSON format. For example, it is sent in the format {"ingredient": "tomato", "quantity": "3", "best before": "2023-10-15"}. The input is the ingredient information entered by the user, and the output is the JSON format data sent to the server.

[1190] Step 3:

[1191] The server stores the ingredient information in a database.

[1192] The server analyzes the received ingredient information and stores it in a database. A relational database such as MySQL or PostgreSQL is used as the database. For example, an SQL query is used to store the information in the form of INSERT INTO ingredient table (ingredient, quantity, expiration date) VALUES ('tomatoes', '3', '2023-10-15'). The input is JSON-formatted data sent to the server, and the output is the ingredient information stored in the database.

[1193] Step 4:

[1194] The server generates recipes using AI models.

[1195] The server generates a recipe using a generative AI model (e.g., GPT-4) based on the stored ingredient information. Specifically, the ingredient information is input into the AI ​​model as a prompt to generate a recipe. For example, the prompt might be "Please generate a simple, nutritionally balanced recipe using the following ingredients: 3 tomatoes, 500g chicken." The input is the ingredient information stored in the database, and the output is the generated recipe.

[1196] Step 5:

[1197] The server sends the recipe and a list of ingredients to the user.

[1198] The server sends the generated recipe and the list of ingredients required for the recipe to the user's device. For example, it sends the HTTP response in the format {"Recipe": "Tomato and Chicken Stew", "Required Ingredients": ["Tomato", "Chicken"]}. The input is the generated recipe and the list of ingredients, and the output is the data sent to the user's device.

[1199] Step 6:

[1200] A user purchases ingredients based on a recipe.

[1201] The user purchases the necessary ingredients based on the recipe and ingredient list sent from the server. For example, you go to a supermarket and purchase "3 tomatoes" and "500g chicken." The input is the recipe and ingredient list sent from the server, and the output is the purchased ingredients.

[1202] Step 7:

[1203] The server makes special suggestions for elderly households and households with children.

[1204] The server makes special purchasing suggestions based on information about the user's health and lifestyle. For example, it suggests "nutritiously balanced meals" to elderly households, and "recipes containing ingredients that children like" to households with children. Specifically, it generates recipes by inputting special prompts into the AI ​​model based on the user's profile information. The input is information about the user's health and lifestyle, and the output is special suggestions.

[1205] (Application example 3)

[1206] 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."

[1207] In modern society, the problem of food waste is serious, and it is particularly difficult for elderly households and those with children to manage and purchase ingredients. Furthermore, there is a lack of suggestions for appropriate ingredients and recipes based on health status and living situations, making it difficult to eat a nutritionally balanced diet. Furthermore, there is little progress in making effective use of surplus ingredients, resulting in waste.

[1208] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes a means for performing machine learning on a daily basis of food market trends and information on ingredients received at retailers, etc., a means for recommending recipes that take into account the customer's preferences, health condition, seasonal ingredients, and nutritional balance, a means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchasing, a means for generating recipes to encourage the consumption of surplus ingredients, and a means for suggesting ingredients and recipes based on the customer's health condition and lifestyle. This makes it possible to reduce food waste and support shopping for elderly households and households with children.

[1209] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[1210] "Retail store" means a store that sells goods directly to consumers.

[1211] "Information on newly arrived ingredients" refers to information on ingredients that have newly arrived at retail stores or markets.

[1212] "Machine learning" refers to a technology in which computers analyze data, learn patterns, and make predictions and classifications.

[1213] "Customer preferences" refers to the types of ingredients and dishes that customers prefer, as well as their taste preferences.

[1214] "Health Status" refers to information about a customer's current health, such as blood pressure, blood sugar levels, allergies, etc.

[1215] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[1216] "Nutritional balance" refers to a state in which the nutrients necessary to maintain health are appropriately consumed.

[1217] A "recipe" is a set of instructions that describes how to prepare a dish.

[1218] "Recommendation" refers to suggesting the best option based on specific criteria.

[1219] "Demand forecasting" refers to predicting future demand.

[1220] "Through" means to do something consistently from start to finish.

[1221] "Excess ingredients" refers to ingredients that have been purchased but left unused.

[1222] "Encouraging consumption" refers to encouraging people to take a particular action.

[1223] "Lifestyle" refers to information about the customer's daily life, such as family composition and lifestyle habits.

[1224] The system for carrying out the present invention operates in cooperation with three entities: a server, a terminal, and a user. A specific embodiment of the system will be described below.

[1225] Server Processing

[1226] The server uses machine learning to learn daily information about food market trends and ingredients arriving at retail stores. This allows it to constantly grasp the latest market information and forecast demand. The server uses a generative AI model to recommend recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. It also generates recipes to encourage the consumption of surplus ingredients. This information is sent to the user's device.

[1227] Terminal handling

[1228] The device accepts user input and displays recipe and ingredient information sent from the server. Users can input their preferences, health status, and surplus ingredients into the device. The device then sends this information to the server and receives suggestions for optimal recipes and ingredients. Based on the recipe selected by the user, the device generates a shopping list of the necessary ingredients and works with the delivery service to support purchasing.

[1229] User operations

[1230] Users input their preferences, health status, and excess ingredients through their device. This allows them to receive optimal recipe and ingredient suggestions from the server. The user then checks the suggested recipes and creates a list for purchasing the ingredients they need. They can also purchase ingredients using a delivery service.

[1231] Hardware and software used

[1232] Hardware: Smartphone

[1233] Software: Python, OpenAI API, requests library

[1234] Specific examples

[1235] For example, if a user prefers "Japanese food, low calorie" and has a health condition of "high blood pressure," the following prompt sentences are used to generate recipes:

[1236] Example prompt sentence:

[1237] User preferences: Japanese food, low calorie

[1238] Health condition: High blood pressure

[1239] Based on this, please suggest the best recipe.

[1240] By inputting this prompt into a generative AI model, the system will suggest the best recipe for the user. The user can then review the suggested recipe, generate a shopping list for the necessary ingredients, and purchase the ingredients using a delivery service.

[1241] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1242] Step 1:

[1243] The server collects daily information on food market trends and incoming ingredients from retailers and other sources, and inputs this information into a machine learning model. This allows the system to learn the latest market information and perform demand forecasts. The input data is market trends and incoming ingredient information, and the output is the demand forecast results.

[1244] Step 2:

[1245] The user inputs their preferences, health status, and excess ingredients through the terminal. The input data is the user's preferences, health status, and excess ingredient information, and this information is sent to the server.

[1246] Step 3:

[1247] The server receives information on preferences, health status, and surplus ingredients sent by the user and generates the optimal recipe using a generative AI model. The input data are the user's preferences, health status, and surplus ingredient information, and the output is the optimal recipe.

[1248] Step 4:

[1249] The server sends the generated recipe to the terminal. The input data is the generated recipe, and the output is the recipe sent to the terminal.

[1250] Step 5:

[1251] The terminal displays the recipe sent from the server. The user checks the suggested recipe and generates a list of ingredients to purchase. The input data is the recipe sent from the server, and the output is the ingredient purchasing list.

[1252] Step 6:

[1253] The terminal then sends the generated food shopping list to the delivery service. The input data is the food shopping list, and the output is order information for the delivery service.

[1254] Step 7:

[1255] The delivery service delivers ingredients to users based on order information sent from the terminal. The input data is the order information, and the output is the delivery of ingredients to the user.

[1256] Step 8:

[1257] The user receives the delivered ingredients and prepares a dish based on the proposed recipe. The input data are the delivered ingredients and the recipe, and the output is the completed dish.

[1258] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1259] "Example 1"

[1260] One embodiment of the present invention is a system that incorporates an emotion engine. This system recognizes a user's emotions and recommends recipes based on those emotions. Specifically, when a user is feeling happy, the system recommends recipes that will amplify that joy, and when a user is feeling sad, the system recommends recipes that will alleviate that sadness. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, and the content of social media posts.

[1261] "Example 2"

[1262] Another embodiment of the present invention is a system that incorporates an emotion engine. This system recognizes the user's emotions and makes food purchase suggestions based on those emotions. Specifically, when the user is feeling stressed, the system suggests foods that will relieve stress, and when the user is relaxed, the system suggests foods that will help maintain that relaxed state. The emotion engine recognizes emotions from the user's biometric information (heart rate, blood pressure, etc.).

[1263] "Example 3"

[1264] Furthermore, another embodiment of the present invention is a system that incorporates an emotion engine. This system recognizes a user's emotions and provides delivery services according to those emotions. Specifically, when a user is feeling angry, a service that alleviates the anger (e.g., expedited delivery) is provided, and when a user is feeling happy, a service that amplifies the joy (e.g., including a surprise gift) is provided. The emotion engine recognizes emotions from the user's behavioral patterns and history data.

