System

Abstract

A system is provided.SOLUTION: The system includes means for collecting the generated data, means for transmitting the collected data to a server, means for learning a specific classification model using the data generated on the server, means for classifying the waste using the classification model, means for optimizing a collection route based on the classification result, and means for delivering sustainable practice information to the terminal.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] Modern urban waste management faces challenges such as inefficient collection routes and recycling contamination. This increases waste collection costs and increases the environmental burden due to improper sorting of recyclable materials. Additionally, there are challenges with citizens' limited access to knowledge about sustainable practices. New methods are needed to address these challenges. [Means for solving the problem]

[0005] The present invention provides a system that collects waste generation data and transmits the collected data to a server. The generated data is used on the server to train a specific classification model, which is then used to classify the waste. The system also designs optimal collection routes based on the classification results and delivers sustainable practice information to terminals. Furthermore, the system also includes a means for dividing the collected data into features and labels and evaluating the performance of the classification model. This enables efficient waste management and promotes recycling.

[0006] "Generated data" refers to information about waste collected by the device's sensors.

[0007] "Means of collection" refers to the function of the terminal to detect characteristic information of waste and convert it into digital form.

[0008] "Means for transmitting" refers to the function of transmitting data collected by the terminal to a server via the Internet or a local network.

[0009] "Server" refers to a centralized computer system for storing and processing collected data.

[0010] "Specific classification model" refers to a model trained on a server using a machine learning algorithm and a program for classifying waste.

[0011] "Means for learning" refers to the ability of the server to train a classification model using training data.

[0012] "Means for classification" refers to the function of analyzing newly collected data using a classification model to classify waste.

[0013] "Means for optimizing collection routes" refers to the function of the server to design and optimize efficient waste collection routes based on the classification results.

[0014] "Terminal" refers to a device for collecting waste characteristic information and transmitting the data to a server.

[0015] "Sustainable practice information" refers to information on environmentally friendly behaviour and proper waste handling methods.

[0016] "Means for distribution" refers to the function by which the server sends and notifies sustainable practice information to the terminal.

[0017] "Means of splitting into features and labels" refers to the process of analyzing data on the server and splitting it into features and labels required for prediction results.

[0018] "Means for evaluating performance" refers to the function of the server to perform tests to evaluate the accuracy and reliability of the classification model. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0027] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0040] This invention is a system that aims to reduce waste and promote recycling by using generative AI. Specifically, terminals installed in each supermarket and community send collected waste data to a server in real time, and the server analyzes and optimizes this data to achieve efficient waste management.

[0041] System Overview

[0042] The sensors in the device collect characteristic information about the waste (e.g., material type, color, weight, etc.) and convert it into digital format as generated data. The collected data is then sent to a server via the Internet or a local network.

[0043] The server receives the data, stores it, and pre-processes it, which includes cleaning and formatting it. The server then splits the pre-processed data into features and labels to train a machine learning model, a specific classification model, that classifies waste as recyclable or not.

[0044] System Operation

[0045] After training the classification model, the server analyzes newly collected data in real time and classifies the waste. Based on the classification results, it designs optimal collection routes and delivers relevant sustainable practice information to users via their devices.

[0046] Specific examples

[0047] For example, smart trash cans installed around town use various sensors to collect information about the characteristics of waste. This information is sent to a server via the user's smartphone or dedicated device. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[0048] The server then uses the classification results to plan optimal collection routes and provide feedback to waste collection companies. It also delivers educational content about non-recyclable waste to users' devices and provides information on sustainable practices, helping users take appropriate recycling actions.

[0049] In this way, the system contributes to efficient waste management, promoting recycling, and realizing a sustainable society through education.

[0050] The processing flow will be explained below.

[0051] Step 1:

[0052] The terminal uses various sensors to collect information about the characteristics of the waste (such as the type of material, color, weight, etc.), which is then converted into a digital format in real time.

[0053] Step 2:

[0054] The terminal transmits the collected data to a server via the Internet or a local network. The transmitted data includes information about the characteristics of the waste.

[0055] Step 3:

[0056] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[0057] Step 4:

[0058] The server separates the preprocessed data into features (e.g., material type, color, weight) and labels (e.g., recyclable / non-recyclable), forming the dataset needed to train the machine learning model.

[0059] Step 5:

[0060] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[0061] Step 6:

[0062] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[0063] Step 7:

[0064] The server then designs the optimal garbage collection route based on the classification results, enabling efficient collection and reducing costs and environmental impact.

[0065] Step 8:

[0066] The server provides feedback to waste collection companies on the designed collection routes and recycling information, allowing them to carry out collection according to the optimal plan.

[0067] Step 9:

[0068] The server generates educational content about non-recyclable waste and delivers sustainable practice information to users via their devices, allowing them to learn about correct recycling methods and environmentally friendly behavior.

[0069] Step 10:

[0070] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[0071] Example 1

[0072] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0073] To efficiently manage waste and promote recycling, it is necessary to classify waste, optimize collection routes, and provide information on sustainable practices. However, conventional systems lack the ability to adequately control the quality of collected data, train classification models, and analyze data in real time, making appropriate waste management difficult. Furthermore, they are also inadequate in providing users with the information they need on sustainable practices, limiting their contribution to a sustainable society.

[0074] 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.

[0075] In this invention, the server includes means for the terminal to collect waste characteristic information using sensors, means for converting the collected data into digital form and sending it to the server, means for storing, cleaning, and formatting the received data on the server, means for dividing the preprocessed data into features and labels and training a machine learning model, means for analyzing the new data in real time and classifying the waste as recyclable or not, means for designing optimal waste collection routes based on the classification results, and means for delivering sustainable practice information to the user through the terminal, thereby enabling efficient waste management and promotion of recycling, as well as providing users with prompt and appropriate information.

[0076] The "terminal" is a device equipped with a sensor to collect characteristic information about waste, convert the collected data into digital format, and send it to a server.

[0077] "Server" is a computer system for receiving, storing, and pre-processing collected data and for training and running machine learning models.

[0078] A "sensor" is a measuring device installed in a terminal to detect and collect characteristic information such as the material, color, and weight of waste.

[0079] "Characteristic information" is data that indicates the material, color, weight, etc. of the waste, and is collected by sensors.

[0080] "Digital format" refers to data that has been converted from analog data so that it can be processed electronically.

[0081] "Data cleaning" is the process of removing noise and missing values ​​from collected data, leaving only the necessary information.

[0082] "Formatting" refers to the conversion of data of different formats or structures into a unified format or structure.

[0083] A "feature" is a value that indicates a specific attribute or characteristic of the data used to train a machine learning model.

[0084] A "label" is a value that indicates the output or classification result corresponding to a feature in a machine learning model.

[0085] A "machine learning model" is an algorithm that learns patterns and relationships from given data and makes predictions and classifications for new data.

[0086] "Classification" is the process of grouping received data based on predefined categories or labels.

[0087] A "waste collection route" is a route used by a waste collection vehicle to efficiently collect waste.

[0088] "Sustainable practice information" is information on actions aimed at realizing a sustainable society, such as reducing environmental impact and promoting recycling.

[0089] "User" means any person or organization that uses this system and receives guidance on how to properly dispose of and recycle waste.

[0090] This invention is a system that aims to reduce waste and promote recycling by utilizing generative AI models. In this system, terminals installed in each supermarket and community send waste data in real time to a server, which then analyzes and optimizes this data to achieve efficient waste management.

[0091] System configuration

[0092] 1. The terminal is equipped with multiple sensors. These sensors collect characteristic information about the waste (such as material type, color, and weight) and convert it into digital form. Specific examples of sensors include material detection sensors, color-identification cameras, and weight sensors.

[0093] 2. The collected data is sent to a server via the Internet or a local network. The device transmits this data in real time, ensuring that it arrives at the server without delay.

[0094] 3. The server first stores the received data, then cleans and formats it. The software used includes database management systems and data cleaning tools, such as PostgreSQL and pandas (a Python library).

[0095] 4. The server splits the preprocessed data into features and labels to train a machine learning model. This model is implemented using the Python machine learning library scikit-learn. In particular, it uses a random forest classifier to classify whether the waste is recyclable or not.