[1265] The processing flow of each embodiment will be described below.

[1266] "Example 1"

[1267] Step 1: Recognize emotions from the user's facial expressions, tone of voice, and social media posts.

[1268] Step 2: Depending on the emotion you recognize, decide whether to increase joy or reduce sadness.

[1269] Step 3: Based on the judgment, recommend an appropriate recipe.

[1270] "Example 2"

[1271] Step 1: Recognize emotions from the user's biometric information (heart rate, blood pressure, etc.).

[1272] Step 2: Depending on the emotion you identify, we'll suggest foods that will help you relieve stress or keep you relaxed.

[1273] Step 3: Based on your judgment, make appropriate food purchasing recommendations.

[1274] "Example 3"

[1275] Step 1: Recognize emotions from user behavior patterns and historical data.

[1276] Step 2: Depending on the emotion you identify, decide whether to offer a service that reduces anger or increases joy.

[1277] Step 3: Use your judgment to provide the appropriate delivery service.

[1278] Example 1

[1279] 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."

[1280] In today's food market, there is a demand for recipes that meet diverse consumer needs, reducing food waste, and providing shopping support for elderly households and those with children. However, to meet these needs, a system is needed that can efficiently collect information on food market trends and retail store stock, and provide recipes that match consumers' preferences, health status, and emotions. Conventional systems have difficulty meeting all of these requirements in a comprehensive manner, and this has led to the issue of not being able to increase consumer satisfaction.

[1281] 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.

[1282] In this invention, the server includes a means for collecting daily information on food market trends and ingredients arriving at retail stores, a means for training an AI model based on the collected data, a means for recognizing user emotions and recommending recipes based on those emotions, a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, and a means for providing a seamless service from demand forecasting and AI recommendations to purchasing. This makes it possible to provide recipes that meet the diverse needs of consumers, reduce food waste, and provide shopping support for elderly households and households with children.

[1283] "Food market trends" refers to information such as fluctuations in food prices in the market, consumer purchasing trends, and seasonal influences.

[1284] "Information on ingredients arriving at retail stores" refers to information such as the arrival date, quantity, and sales status of each product at a retail store.

[1285] An "AI model" refers to an artificial intelligence model built using a machine learning framework that learns from specific data and makes predictions and classifications.

[1286] "User emotions" refers to the emotional state recognized from the user's facial expressions, tone of voice, and the content of their social media posts.

[1287] "Means for recommending recipes" refers to a function that suggests recipes suitable for the user based on collected data and the learning results of the AI ​​model.

[1288] "Demand forecasting" refers to predicting future demand for a product based on collected data.

[1289] "A means of providing a seamless service" refers to a function that consistently provides a series of processes from data collection to recipe recommendations and purchase.

[1290] "Food waste contribution" refers to reducing food waste and promoting sustainable consumption.

[1291] "Shopping support for elderly households and households with children" refers to a function that helps elderly people and households with children to shop more conveniently.

[1292] "Customer attributes" refers to individual information about customers, such as family composition, age, and preferences.

[1293] A "comprehensive platform" refers to a system that integrates multiple functions and services and provides them to users in a centralized manner.

[1294] MODE FOR CARRYING OUT THE INVENTION

[1295] This invention is a system that collects daily information on food market trends and ingredients arriving at retail stores, and uses an AI model to recommend recipes based on customer preferences, health status, and emotions. Specific embodiments of this system are described below.

[1296] Server Roles

[1297] The server collects daily information on food market trends and food ingredients arriving at retail stores. Specifically, the server obtains food price fluctuation data, consumer purchasing trend data, and seasonal influence data via API. It also downloads CSV files from the FTP server to collect arrival information from each retail store. This data is stored in a database such as MySQL or PostgreSQL.

[1298] Next, the server preprocesses the collected data. Specifically, it complements missing values, normalizes the data, and detects and corrects outliers. Based on the preprocessed data, an AI model is trained using a machine learning framework such as TensorFlow or PyTorch. Once trained, the AI ​​model generates recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[1299] Device Role

[1300] The device uses a camera and microphone to recognize the user's emotions. Specifically, the camera captures the user's facial expressions and the microphone records the tone of voice. With the user's permission, the device also accesses their social media accounts and analyzes their posts. This data is sent to the emotion engine, which uses TensorFlow to analyze the facial and voice data and identify the user's emotions.

[1301] User Roles

[1302] The user receives recipe recommendations through their device. For example, if the user is feeling happy, the system will recommend recipes that will amplify that joy. Specifically, if the user is smiling, the system will display a recipe for a "fruit parfait." Conversely, if the user is feeling sad, the system will display a recipe for a "heartwarming chicken soup."

[1303] Specific examples

[1304] Examples of data collection

[1305] The server retrieves food price fluctuation data from "https: / / api.marketdata.com / prices", authenticates using an API key, and receives the data in JSON format.

[1306] Retail store arrival information is obtained by downloading a CSV file from the FTP server.

[1307] Specific examples of data preprocessing

[1308] The server imputes missing values ​​with the mean and normalizes the data, e.g., scaling price data to a range of 0 to 1.

[1309] To detect outliers, we use Z scores, with data exceeding 3 standard deviations being treated as outliers.

[1310] Specific examples of emotion recognition

[1311] The device captures the user's facial expressions using OpenCV and sends them to the emotion engine, which uses TensorFlow to analyze the facial expression data and identify the user's emotions.

[1312] Example of recipe recommendation

[1313] If the user is feeling happy, the system displays a recipe for a "fruit parfait," detailing the ingredients needed and how to make it.

[1314] Prompt Sentence Examples

[1315] "Build an AI model that can recommend recipes based on market trend data and retail store stock information, taking into account customer preferences and health conditions. Also, combine it with a system that recognizes user emotions and recommends recipes based on those emotions."

[1316] In this way, by clarifying the roles of the server, terminal, and user, and explaining the specific processing content and the hardware and software used, it becomes easier to understand the overall picture of the system.

[1317] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1318] Step 1:

[1319] The server collects food market trend data and information on ingredients arriving at retail stores. Specifically, it obtains food price fluctuation data, consumer purchasing trends, and seasonal influence data via API. It also downloads CSV files from the FTP server to collect arrival information from each retail store. This data is stored in a database such as MySQL or PostgreSQL.

[1320] Input: Market trend data obtained from API, stock information downloaded from FTP server

[1321] Output: Market trend data and stock information stored in a database

[1322] Step 2:

[1323] The server preprocesses the collected data by imputing missing values, normalizing the data, and detecting and correcting outliers. For example, it imputes missing values ​​with the mean and scales price data to a range of 0 to 1. For outlier detection, it uses Z-scores and treats data that is greater than 3 standard deviations as an anomaly.

[1324] Input: Market trend data and stock information stored in the database

[1325] Output: Preprocessed data

[1326] Step 3:

[1327] The server trains an AI model based on the preprocessed data. TensorFlow and PyTorch are used as machine learning frameworks. Once trained, the AI ​​model generates recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[1328] Input: Preprocessed data

[1329] Output: Trained AI model

[1330] Step 4:

[1331] The device uses a camera and microphone to recognize the user's emotions. Specifically, the camera captures the user's facial expressions and the microphone records the user's tone of voice. With the user's permission, the device also accesses their social media accounts and analyzes their posts. This data is then sent to the emotion engine.

[1332] Input: facial expression data captured by the camera, audio data recorded by the microphone, and social media posts

[1333] Output: Data sent to the emotion engine

[1334] Step 5:

[1335] The emotion engine analyzes the collected data and identifies the user's emotions. Specifically, it uses TensorFlow to analyze facial expression and voice data, recognizing "joy" when the user is smiling and "sadness" when the user is crying.

[1336] Input: Data sent to the emotion engine

[1337] Output: Identified user sentiment

[1338] Step 6:

[1339] The user receives recipe recommendations through their device. For example, if the user is feeling happy, the system will recommend recipes that will amplify that joy. Specifically, if the user is smiling, the system will display a recipe for a "fruit parfait." Conversely, if the user is feeling sad, the system will display a recipe for a "heartwarming chicken soup."

[1340] Input: Identified user sentiment

[1341] Output: Recommended recipes

[1342] (Application example 1)

[1343] 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."

[1344] Modern consumers have diverse needs when it comes to grocery shopping. In particular, they want to be able to select ingredients based on their health-conscious or personal preferences, receive recipe suggestions using seasonal ingredients, and even have meal suggestions tailored to their emotions. However, meeting these needs requires a system that can process vast amounts of information in real time and make appropriate recommendations. There is also a demand for smartphone applications to improve the shopping experience in brick-and-mortar stores. The lack of a comprehensive system to solve these challenges presents a problem.