[0096] 5. The server uses the trained model to analyze newly collected data in real time and classify the waste. Based on this classification, it designs optimal waste collection routes. Collection route planning is done using Geographic Information System (GIS) software.

[0097] 6. The server sends the sorting results and collection route information to the device and also provides the user with sustainable practice information, such as educational content about non-recyclable waste and the location of specific recycling facilities.

[0098] Specific examples

[0099] For example, a smart trash can installed in a city uses various sensors to collect information about the characteristics of waste. When a user throws a plastic bottle into the trash can, a material detection sensor detects the plastic, a color-identification camera identifies the color blue, and a weight sensor measures 200g. This data is converted into digital format and sent to a server in real time.

[0100] The server stores the received data, cleans and formats it using pandas, then splits the data ("plastic, blue, 200g") into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[0101] The server analyzes newly received data in real time and classifies it as "recyclable." Based on this classification, the GIS software designs optimal garbage collection routes and provides them to collection companies. At the same time, the user receives a notification via their device saying, "This waste is recyclable. Please separate it appropriately."

[0102] Below is an example of a prompt sentence:

[0103] "What is the recycling status of the waste I put in yesterday?"

[0104] In this way, the present invention can efficiently manage waste and promote recycling, and further contributes to the realization of a sustainable society through user education.

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Step 1:

[0107] The terminal uses sensors to collect information about the characteristics of the waste. The terminal is equipped with a material detection sensor, a color recognition camera, a weight sensor, etc. The data obtained by these sensors is information about the material, color, and weight of the waste.

[0108] Input: Waste

[0109] Specific operation: The sensor detects plastic waste, the color-detection camera identifies the color blue, and the weight sensor measures 200g.

[0110] Output: Feature information: "Plastic, Blue, 200g"

[0111] Step 2:

[0112] The device converts the collected data into a digital format and transmits it to a server via the internet or a local network. The device transmits this data in real time, so it arrives at the server without delay.

[0113] Input: Feature information "Plastic, Blue, 200g"

[0114] Specific operation: The device converts the analog data into digital format and sends the data to the specified address on the server.

[0115] Output: Digital characteristic information (data sent to the server)

[0116] Step 3:

[0117] The server stores, cleans, and formats the data it receives using a database management system (e.g., PostgreSQL) and a data cleaning tool (e.g., pandas).

[0118] Input: Digital feature information

[0119] Specific operation: The server stores the received data in a database and performs cleaning such as filling in missing values ​​and removing outliers. After that, the data format is unified.

[0120] Output: Preprocessed feature information

[0121] Step 4:

[0122] The server splits the preprocessed data into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[0123] Input: Preprocessed feature information

[0124] Specific operation: The server splits the data "plastic, blue, 200g" into "features (plastic, blue, 200g)" and "label (recyclable)" and supplies them to the random forest classifier as training data.

[0125] Output: A trained machine learning model

[0126] Step 5:

[0127] The server analyzes newly received data in real time and uses machine learning models to classify the waste.

[0128] Input: Newly received data (e.g. "glass, blue, 300g")

[0129] What it does: Use the trained model to analyze new data: "glass, blue, 300g" and classify it as "recyclable."

[0130] Output: Classification result (e.g. "Recyclable")

[0131] Step 6:

[0132] The server uses Geographic Information System (GIS) software to design optimal waste collection routes based on the classification results and provide feedback to collection companies.

[0133] Input: Classification results and permitted information (e.g. waste type and collection amount)

[0134] How it works: The server analyzes all waste data, calculates efficient collection routes based on the amount and type of waste, and provides collection schedules and efficient route guidance to collection companies.

[0135] Output: Optimized collection route information

[0136] Step 7:

[0137] The server delivers sustainable practice information to users through their devices, including, for example, educational content about non-recyclable waste and the locations of specific recycling facilities.

[0138] Input: Classification results and sustainable practice information

[0139] What it does: Generates an educational video on "How to dispose of non-recyclable plastics" and sends a notification to the user's smartphone.

[0140] Output: User notification and educational content

[0141] The above is the processing flow of this system.

[0142] (Application example 1)

[0143] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0144] In modern society, the increase in waste and the need for recycling have created a demand for efficient waste management. However, proper sorting and promotion of recycling are often insufficient. Brick-and-mortar stores in particular generate large amounts of waste, and a system is needed to efficiently manage this waste and promote sustainable practices.

[0145] 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.

[0146] In this invention, the server includes means for collecting the generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for providing the waste classification results in real time through the user's information terminal, and means for delivering educational content for promoting recycling to the user's information terminal, thereby enabling efficient waste management, promotion of recycling, and realization of a sustainable society.

[0147] "Generated data" means waste characteristics collected by a waste management system and converted into digital form.

[0148] The "server" is a central control unit that receives collected data and performs analysis and training of classification models.

[0149] A "classification model" is a model that uses machine learning algorithms to classify waste as recyclable or not.

[0150] "Collection route optimization" is a method for designing optimal waste collection routes based on classification results, thereby achieving efficient collection activities.

[0151] "Sustainable Practice Information" is educational content and guidelines for recycling and waste reduction provided to users.

[0152] "User's information terminal" refers to a device such as a smartphone or tablet that is used to receive and display information in real time.

[0153] "Means for providing waste classification results in real time" refers to a function that displays the classification results on the user's information terminal the moment the waste is collected.

[0154] "Educational content to promote recycling" is content that provides users with information on the correct methods of waste disposal and recycling.

[0155] As an embodiment of this invention, the following system can be constructed. This system improves the efficiency of waste management in physical stores, promotes recycling, and supports sustainable practices. The system is mainly composed of a server, terminals, and users.

[0156] The server has the following functions: First, it collects data generated by sensors and transmits it via the Internet or local network. This digitally aggregates waste characteristic information (such as material type, color, and weight) on the server. Next, the server analyzes the received data and performs preprocessing (data cleaning, formatting, etc.). It uses the preprocessed data to train a specific classification model, which is then used to classify waste as recyclable or not. It designs optimal collection routes based on the classification results and delivers this information to users' devices in real time. It also provides users with educational content to promote recycling as needed.

[0157] The device has the function of receiving and displaying sustainable practice information and waste classification results sent from the server, allowing users to check important information on how to correctly classify waste and recycling in real time. Specifically, this applies to users' information devices such as smartphones and tablets.

[0158] Users receive information from the server via their devices and implement appropriate waste management practices, which promotes efficient waste disposal and recycling. Furthermore, educational content helps users deepen their own understanding of sustainable practices.

[0159] For example, supermarket staff can send data obtained from waste management devices (smart trash cans) to a server via a smartphone app. This data is analyzed on the server, and real-time information on whether the waste can be recycled is provided to staff. Based on this information, staff can properly classify the waste. The app also provides educational content about non-recyclable waste, promoting sustainable behavior.

[0160] Example prompt sentence:

[0161] Write a program that uses data obtained from waste management terminals by supermarket staff to determine in real time whether waste can be recycled and displays the results on the staff's smartphone app, "Eco Shopper." Data obtained from the sensors includes material type, color, and weight. Based on the results, provide staff with information about recyclability.

[0162] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0163] Step 1:

[0164] The server collects the data generated by the sensors. Specifically, the sensors in the waste management terminal collect characteristic information such as the type, color, and weight of the waste material, converts this data into a digital format, and transmits it to the server via the Internet or a local network.

[0165] Input: Waste characteristics (type of material, color, weight)

[0166] Output: Digital feature information data

[0167] Step 2:

[0168] The server preprocesses the received data. Specifically, the server performs data cleaning and formatting to generate preprocessed data.

[0169] Input: Digital feature data

[0170] Output: Preprocessed data

[0171] Step 3:

[0172] The server uses the preprocessed data to train a classification model. Specifically, it splits the preprocessed data into features and labels, and uses them as input to train a machine learning model (e.g., random forest).

[0173] Input: Preprocessed data

[0174] Output: A trained classification model

[0175] Step 4:

[0176] The server classifies newly collected waste data by analyzing the newly received data using a trained classification model and classifying the waste as recyclable or not in real time.