[1345] 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.

[1346] In this invention, the server includes a means for performing machine learning on a daily basis on food market trends and information on ingredients in stock at retailers, a means for recommending recipes that take into account a customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchase, a means for recognizing a user's emotions and recommending recipes according to those emotions, and a means for providing a smartphone application for improving the shopping experience in physical stores. This enables recommendations that meet the diverse needs of consumers and also improves the shopping experience in physical stores.

[1347] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[1348] A "retail store" is a store that sells goods directly to consumers.

[1349] "Incoming food information" is information about food ingredients arriving at the store, such as the arrival date, amount of each product, and sales status.

[1350] "Machine learning" is a technology that allows computers to automatically learn from data and make predictions and classifications.

[1351] "Customer preferences" refers to information about the preferences of individual customers, such as the ingredients and types of dishes that customers prefer, and their taste preferences.

[1352] "Health Status" refers to information about a customer's current health, including, for example, allergies or specific nutritional needs.

[1353] "Seasonal ingredients" are ingredients that are considered to be at their most delicious and nutritious during that season.

[1354] "Nutritional balance" refers to the appropriate balance of nutrients in a diet.

[1355] A "recipe" is a list of steps and ingredients for making a particular dish.

[1356] "Recommendation" means suggesting the best option based on specific criteria.

[1357] "Demand forecasting" is the process of predicting future demand for a product.

[1358] "Through" means to do something consistently from start to finish.

[1359] "User emotion" refers to the emotional state that the user is feeling, such as joy, sadness, or anger.

[1360] A "brick and mortar store" is a store that exists in a physical location and where consumers visit in person to purchase products.

[1361] A "smartphone application" is a software program that runs on a smartphone.

[1362] A system for implementing this invention is configured as follows: A server collects daily information on food market trends and food ingredients arriving at retail stores, and inputs this data into a machine learning model for learning. Specifically, the server collects market trend data such as food price fluctuations, consumer purchasing habits, and seasonal influences, and collects retail store arrival information such as the arrival date, arrival volume, and sales status of each product.

[1363] The server uses this data to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. Furthermore, to recognize the user's emotions, the server receives input from the user's facial expressions, tone of voice, and social media posts, and uses an emotion engine to recognize their emotions. Based on this emotional information, the server recommends recipes that match the user's emotions.

[1364] The server then provides the generated recipes to users through a smartphone application. This application is designed to enhance the shopping experience in brick-and-mortar stores, allowing users to receive real-time recipe recommendations in-store. It also helps users efficiently purchase the ingredients they need based on demand forecasts.

[1365] The hardware used includes a server and a smartphone. The software used includes Python, a requests library, and an emotion recognition library. The server uses this software to collect data, train machine learning models, recognize emotions, and generate recipes.

[1366] As a concrete example, the following prompt sentences can be used to recognize the user's emotions and recommend appropriate recipes:

[1367] Example prompt sentence:

[1368] "The user's facial expression is smiling, their voice tone is cheerful, and their social media post states, 'Today was such a fun day!'"

[1369] Based on this prompt, the server can recognize that the user is feeling joy and recommend recipes (e.g., party dishes or desserts) that will enhance that joy.

[1370] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1371] Step 1:

[1372] The server collects food market trend data. Specifically, it obtains market trend data (price fluctuations, consumer purchasing trends, seasonal influences, etc.) through API. The input is market trend data from the API, and the output is the collected market trend data.

[1373] Step 2:

[1374] The server collects information on incoming ingredients from retail stores. Specifically, it obtains data such as the arrival date, quantity, and sales status of each product through the API. The input is the incoming ingredient information from the API, and the output is the collected incoming ingredient information.

[1375] Step 3:

[1376] The server inputs the collected market trend data and information on incoming ingredients into a machine learning model for learning. Specifically, it preprocesses the data using Python and machine learning libraries and inputs it into the model. The input is market trend data and information on incoming ingredients, and the output is the trained model.

[1377] Step 4:

[1378] The server generates recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance. Specifically, it uses a trained model to generate recipes based on the user's profile information. The input is the user's profile information, and the output is the generated recipe.

[1379] Step 5:

[1380] The server recognizes the user's emotions. Specifically, it receives the user's facial expressions, tone of voice, and social media posts as input, and uses an emotion engine to recognize emotions. The input is the user's emotional data, and the output is the recognized emotional information.

[1381] Step 6:

[1382] The server recommends recipes based on the recognized emotional information. Specifically, the emotional information is input into a trained model to generate an appropriate recipe. The input is the recognized emotional information, and the output is a recipe based on the emotion.

[1383] Step 7:

[1384] The server provides the generated recipe to the user through a smartphone application. Specifically, it sends recipe information to the application so that the user can check the recipe in real time. The input is the generated recipe, and the output is the recipe information displayed on the user's smartphone.

[1385] Step 8:

[1386] Users use a smartphone application to enhance their shopping experience in a brick-and-mortar store by purchasing ingredients based on recipes recommended through the application. The input is recipe information from the smartphone application, and the output is the purchased ingredients.

[1387] Example 2

[1388] Next, a description will be given of 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."

[1389] In today's modern diet, it is difficult to select and purchase the right ingredients based on customers' preferences and health status. It is also difficult to grasp food market trends and retailer stock information in real time and forecast demand. Furthermore, there is a lack of systems that can suggest ingredients based on the customer's emotional state and efficiently carry out the entire process from purchase to delivery. To solve these issues, a comprehensive system is needed.

[1390] 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.

[1391] In this invention, the server includes a means for performing machine learning on a daily basis on food market trends and information on ingredients received at retailers, a means for recommending recipes that take into account a customer's preferences, health condition, seasonal ingredients, and nutritional balance, a means for providing an end-to-end process from demand forecasting and machine learning recommendations to purchasing, a means for analyzing recipes and generating a list of necessary ingredients, a means for collecting biometric information and analyzing emotions, a means for suggesting ingredients to purchase based on the customer's emotions, and a means for arranging delivery after the purchase is confirmed. This enables the selection and purchase of appropriate ingredients based on the customer's preferences, health condition, and emotional state, and allows the entire process from purchase to delivery to be carried out efficiently.

[1392] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[1393] "Retail store" refers to a store or business that sells products directly to consumers.

[1394] "Machine learning" refers to the technology that allows computers to learn patterns from data and make predictions and classifications.

[1395] "Preference" refers to an individual's tendency to prefer certain ingredients or dishes.

[1396] "Health status" refers to the state of an individual's physical and mental health.

[1397] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[1398] "Nutritional balance" refers to a state in which necessary nutrients are consumed in appropriate proportions.

[1399] A "recipe" is a list of steps or ingredients for making a particular dish.

[1400] A "recommendation" is a recommendation for a particular choice or action.

[1401] "Demand forecasting" refers to the process of predicting future demand.

[1402] "Through" means to do something consistently from start to finish.

[1403] "Recipe analysis" refers to the process of analyzing the contents of a recipe and extracting the necessary information.

[1404] "Ingredient list" refers to the list of ingredients needed to make a particular dish.

[1405] "Biometric information" refers to an individual's physical data, such as heart rate and blood pressure.

[1406] "Emotion analysis" refers to the process of analyzing an individual's emotional state based on biometric and other data.

[1407] "Purchase Offer" means an offer to purchase a particular product or service.

[1408] "Confirmation of purchase" refers to a customer's final decision to purchase a product or service.

[1409] "Delivery arrangements" refers to the procedures for delivering purchased products to customers.

[1410] MODE FOR CARRYING OUT THE INVENTION

[1411] This invention is a system that uses machine learning to learn daily information on food market trends and ingredients in stock at retail stores, recommends recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance, and provides a seamless service from demand forecasting to purchasing and delivery. Specific embodiments of this system are described below.

[1412] Hardware and software used

[1413] 1. Server

[1414] The server collects, analyzes, and stores data, and runs machine learning models, using the following software and frameworks:

[1415] Machine learning frameworks: TensorFlow, PyTorch

[1416] Database: MySQL, PostgreSQL

[1417] Natural language processing libraries: NLTK, spaCy

[1418] Emotion analysis API: IBM Watson, Microsoft Azure

[1419] 2. Terminal

[1420] The terminal provides an interface with the user and inputs and displays data. Specifically, the following devices and software are used:

[1421] Devices: Smartphones, tablets, computers

[1422] Applications: Web applications, mobile applications

[1423] 3. Users

[1424] Users use the system to purchase ingredients. They access the system through their terminal, check the suggested recipes and ingredients, and then decide on the purchase.

[1425] Program processing

[1426] 1. Recipe analysis

[1427] The server receives the recipe recommendations using the generative AI model and analyzes them using natural language processing techniques, which generates a list of the ingredients needed.

[1428] 2. Demand forecast

[1429] The server then uses the generated ingredient list to forecast demand, taking into account past purchase data, seasonality, and trends. Time series analysis is performed using Facebook Prophet and ARIMA models.

[1430] 3. Purchase proposal

[1431] The server then makes food purchasing suggestions to customers based on predicted demand, which are displayed on their devices via a web or mobile application.

[1432] 4. Biometric data collection and emotion analysis

[1433] The device collects the user's biometric information (heart rate, blood pressure, etc.) through sensors and sends it to the server. The server then uses an emotion engine to analyze the user's emotions and makes food purchase suggestions based on the analysis results.

[1434] 5. Purchase confirmation and delivery arrangements

[1435] The user checks the suggested ingredients and decides to purchase them. The purchase information is sent to the server, and the order is confirmed. The server then sends the order information to a delivery service, and the ingredients are delivered to the user.

[1436] Specific examples

[1437] For example, if the AI ​​recommends a recipe for "chicken curry," the server generates a list of ingredients, such as chicken, curry powder, onions, and tomatoes. The server uses past data to predict demand for these ingredients and suggests them to the user. Once the user decides to purchase them, the ingredients are delivered to the user via a delivery service.

[1438] Prompt Sentence Examples

[1439] "Generate a list of suggested foods for when the user is feeling stressed."

[1440] "Create a list of ingredients needed for the following recipe and forecast demand: Chicken curry."

[1441] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1442] Step 1: Recipe Analysis

[1443] The server receives recipes recommended using a generative AI model. As input, it receives the recipe's text data. The server analyzes the recipe content using natural language processing techniques (e.g., NLTK or spaCy) and generates a list of required ingredients. As output, it obtains a list of ingredients. This list includes the name and required quantity of each ingredient. Specifically, it tokenizes the recipe text and extracts the ingredient names and their quantities.

[1444] Step 2: Demand forecast

[1445] The server performs demand forecasting based on the ingredient list generated in step 1. As input, it receives the ingredient list, past purchase data, seasonal data, and trend data. The server uses a time series analysis model to predict the demand for each ingredient. As output, it obtains the predicted demand amount. Specifically, it inputs past data into the time series model to predict future demand.

[1446] Step 3: Purchase proposal

[1447] The server makes food purchase suggestions to customers based on the demand predicted in step 2. The inputs are the predicted demand, customer preference data, and health status data. The server integrates these data and generates optimal purchase suggestions. The output is a list of purchase suggestions. This list is displayed on the device via a web or mobile application. Specifically, the suggestion list is filtered based on the customer's preferences and health status to select the most suitable ingredients.

[1448] Step 4: Biometric data collection and emotion analysis

[1449] The device collects the user's biometric information (heart rate, blood pressure, etc.) through sensors. The biometric data is given as input. The device sends this data to a server. The server uses an emotion engine (e.g., IBM Watson or Microsoft Azure emotion recognition API) to analyze the user's emotions. The output is the user's emotional state. Specifically, the biometric information is input into the emotion engine, and the emotional state is analyzed.

[1450] Step 5: Emotion-based purchase recommendations

[1451] The server makes food purchase suggestions based on the emotional state analyzed in step 4. As input, the emotional state data and a list of ingredients are given. The server selects ingredients according to the emotional state and generates a list of purchase suggestions. As output, a list of purchase suggestions based on emotions is obtained. This list is displayed on the terminal. Specifically, the server filters ingredients according to the emotional state and makes the most suitable suggestions.

[1452] Step 6: Confirm your purchase and arrange delivery

[1453] The user checks the suggested ingredients and decides to purchase them. Purchase confirmation information is provided as input. The terminal sends this information to the server. The server receives the purchase information and confirms the order. Confirmed order information is obtained as output. The server sends this information to the delivery service and arranges for delivery. Specifically, the order information is entered into the delivery service's system and the delivery procedure begins.

[1454] (Application example 2)

[1455] 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."

[1456] Conventional food purchasing systems have problems such as difficulty in suggesting ingredients that match the customer's preferences, health condition, and emotions, and the entire process from purchase to delivery is not efficient.Furthermore, there are issues such as insufficient efforts to reduce food waste and insufficient shopping support for elderly households and households with children.

[1457] 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.

[1458] In this invention, the server includes a means for performing machine learning on a daily basis to acquire information on food market trends and ingredients available at retailers, a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing a seamless service from demand forecasting and machine learning recommendations to purchasing, a means for acquiring biometric information and analyzing it with an emotion recognition engine, a means for suggesting ingredients and recipes based on emotions, and a means for purchasing the suggested ingredients and delivering them to the customer using a delivery service. This allows for the recommendation of ingredients based on the customer's preferences, health status, and emotions, and efficiently carries out the entire process from purchase to delivery. It also enables the reduction of food waste and shopping support for elderly households and households with children.

[1459] "Food market trends" refers to fluctuations and trends in supply and demand in the food market.

[1460] "Retail store" means a store that sells goods directly to consumers.

[1461] "Information on newly arrived ingredients" refers to information on ingredients that have newly arrived at retail stores or markets.

[1462] "Daily" means something that happens every day.

[1463] "Machine learning" refers to a technology in which computers analyze data and learn patterns to automatically improve their performance.

[1464] "Customer preferences" refers to the types of ingredients and dishes that customers prefer.

[1465] "Health Status" refers to the state of a Customer's physical health.

[1466] "Seasonal ingredients" refer to ingredients that are at their most delicious and nutritious during that season.

[1467] "Nutritional balance" refers to the balance of nutrients contained in a meal.

[1468] A "recipe" refers to how to prepare a dish or how to combine ingredients.

[1469] "Recommendation" refers to suggesting the best option based on specific criteria.

[1470] "Demand forecasting" refers to predicting future demand.

[1471] "Through" means to do something consistently from start to finish.

[1472] "Biometric information" refers to physical data such as heart rate and blood pressure.

[1473] An "emotion recognition engine" refers to technology that analyzes biometric information to recognize a user's emotions.

[1474] "Recommendation" refers to showing the best options based on specific criteria.

[1475] "Delivery service" refers to a service that delivers purchased products to customers.

[1476] "Food waste reduction" refers to reducing food waste.

[1477] An "elderly household" refers to a household in which an elderly person primarily lives.

[1478] "Child-rearing households" refers to households raising children.

[1479] "Shopping assistance" refers to providing support when shopping.

[1480] "Customer attributes" refer to characteristics such as a customer's family structure, age, and preferences.

[1481] A "comprehensive platform" refers to a foundation that provides multiple functions and services in a single system.

[1482] A system for carrying out this invention is configured as follows: The server includes means for performing machine learning on a daily basis information on food market trends and ingredients received at retailers, etc., means for recommending recipes that take into consideration the customer's preferences and health status, seasonal ingredients, and nutritional balance, means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchasing, means for acquiring biometric information and analyzing it with an emotion recognition engine, means for suggesting ingredients and recipes according to emotions, and means for purchasing the suggested ingredients and delivering them to the customer using a delivery service.

[1483] Hardware and software used

[1484] Hardware: Smartphones, wearable devices (heart rate, blood pressure measurement)

[1485] Software: EmotionEngine (emotion recognition engine), RecipeAI (recipe recommendation engine), DeliveryAPI (delivery service API)

[1486] Data processing and calculation

[1487] 1. Acquisition of biometric information: The device (smartphone or wearable device) acquires the user's biometric information such as heart rate and blood pressure.

[1488] 2. Emotion recognition: The biometric information acquired by the server is analyzed using EmotionEngine to recognize the user's emotions.

[1489] 3. Recipe recommendation: The server uses RecipeAI to recommend appropriate recipes based on the emotions recognized by the server.

[1490] 4. Purchase and delivery: The server purchases the necessary ingredients based on the recommended recipe using the Delivery API and delivers them to the user using a delivery service.

[1491] Specific examples

[1492] For example, if a user is feeling stressed, EmotionEngine will recognize this and RecipeAI will recommend recipes containing ingredients that have stress-relieving properties, such as chamomile tea and dark chocolate.

[1493] Prompt Sentence Examples

[1494] "Suggest recipes that are suitable for when the user's heart rate is high and they are feeling stressed."

[1495] In this way, a system can be realized that provides food delivery services according to the user's preferences, health condition, and emotions.