[0177] Input: Newly collected waste data

[0178] Output: Waste classification results

[0179] Step 5:

[0180] The server optimizes collection routes based on the classification results, specifically by designing optimal collection routes depending on the type and amount of waste, and preparing feedback for collection companies.

[0181] Input: Waste classification results

[0182] Output: Optimized collection route information

[0183] Step 6:

[0184] The server delivers sustainable practice information and classification results to the device, specifically, information on recyclable waste and educational content to promote recycling to the user's smartphone or tablet.

[0185] Input: Sustainable practice information and classification results

[0186] Output: Information displayed on the user's terminal

[0187] Step 7:

[0188] The user's device displays the received information. Specifically, the smartphone app displays the classification results and sustainable practice information in real time, and guides the user to appropriate waste disposal methods.

[0189] Input: Information from the server (classification results, sustainable practice information)

[0190] Output: Information displayed on the terminal screen

[0191] 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.

[0192] This invention combines a system that uses generative AI to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[0193] System Overview

[0194] The device's sensors collect waste characteristics (e.g., material type, color, weight, etc.) and simultaneously acquire the user's emotional data. For example, the device can recognize the user's emotions by analyzing their facial expressions and tone of voice as they sort the waste. The collected information is converted into digital form in real time and sent to a server via the Internet or a local network.

[0195] System Operation

[0196] The server stores the received data and pre-processes it. This includes cleaning and formatting the data. The server then splits the pre-processed data into features and labels to train a machine learning model. This model is a specific classification model that classifies waste as recyclable or not.

[0197] Furthermore, the server is equipped with an emotion engine that analyzes and classifies the user's emotion data. Based on the emotions recognized by the emotion engine, the server personalizes sustainable practice information and delivers it to the user via their device.

[0198] Specific examples

[0199] For example, smart trash cans installed in urban areas use various sensors to collect information on the characteristics of waste and user emotional data. The sensors detect anxiety and stress felt by users when throwing away trash, and this information is sent to a server. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[0200] At the same time, the emotion engine analyzes the user's emotions, and if it determines that the user is experiencing high levels of anxiety or stress, for example, easy-to-understand recycling instruction videos and positive messages are delivered to the user via the device.

[0201] The server also uses the classification results to design optimal collection routes and provides feedback to waste collection companies, allowing them to carry out collection work according to the optimal plan.For non-recyclable waste, the server notifies users of sustainable practices, promoting environmental education and raising awareness.

[0202] This will enable efficient waste management, promote recycling, and use emotional data to personalize sustainable practices.

[0203] The processing flow will be explained below.

[0204] Step 1:

[0205] The device uses various sensors to collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and simultaneously acquires the user's emotional data (e.g., facial expression, tone of voice, etc.). The collected information is converted into digital form in real time.

[0206] Step 2:

[0207] The terminal transmits the collected data (waste characteristic information and user emotion data) to a server via the Internet or a local network.

[0208] Step 3:

[0209] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[0210] Step 4:

[0211] The server divides the preprocessed data into features (characteristic information about the waste) and labels (whether it can be recycled or not), forming the dataset needed for training the machine learning model.

[0212] Step 5:

[0213] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[0214] Step 6:

[0215] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[0216] Step 7:

[0217] The server runs an emotion engine to analyze the user's emotion data, for example, by using facial recognition or voice analysis to estimate the user's emotion and record the results in a database.

[0218] Step 8:

[0219] The server personalizes sustainable practice information based on the user's emotions recognized by the emotion engine. For example, if the user is feeling anxious, it will recommend an easy-to-understand video on recycling.

[0220] Step 9:

[0221] The server sends the classification results and personalized sustainable practice information to the terminal, where the user can receive the information.

[0222] Step 10:

[0223] The device displays the information received from the server to the user, allowing them to learn how to properly classify waste, how to recycle, and other emotionally relevant practices.

[0224] Step 11:

[0225] The server then designs optimal garbage collection routes based on the classification results and provides feedback to garbage collection companies, enabling them to collect waste efficiently and along optimal routes.

[0226] Step 12:

[0227] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[0228] Example 2

[0229] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0230] Conventional waste management systems not only have difficulty collecting information on waste characteristics and classifying them appropriately, but also lack the ability to provide sustainable practice information that takes users' emotions into account. This makes it difficult to maintain users' interest in recycling and manage waste efficiently. Furthermore, there is a lack of means to provide appropriate support to users to alleviate the anxiety and stress they may feel during the waste sorting process.

[0231] 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.

[0232] In this invention, the server includes means for collecting waste characteristic information, means for collecting user emotion data, means for transmitting the collected data to the server, means for learning a specific classification model using data generated on the server, means for classifying waste using the classification model, means for analyzing the emotion data and recognizing the user's emotion, means for generating personalized sustainable practice information based on the classification results and the emotion analysis results, means for transmitting the generated sustainable practice information to the terminal, and means for delivering the sustainable practice information to the terminal, thereby enabling efficient waste classification and management and personalization of sustainable practice information based on the user's emotion.

[0233] "Waste characteristic information" refers to the physical and chemical attributes of waste materials, such as type, color, and weight.

[0234] "User emotion data" is data related to emotions obtained by analyzing the user's facial expressions, tone of voice, body movements, etc. when sorting waste.

[0235] "Server" refers to the central computer system for storing, analyzing, and processing collected data.

[0236] "Specific classification model" refers to a machine learning algorithm for classifying waste as recyclable or not.

[0237] "Emotion engine" refers to an algorithm for analyzing a user's emotion data and recognizing and classifying the emotion.

[0238] "Personalized sustainable practice information" refers to information and guidance on sustainable waste management and recycling that is individually optimized based on the user's emotional data.

[0239] "Terminal" refers to a device for collecting waste characteristic information and user emotion data, and for delivering sustainable practice information from a server to users.

[0240] "Emotion analysis results" refers to the results of the user's emotion data analyzed by the emotion engine, and refers to information indicating what emotions the user is feeling.

[0241] This invention combines a system that uses generative AI models to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[0242] The sensors on the device collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and also use a camera and microphone to simultaneously acquire the user's emotional data (facial expression, tone of voice, etc.). Specific examples of sensors include image sensors and weight sensors, and it is desirable for the camera to have high resolution. The collected information is converted into digital format in real time and sent to a server via the Internet or local network. The server requires a high-performance processor and large amount of memory, and it is recommended to use a database system such as MongoDB or MySQL.

[0243] The server stores the received data and performs preprocessing. Preprocessing includes data cleaning (removing noise and missing values) and formatting. Python scripts and the Pandas library can be used for data preprocessing. The server then splits the preprocessed data into features and labels and trains a machine learning model. The server uses machine learning libraries such as TensorFlow and PyTorch to build and train a classification model.

[0244] The machine learning model classifies waste as recyclable or not. This classification result contributes to more efficient waste management. In addition, the server is equipped with an emotion engine that analyzes and classifies users' emotional data. The emotion engine uses TensorFlow to analyze the user's facial expression data and recognizes emotions such as "anxiety" and "stress." Based on these emotions, the engine personalizes sustainable practice information and sends the generated information to the device. For example, if the emotion engine recognizes high levels of "anxiety" or "stress," it will provide the user with easy-to-understand recycling instruction videos and positive messages.

[0245] The device receives personalized sustainable practice information sent from the server and delivers it to the user via a display screen and audio output device, helping users deepen their understanding of recycling and providing information that helps reduce stress.

[0246] Examples:

[0247] Smart trash cans installed in urban areas use various sensors (image sensors, weight sensors, etc.) to collect information on the characteristics of waste. When a user throws away a plastic bottle, their facial expressions and tone of voice are collected by a camera and microphone, and this information is sent to a server. The server's emotion engine analyzes the user's emotional data, and if it determines that the user is experiencing high levels of "anxiety" or "stress," it will provide the user with recycling instruction videos and positive messages via their device.

[0248] Example prompt sentence:

[0249] "Analyze facial and vocal data as users sort waste, and suggest guidance and messages to provide if they feel anxious or stressed."