[1496] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1497] Step 1:

[1498] The terminal acquires the user's biometric information.

[1499] Input: User's vital signs such as heart rate and blood pressure

[1500] Data processing: The device uses sensors to measure biometric information and acquires it as digital data.

[1501] Output: Acquired biometric data

[1502] Step 2:

[1503] The server sends biometric information to the Emotion Engine to recognize emotions.

[1504] Input: Acquired biometric data

[1505] Data calculation: EmotionEngine analyzes biometric information and recognizes the user's emotions

[1506] Output: Recognized emotion data (e.g., stress, relaxation, etc.)

[1507] Step 3:

[1508] Based on the recognized emotional data, the server sends prompts to RecipeAI and recommends appropriate recipes.

[1509] Input: Recognized emotion data

[1510] Data processing: The server generates prompts based on the emotion data and sends them to RecipeAI.

[1511] Output: Recommended recipe data

[1512] Step 4:

[1513] The server generates a list of necessary ingredients based on the recommended recipe data.

[1514] Input: Recommended recipe data

[1515] Data processing: The server analyzes the recipe data and generates a list of required ingredients.

[1516] Output: List of ingredients needed

[1517] Step 5:

[1518] The server purchases ingredients using the Delivery API based on the list of ingredients needed.

[1519] Input: List of ingredients needed

[1520] Data calculation: The server sends the ingredients list to the Delivery API and completes the purchase process.

[1521] Output: Purchase confirmation data

[1522] Step 6:

[1523] Based on the purchase confirmation data, the server uses a delivery service to deliver the ingredients to the user.

[1524] Input: Purchase confirmation data

[1525] Data processing: The server sends the necessary information to the delivery service and processes the delivery.

[1526] Output: Delivery confirmation data

[1527] In this way, a series of processes is realized, including recognizing the user's emotions based on their biometric information, recommending appropriate recipes, and purchasing and delivering the necessary ingredients.

[1528] Example 3

[1529] Next, a description will be given of 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."

[1530] Food waste and food loss are major issues in modern society. Elderly households and those with children face the added challenge of a heavy shopping burden, making it difficult for them to select the right ingredients for their health and living situation. Furthermore, there is a demand for services that respond to the emotions of users, but the current system does not adequately address these needs.

[1531] 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.

[1532] In this invention, the server includes: a means for performing machine learning on food market trends and information on ingredients available at retail stores on a daily basis; a means for recommending recipes that take into account a user's preferences, health status, seasonal ingredients, and nutritional balance; a means for providing a comprehensive service from demand forecasting and machine learning recommendations to purchasing; a means for inputting a user's ingredient information; a means for analyzing the input ingredient information and generating optimal recipes; a means for recognizing emotions based on the user's behavioral patterns and historical data; and a means for providing delivery services based on the results of emotion recognition. This makes it possible to reduce food waste and food loss. It also makes it possible to provide elderly households and households with children with support in selecting and purchasing ingredients appropriate for their health status and lifestyle. Furthermore, it is possible to provide services tailored to the user's emotions, which is expected to improve user satisfaction.

[1533] "Food market trends" refers to fluctuations and trends in the supply and demand of food products in the market.

[1534] "Information on ingredients arriving at retail stores" refers to information on the types and quantities of ingredients newly arrived at retail stores.

[1535] "Machine learning" refers to a technology in which a computer automatically learns from data and makes predictions and classifications.

[1536] "User preferences" refers to personal preferences such as the types of ingredients and dishes preferred by the user, and taste preferences.

[1537] "Health Condition" refers to a User's physical health status or specific health condition (e.g., high blood pressure, diabetes, etc.).

[1538] "Seasonal ingredients" refer to ingredients that are considered to be at their most delicious and nutritious during that season.

[1539] "Nutritional balance" refers to a state in which a diet contains an appropriate amount of necessary nutrients.

[1540] A "recipe" is a list of steps or ingredients for making a particular dish.

[1541] "Demand forecasting" refers to the analysis and calculations used to predict future demand.

[1542] "Ingredient information" refers to information regarding the types and quantities of ingredients owned by the user.

[1543] "Behavioral patterns" refer to a user's past actions and habits.

[1544] "Historical data" refers to a user's past purchase history and behavioral history.

[1545] "Emotion recognition" refers to technology that analyzes and recognizes the user's emotions.

[1546] "Delivery Service" refers to a service that delivers purchased products to a location designated by the user.

[1547] MODE FOR CARRYING OUT THE INVENTION

[1548] This invention is a system for optimizing food consumption and reducing food waste. This system inputs the user's ingredient information, generates optimal recipes, and suggests purchasing the necessary ingredients. It also makes purchasing suggestions for elderly households and households with children, taking into account their health status and living situation. Furthermore, by combining it with an emotion engine, it provides a delivery service that responds to the user's emotions.

[1549] Hardware and software used

[1550] Server: Database server, AI model server

[1551] Device: User's smartphone or tablet

[1552] Software: AI recommendation engine, emotion recognition engine, delivery management system

[1553] Specific operation of the system

[1554] 1. Enter the user's ingredient information

[1555] Users use their smartphones or tablets to enter current food inventory information into the app.

[1556] Specifically, the user enters information such as "3 tomatoes, 1 pack of cheese, 1 bunch of basil" into the app's input form.

[1557] 2. Sending food information

[1558] The terminal transmits the input ingredient information to the server.

[1559] Specifically, the terminal converts the input data into JSON format and sends it to the server using the HTTPS protocol.

[1560] 3. Optimizing food consumption

[1561] The server passes the received ingredient information to the AI ​​recommendation engine.

[1562] The server uses an AI recommendation engine to analyze the input ingredient information and generate the optimal recipe.

[1563] Specifically, the server generates a recipe for "Caprese salad" based on "tomato, cheese, and basil" and suggests it to the user.

[1564] 4. Generate purchase offers

[1565] The server generates a shopping list of necessary ingredients based on the generated recipe.

[1566] The server makes purchasing suggestions to elderly households and households with children, taking into account their health and living conditions.

[1567] Specifically, the server adds the olive oil, salt, and pepper needed for a Caprese salad to the shopping list, and suggests low-sodium salt for elderly households.

[1568] 5. Emotion recognition

[1569] The server uses an emotion recognition engine to analyze the user's emotions based on the user's behavioral patterns and historical data.

[1570] Specifically, the server analyzes the user's past purchase history and reviews, and if the user is dissatisfied with "slow delivery," it recognizes that emotion as "anger."

[1571] 6. Providing delivery services

[1572] The server provides appropriate delivery services based on the results of emotion recognition.

[1573] The server offers rush delivery if the user is angry, or a surprise gift package if the user is happy.

[1574] Specifically, the server will offer "free delivery on the next order" and, if the user is happy, will include a "surprise gift."

[1575] Examples and prompts

[1576] Specific examples

[1577] The user inputs "ingredients in the refrigerator" on their smartphone.

[1578] The terminal transmits the input data to the server.

[1579] The server generates a recipe for "Caprese salad" based on "tomato, cheese, and basil" and suggests it to the user.

[1580] The server adds the olive oil, salt, and pepper needed for a Caprese salad to the shopping list, and suggests low-sodium salt for elderly households.

[1581] The server analyzes the user's past purchase history and reviews, and if the user is dissatisfied with a slow delivery, it recognizes that emotion as anger.

[1582] The server offers "free shipping on your next delivery" and, if the customer is happy, includes a "surprise gift."

[1583] Prompt Sentence Examples

[1584] Please enter the ingredients you have in your refrigerator.

[1585] "We recommend purchasing the following ingredients based on your health status."

[1586] "Based on current sentiment, we'll offer your next delivery free."

[1587] The above is a specific embodiment of the system of the present invention. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1588] Step 1:

[1589] Users use their smartphones or tablets to enter current food inventory information into the app.

[1590] Input: The type and quantity of ingredients the user owns (e.g., "3 tomatoes, 1 pack of cheese, 1 bunch of basil")

[1591] Output: The entered ingredient information is saved on the device.

[1592] Specifically, the user enters ingredient information into the application's input form and presses the submit button.

[1593] Step 2:

[1594] The terminal transmits the input ingredient information to the server.

[1595] Input: Ingredient information entered by the user

[1596] Output: Ingredient information sent to the server

[1597] Specifically, the terminal converts the input data into JSON format and sends it to the server using the HTTPS protocol.

[1598] Step 3:

[1599] The server passes the received ingredient information to the AI ​​recommendation engine.

[1600] Input: Ingredient information sent from the device

[1601] Output: Ingredient information passed to the AI ​​recommendation engine

[1602] Specifically, the server analyzes the received data and passes it on to the AI ​​recommendation engine.

[1603] Step 4:

[1604] The server uses an AI recommendation engine to analyze the input ingredient information and generate the optimal recipe.

[1605] Input: Ingredient information passed to the AI ​​recommendation engine

[1606] Output: The generated recipe (e.g. "Caprese Salad")

[1607] Specifically, the server generates a recipe for "Caprese salad" based on "tomato, cheese, and basil" and suggests it to the user.

[1608] Step 5:

[1609] The server generates a shopping list of necessary ingredients based on the generated recipe.

[1610] Input: Generated recipe

[1611] Output: Shopping list (e.g. "olive oil, salt, pepper")

[1612] Specifically, the server lists the ingredients needed for a recipe and suggests "low-sodium salt" for elderly households.

[1613] Step 6:

[1614] The server uses an emotion recognition engine to analyze the user's emotions based on the user's behavioral patterns and historical data.

[1615] Input: User behavior patterns and historical data

[1616] Output: Recognized emotion (e.g., "anger")

[1617] Specifically, the server analyzes the user's past purchase history and reviews, and if the user is dissatisfied with "slow delivery," it recognizes that emotion as "anger."

[1618] Step 7:

[1619] The server provides appropriate delivery services based on the results of emotion recognition.

[1620] Input: Recognized emotion

[1621] Output: Delivery service offered (e.g. "Next delivery is free")

[1622] Specifically, the server will offer "free delivery on the next order" and, if the user is happy, will include a "surprise gift."

[1623] (Application example 3)

[1624] 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."

[1625] In modern society, food waste is an issue, and the need for shopping support for elderly households and those with children is growing. There is also a demand for services that respond to users' emotions, but no effective system exists to achieve this. Another issue is the lack of a system that can make recommendations based on customer preferences and health status, and provide a seamless service from demand forecasting to purchasing.

[1626] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes a means for performing machine learning on a daily basis of food market trends and information on ingredients received at retailers, etc., a means for recommending recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance, a means for providing an end-to-end service from demand forecasting and machine learning recommendations to purchasing, and a means for recognizing user emotions and providing delivery services in accordance with those emotions. This makes it possible to reduce food waste, provide shopping support to elderly households and households with children, and further realize the provision of services in accordance with the user's emotions.

[1627] "Food market trends" refers to fluctuations and trends in food supply and demand.

[1628] "Retail store" means a store that sells goods directly to consumers.

[1629] "Information on newly arrived ingredients" refers to information on the types and quantities of ingredients newly arrived at retail stores and markets.

[1630] "Machine learning" refers to a technology in which computers analyze data, recognize patterns, and make predictions and judgments.

[1631] "Customer preferences" refer to the types and characteristics of products and services that customers prefer.

[1632] "Health status" refers to the state of an individual's physical and mental health.

[1633] "Seasonal ingredients" refer to ingredients that are at their most delicious and nutritious during that season.

[1634] "Nutritional balance" refers to the proper distribution of nutrients necessary to maintain health.

[1635] A "recipe" is a set of instructions that describes how to prepare a dish.

[1636] "Recommendation" refers to presenting the best options based on specific criteria.

[1637] "Demand forecasting" refers to predicting future demand.

[1638] "Through" means to do something consistently from start to finish.

[1639] "User emotion" refers to the emotions or moods that a user is feeling.

[1640] "Delivery service" refers to a service that delivers products to a location designated by the customer.

[1641] A system for implementing the present invention includes the following major components:

[1642] 1. Server:

[1643] The server uses machine learning to learn daily information on food market trends and incoming ingredients from retailers, etc. This allows it to constantly monitor the latest market trends and incoming information and store it in a database. The software used is a machine learning algorithm (e.g., TensorFlow, PyTorch).

[1644] 2. Terminal:

[1645] The device (smartphone, tablet, etc.) recommends recipes that take into consideration the customer's preferences, health status, seasonal ingredients, and nutritional balance. The device receives data sent from the server and displays the optimal recipe for the user. Recipe recommendations use a recommendation engine (e.g., Collaborative Filtering, Content-Based Filtering).

[1646] 3. User:

[1647] Users can check recipes and purchase the necessary ingredients through the device. The device is equipped with emotion recognition software to recognize the user's emotions. The device analyzes emotions from text and voice data entered by the user and provides delivery services according to those emotions.

[1648] 4. Delivery Service:

[1649] The server provides a delivery service according to the user's emotions. For example, if the user is feeling angry, the server will send a rush delivery, and if the user is feeling happy, the server will include a surprise gift. A delivery management system is used to manage the delivery service.

[1650] Examples:

[1651] User profile: 70-year-old elderly person with diabetes, prefers a low-carb, high-fiber diet

[1652] User input: "I'm feeling very stressed today."

[1653] Example prompt for a generative AI model:

[1654] User profile: Age: 70, Household type: Seniors, Health condition: Diabetic, Preferences: Low sugar, High fiber

[1655] User input: "I'm feeling very stressed today."

[1656] Design an application to recommend appropriate recipes to this user and provide delivery services based on their emotions.

[1657] In this way, we can realize a system that reduces food waste and provides shopping support to elderly households and families with children. Furthermore, it will be possible to provide services that respond to the user's emotions.

[1658] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1659] Step 1:

[1660] The server collects daily information on food market trends and incoming ingredients from retailers, etc., and analyzes it using machine learning algorithms. Input includes arrival information from each retailer and market trend data. This data is analyzed and the latest market trends and arrival status are stored in a database. The output is the analysis results, which are saved in the database.

[1661] Step 2:

[1662] The server recommends recipes based on the customer's preferences, health status, seasonal ingredients, and nutritional balance. The input includes customer profile information (age, health status, preferences, etc.). A recommendation engine is used to analyze this information and generate optimal recipes. The output is the recommended recipe sent to the device.

[1663] Step 3:

[1664] The terminal displays the recipe sent from the server to the user. The input includes the recipe data sent from the server. The terminal visually displays this to the user, allowing the user to confirm the recipe. The output is a screen where the user can confirm the recipe.

[1665] Step 4:

[1666] The user checks the recipe through the terminal and purchases the necessary ingredients. The input includes the recipe information displayed on the terminal. The user selects the necessary ingredients based on the recipe and completes the purchase procedure. The output is a shopping list that is generated and sent to the server.

[1667] Step 5:

[1668] The device uses emotion recognition software to recognize the user's emotions. Inputs include the user's text and voice data. The emotion recognition software analyzes this data and identifies the user's emotions. The output is recognized emotion data, which is sent to the server.

[1669] Step 6:

[1670] The server provides delivery services according to the user's emotions. The input includes the recognized emotion data. The server selects an appropriate delivery service based on the emotion data and issues instructions to the delivery management system. For example, if the user is feeling angry, the server will expedite the delivery, and if the user is feeling happy, the server will include a surprise gift. The output is a delivery instruction that is generated and sent to the delivery management system.

[1671] Step 7:

[1672] The delivery management system executes deliveries based on instructions from the server. The input includes delivery instructions from the server. The delivery management system follows the instructions to properly package the product and deliver it to the user using the specified delivery method. The output is the product being delivered to the user.

[1673] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1674] 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 performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1675] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1676] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1677] [Third embodiment]

[1678] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1679] 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.

[1680] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1681] 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.

[1682] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1683] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1684] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1685] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1686] 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.

[1687] 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.

[1688] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1689] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1690] "Example 1"

[1691] A first embodiment of the present invention is a system that uses AI to learn daily information about food market trends and supermarket stock information. This system collects market trend data and supermarket stock information, and the AI ​​learns based on this information. Specifically, market trend data includes, for example, food price fluctuations, consumer purchasing habits, and seasonal influences. Furthermore, supermarket stock information includes, for example, the arrival date, stock volume, and sales status of each product. Based on this information, the AI ​​recommends recipes that take into account the customer's preferences, health condition, seasonal ingredients, and nutritional balance.

[1692] "Example 2"

[1693] A second embodiment of the present invention is a system that provides a seamless service from demand forecasting to purchasing and delivery. This system predicts the demand for ingredients based on recipes recommended by AI, and enables customers to purchase them. Specifically, the AI ​​creates a list of ingredients included in the recommended recipes, and suggests purchasing ingredients based on that list. The purchased ingredients are then delivered to the customer using a delivery service.

[1694] "Example 3"

[1695] A third embodiment of the present invention is a system that contributes to reducing food waste and provides shopping support to elderly households and households with children. This system makes suggestions to optimize food consumption and reduce food waste. Specifically, it suggests purchasing necessary ingredients based on recipes recommended by AI, encouraging the consumption of surplus ingredients. In addition, for elderly households and households with children, it makes purchasing suggestions that take into account their health status and living situation.

[1696] The processing flow of each embodiment will be described below.

[1697] "Example 1"

[1698] Step 1: Collect market trend data and supermarket inventory information. This information includes, for example, food price fluctuations, consumer purchasing habits, seasonal influences, arrival dates, quantities, and sales status of each product.

[1699] Step 2: The AI ​​learns from the collected information. Based on market trends and arrival information, the AI ​​learns a model to recommend recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[1700] Step 3: Using the AI-trained model, the system recommends recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance.

[1701] "Example 2"

[1702] Step 1: Based on the recipe recommended by AI, the demand for ingredients is predicted. Specifically, a list of ingredients included in the recipe is created and purchasing suggestions are made based on that list. Step 2: The ingredients that the customer has decided to purchase are delivered to the customer using a delivery service.

[1703] "Example 3"

[1704] Step 1: Based on the recipes recommended by AI, suggestions are made to optimize ingredient consumption and reduce food waste. Specifically, a list of ingredients included in the recipe is created, and based on that, suggestions are made to encourage the consumption of surplus ingredients.

[1705] Step 2: Make purchasing suggestions to elderly households and households with children, taking into account their health and living conditions.

[1706] Example 1

[1707] 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 headset type terminal 314 will be referred to as a "terminal."

[1708] In today's grocery market, consumers demand ingredient selection and recipe suggestions that accommodate diverse tastes, health conditions, and seasonal changes. However, providing appropriate recipes that meet these needs is not easy. Reducing food waste and providing shopping support for elderly households and families with children are also important issues. Furthermore, there is a need to build a comprehensive platform that delivers information based on customer attributes. To solve these issues, an efficient and effective system is required.

[1709] 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.

[1710] In this invention, the server includes: a means for performing machine learning on daily food market trends and retail store inventory information; a means for storing the collected data in a database; a means for preprocessing the stored data; a means for inputting the preprocessed data into a machine learning model; a means for generating recipes using the machine learning model that take into account customer preferences, health status, seasonal ingredients, and nutritional balance; a means for providing the generated recipes to a user terminal; and a means for collecting user feedback and using it to improve the system. This enables the provision of recipes that meet diverse consumer needs, reducing food waste and providing shopping support to elderly households and households with children. It also enables the creation of a comprehensive platform that distributes information based on customer attributes.

[1711] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[1712] "Retail store stock information" is information about product stock at retail stores, such as the arrival date, stock volume, and sales status of each product.

[1713] "Machine learning" is a technology that allows computers to learn patterns based on data and make predictions and classifications.

[1714] A "database" is a system for efficiently storing, managing, and retrieving data.

[1715] "Preprocessing" refers to the process of preparing data before inputting it into a machine learning model, such as cleaning and normalizing the data.

[1716] A "machine learning model" is an algorithm or mathematical model that learns from data and makes predictions or classifications.

[1717] A "recipe" is a set of instructions that shows how to prepare a dish and how to combine ingredients.

[1718] A "user terminal" is a device that allows a user to receive information and perform operations.

[1719] "Feedback" is information that is useful for improving the system, such as user ratings and comments.

[1720] "Food waste reduction" is an initiative aimed at reducing food waste and using it efficiently.

[1721] "Shopping support" is a service that provides shopping support to consumers with specific needs, such as elderly households and households with children.

[1722] "Customer attributes" refers to characteristics and information about customers, such as family structure, age, and preferences.

[1723] An "integrated platform" is a system that integrates multiple functions and services to provide users with a consistent experience.

[1724] This invention is a system that collects food market trends and retail store arrival information daily, and uses machine learning to generate recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance. Specific embodiments of this system are described below.

[1725] Server Processing

[1726] 1. Data Collection

[1727] The server collects food market trend data and retail store arrival information every day at 9:00 a.m. Market trend data includes food price fluctuations, consumer purchasing trends, seasonal influences, etc. This data is obtained using the API of the food market information service.

[1728] The retail store's stock information includes the arrival date, stock quantity, sales status, etc. of each product. This data is obtained from the retail store's inventory management system.

[1729] 2. Data storage

[1730] The server stores the collected data in a MySQL database, which stores data such as food price fluctuation data, consumer purchasing trends, and retail store inventory information as tables.

[1731] 3. Data Preprocessing

[1732] The server preprocesses the stored data, including cleaning the data (filling in missing values ​​and removing outliers) and normalizing it. The server uses the Python pandas library to clean and normalize the data.

[1733] 4. Inputting data into the AI ​​model

[1734] The server then inputs the preprocessed data into an AI model built with TensorFlow, which learns from food price fluctuations, consumer purchasing trends, and retailer inventory information.

[1735] 5. Recipe Generation

[1736] The server uses an AI model to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. The generated recipes are then stored in a database.

[1737] 6. Recipe provision

[1738] The server sends the generated recipe to the terminal via the API.

[1739] Terminal handling

[1740] 1. Receiving recipes

[1741] The terminal receives the recipe information sent from the server and stores the received recipe information in a local database of the terminal.

[1742] 2. Recipe display

[1743] The device then displays the received recipe information to the user. Front-end frameworks such as React and Vue.js are used for displaying the information, and detailed recipe information (ingredients, cooking instructions, nutritional information, etc.) is displayed in an easy-to-read format.

[1744] 3. Feedback Collection

[1745] The device collects feedback from users, including recipe ratings and comments, and stores the feedback in a local database on the device.

[1746] 4. Send Feedback

[1747] The terminal sends the collected feedback to the server.

[1748] User operations

[1749] 1. View recipes

[1750] The user browses the provided recipes through the device, checks the detailed information of the recipes, and selects the recipes they want to try.

[1751] 2. Recipe Trial

[1752] The user then tries out the selected recipe and cooks the dish according to the recipe.

[1753] 3. Feedback input

[1754] Users can input their ratings and comments on the recipes into the device, including feedback on the cooking results, taste, ease of use, etc.

[1755] 4. Send Feedback

[1756] The user sends the input feedback to the server through the terminal.

[1757] Specific examples

[1758] Example 1: Collecting market trend data

[1759] The server uses the API of a food market information service to obtain food price fluctuation data at 9:00 AM every day. For example, it sends the following API request:

[1760] GET / api / market_trends?date=2023-10-01

[1761] Example 2: Data input to an AI model

[1762] The server inputs the preprocessed data into an AI model built with TensorFlow, for example using a prompt like this:

[1763] Prompt: Based on food price fluctuation data and retail store stock information, predict food prices for the next week.

[1764] Example 3: Creating and serving recipes

[1765] The server uses the AI ​​model to generate recipes, which are then sent to the device. For example, the following recipes are provided:

[1766] Recipe name: Healthy salad made with seasonal vegetables

[1767] Ingredients: Tomato, cucumber, lettuce, olive oil, salt

[1768] Directions: 1. Thinly slice the tomato and cucumber. 2. Tear the lettuce into bite-sized pieces. 3. Place all ingredients in a bowl and toss with olive oil and salt.

[1769] This way, users receive fresh, healthy recipes every day.

[1770] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1771] Step 1: Data collection

[1772] The server collects food market trend data and retail store arrival information every day at 9:00 AM. Market trend data includes food price fluctuations, consumer purchasing trends, seasonal effects, etc. This data is obtained using the API of the food market information service. Retail store arrival information includes the arrival date, arrival quantity, sales status, etc. of each product. This data is obtained from the retail store's inventory management system. The input is the API request, and the output is the collected market trend data and arrival information.

[1773] Step 2: Save data

[1774] The server stores the collected data in a MySQL database. The database stores data such as food price fluctuation data, consumer purchasing trends, and retail store inventory information as tables. The input is the collected data, and the output is the data stored in the database.

[1775] Step 3: Data Preprocessing

[1776] The server preprocesses the stored data. Preprocessing includes data cleaning (filling in missing values ​​and removing outliers) and normalization. The Python pandas library is used to clean and normalize the data. The input is the raw data stored in the database, and the output is the preprocessed data.

[1777] Step 4: Input data into the AI ​​model

[1778] The server inputs the preprocessed data into an AI model built with TensorFlow. The AI ​​model learns based on food price fluctuations, consumer purchasing trends, and retail store inventory information. The input is the preprocessed data, and the output is the learning results.

[1779] Step 5: Recipe Generation

[1780] The server uses an AI model to generate recipes that take into account the customer's preferences, health status, seasonal ingredients, and nutritional balance. The generated recipes are stored in a database. The input is the learning results of the AI ​​model, and the output is the generated recipe.

[1781] Step 6: Provide the recipe

[1782] The server sends the generated recipe to the terminal via API. The input is the generated recipe, and the output is the recipe information sent to the terminal.

[1783] Step 7: Receiving the recipe

[1784] The terminal receives recipe information sent from the server. The received recipe information is stored in the terminal's local database. The input is the recipe information sent from the server, and the output is the recipe information stored in the local database.

[1785] Step 8: View the recipe

[1786] The device displays the received recipe information to the user. Front-end frameworks such as React and Vue.js are used for display. Detailed information about the recipe (ingredients, cooking instructions, nutritional information, etc.) is displayed in an easy-to-read format. The input is the recipe information stored in the local database, and the output is the recipe information displayed to the user.

[1787] Step 9: Gather feedback

[1788] The device collects feedback from users. The feedback includes recipe ratings and comments. The collected feedback is stored in a local database on the device. The input is the feedback from the user, and the output is the feedback stored in the local database.

[1789] Step 10: Send feedback

[1790] The terminal sends the collected feedback to the server, where the input is the feedback stored in the local database and the output is the feedback sent to the server.

[1791] Step 11: Use feedback

[1792] The server receives feedback sent by users and uses it to improve the system. The input is the feedback sent by users, and the output is the feedback used to improve the system.

[1793] (Application example 1)

[1794] 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."

[1795] Modern consumers have diverse needs when it comes to purchasing groceries. In particular, they want recipe suggestions based on their health status and preferences, meal suggestions using seasonal ingredients, and efficient purchasing that takes into account market trends and supermarket stock information. However, there is no consistent system in place to meet these needs. Reducing food waste and providing shopping support for elderly households and households with children are also important issues.

[1796] 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.

[1797] In this invention, the server includes: means for daily AI learning of food market trends and supermarket inventory information; means for recommending recipes that take into account customer preferences, health status, seasonal ingredients, and nutritional balance; means for providing an end-to-end process from demand forecasting and AI recommendations to purchase; means for recommending recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance through an application installed on a smartphone; and means for proposing recipes that suit the user's wishes using a generative AI model. This meets the diverse needs of consumers, enabling efficient grocery shopping and healthy eating.

[1798] "Food market trends" refers to information about food market trends, such as food price fluctuations, consumer purchasing habits, and seasonal influences.

[1799] "Information on ingredients arriving at supermarkets" refers to information on the arrival of ingredients at supermarkets, such as the arrival date, amount of each product, and sales status.

[1800] "AI learning" is the learning process that enables artificial intelligence to recognize patterns based on large amounts of data and make predictions and judgments.

[1801] "Customer preferences" refers to information about the types of ingredients and dishes that a particular customer prefers.

[1802] "Health conditions" refers to information about the health of a particular customer, including, for example, allergies and restrictions on the intake of certain nutrients.

[1803] "Seasonal ingredients" are ingredients that are considered to be most delicious and nutritious during a particular season.

[1804] "Nutritional balance" refers to a diet in which the nutrients necessary to maintain health are properly distributed.

[1805] "Means for recommending recipes" refers to a method of suggesting optimal cooking recipes that take into consideration the customer's preferences, health condition, seasonal ingredients, and nutritional balance.

[1806] "Demand forecasting" is an analytical method for predicting future demand for a product.

[1807] "A means of providing a seamless service" is a method of consistently providing a series of processes from demand forecasting to recommendations and purchase.

[1808] An "application installed on a smartphone" is a software program that runs on a smartphone.

[1809] A "generative AI model" is a model that allows artificial intelligence to generate new information and suggestions based on data.

[1810] A "prompt" is an instruction given to a generative AI model, serving as a guideline for generating specific information or suggestions.

[1811] The following system configuration will be described as an embodiment of the present invention.

[1812] System Configuration

[1813] The system includes a server, the user's smartphone, and a generative AI model. The server collects daily information on food market trends and ingredients arriving at supermarkets, and uses this information for AI learning. A dedicated application is installed on the user's smartphone, which recommends recipes that take into account the user's preferences, health status, seasonal ingredients, and nutritional balance.

[1814] Program processing

[1815] The server processes the data using the following means:

[1816] 1. Data collection methods:

[1817] The server collects daily food market trend data (price fluctuations, consumer purchasing habits, seasonal influences, etc.) and supermarket inventory information (arrival date, amount, sales status, etc.). This data is obtained using APIs and web scraping technology.

[1818] 2. AI learning methods:

[1819] The server performs AI learning based on the collected data. Specifically, it uses Python machine learning libraries (e.g., TensorFlow, PyTorch) to build food demand forecasting models and recipe recommendation models.

[1820] 3. Recipe recommendation methods:

[1821] The server recommends optimal recipes taking into consideration the user's preferences, health status, seasonal ingredients, and nutritional balance. Recipe information is provided through an application installed on the user's smartphone.

[1822] 4. Generative AI model means:

[1823] The server uses a generative AI model to suggest recipes that match the user's preferences, and the generative AI model generates new recipes based on the user's input (prompt sentence).

[1824] Hardware and software used

[1825] Hardware: Servers, smartphones

[1826] Software: Python, TensorFlow, PyTorch, dedicated applications

[1827] Specific examples

[1828] For example, if a user wants "low-carb recipes," they can input the following prompt into the generative AI model:

[1829] Example prompt sentence:

[1830] "The user wants low-carb recipes. Please recommend the best recipes based on market trend data and supermarket stock information."

[1831] By inputting this prompt into a generative AI model, the AI ​​will suggest recipes that match the user's preferences. The recipe information is provided through an application installed on the user's smartphone, and the user can purchase ingredients and cook based on the recipe.

[1832] This will meet the diverse needs of consumers, enable efficient grocery shopping, and encourage healthy eating habits.

[1833] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1834] Step 1:

[1835] The server collects food market trend data and supermarket stock information daily. Specifically, it uses API and web scraping technology to obtain data such as price fluctuations, consumer purchasing trends, seasonal influences, stock arrival dates, stock quantities, and sales status. The input is market trend data and stock arrival information, and the output is the collected data.

[1836] Step 2:

[1837] The server performs AI training based on the collected data. It uses Python machine learning libraries (e.g., TensorFlow, PyTorch) to build food demand forecasting models and recipe recommendation models. The input is the collected market trend data and stock information, and the output is a trained AI model.

[1838] Step 3:

[1839] Users input their preferences, health status, and desired recipe conditions through a dedicated application installed on their smartphone. The input is the user's preferences, health status, and desired recipe conditions, and the output is this information.

[1840] Step 4:

[1841] The server receives user input and uses an AI model to recommend optimal recipes. Specifically, it filters recipes and selects the most suitable one, taking into consideration the user's preferences, health status, seasonal ingredients, and nutritional balance. The input is the user's preferences, health status, and desired recipe conditions, and the output is the recommended recipe.

[1842] Step 5:

[1843] The server uses a generative AI model to generate a new recipe that matches the user's preferences. The prompt sentence is input into the generative AI model, which generates a recipe based on the user's preferences. The input is the prompt sentence, and the output is the generated recipe.

[1844] Step 6:

[1845] The server sends the recommended recipes and generated recipes to the user's smartphone. The user can review and select the suggested recipes through the smartphone application. The input is the recommended recipes and generated recipes, and the output is the recipe displayed on the user's smartphone.

[1846] Step 7:

[1847] The user purchases ingredients and cooks based on the selected recipe. A shopping list and cooking instructions are provided through a smartphone application. The input is the selected recipe, and the output is the shopping list and cooking instructions.

[1848] This will meet the diverse needs of consumers, enable efficient grocery shopping, and encourage healthy eating habits.

[1849] Example 2

[1850] Next, a description will be given of 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."

[1851] Conventional food purchasing systems have problems such as difficulty in proposing recipes based on customer preferences and health status, and the entire process from demand forecasting and purchasing to delivery is not carried out efficiently.Furthermore, ther...

Claims

1. A method to use machine learning to learn about incoming ingredients on a daily basis, A means for predicting demand for ingredients using the machine learning model trained by the machine learning means; A means for acquiring biometric information of a user and analyzing the emotion of the user using an emotion recognition engine based on the acquired biometric information; A means for generating a purchase proposal based on the demand forecast, the analyzed information indicating the user's emotions, and at least one of the information on the user's preferences and health condition; A system including:

2. A means for purchasing the food suggested for purchase and delivering the purchased food to the user using a delivery service is included. The system of claim 1 .

3. The means for generating a purchase suggestion generates a recipe using a generative AI model based on a prompt sentence corresponding to the information indicating the user's emotion. The system of claim 1 .

Images