[0250] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0251] Step 1:

[0252] The terminal collects waste characteristic information and user emotion data.

[0253] Specific operation: The device is equipped with an image sensor, weight sensor, camera, and microphone to collect characteristic information such as the type, color, and weight of the waste material. The camera and microphone also collect the user's facial expression and tone of voice to obtain emotional data.

[0254] Input: waste (bottles, paper, etc.), user's face, voice

[0255] Output: Waste characteristic information (material, color, weight, etc.), user emotion data (facial expression, tone of voice)

[0256] Step 2:

[0257] The terminal transmits the collected data to the server.

[0258] What it does: Converts collected data into digital form in real time and sends it to a server over the Internet or local network using HTTP or HTTPS protocols.

[0259] Input: waste characteristics, user emotion data

[0260] Output: Raw data sent to the server

[0261] Step 3:

[0262] The server stores the received data and performs preprocessing.

[0263] Specific operations: The received data is stored in a database (e.g., MongoDB, MySQL), and the data is cleaned (removed noise and missing values) and formatted using Python scripts and the Pandas library.

[0264] Input: Raw data (waste characteristics, user emotion data)

[0265] Output: Preprocessed data

[0266] Step 4:

[0267] The server uses the preprocessed data to train a machine learning model.

[0268] Specific operation: Split the preprocessed data into features and labels, and train a neural network model using TensorFlow or PyTorch.

[0269] Input: Preprocessed data (features, labels)

[0270] Output: A trained classification model

[0271] Step 5:

[0272] The server classifies the waste as recyclable or not.

[0273] How it works: Using the trained classification model, newly received waste data is classified as recyclable or not in real time.

[0274] Input: New waste data

[0275] Output: Waste classification results (recyclable or not)

[0276] Step 6:

[0277] The server analyzes the user's emotional data using an emotion engine and generates personalized sustainable practice information.

[0278] Specific operation: Analyzes user emotion data using TensorFlow and recognizes emotions such as "anxiety" and "stress." Based on the recognition results, generates personalized sustainable practice information using Jinja2 templates.

[0279] Input: User emotion data

[0280] Output: Personalized sustainable practice information

[0281] Step 7:

[0282] The server sends the generated information to the terminal, which then distributes it to the user.

[0283] Specific operation: The generated sustainable practice information is sent to the terminal, which receives it and delivers it to the user using a display screen or audio output device.

[0284] Input: Personalized sustainable practice information

[0285] Output: Providing information to users (e.g., recycling instruction videos, positive messages)

[0286] This details how each processing step collects, transmits, analyzes, generates, and distributes data.

[0287] (Application example 2)

[0288] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0289] Conventional waste management systems lack the means to streamline waste sorting and collection, and are unable to provide sustainable practice information that takes into account the emotions of users. Especially in the case of waste disposal in stores, employees often feel stressed and anxious, which can lead to delays in proper waste sorting and recycling.

[0290] 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.

[0291] In this invention, the server includes means for collecting generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for collecting user emotion data, and means for analyzing the emotion data to personalize the sustainable practice information, thereby enabling efficient waste classification and optimization of collection routes, and further enabling the provision of personalized sustainable practice information based on the user's emotions.

[0292] "Generated data" refers to all digital information collected from devices, including emotional data and waste information.

[0293] The "collection means" refers to devices such as sensors and cameras for acquiring waste characteristic information and user emotional data.

[0294] The "means for transmitting to the server" is a communication module or protocol for transferring collected data to a server on the cloud via the Internet or a local network.

[0295] A "specific classification model" is a model trained using a machine learning algorithm to classify waste as recyclable or non-recyclable.

[0296] The "classification means" is a component that inputs collected data into a specific classification model and processes it to classify waste as recyclable or not.

[0297] The "means for optimizing collection routes" refers to algorithms or systems that design optimal collection routes for waste collection companies based on the classification results and enable efficient waste collection.

[0298] "Sustainable practice information" is information that helps users take sustainable actions, such as recycling methods and knowledge about environmental protection.

[0299] The "means of distribution" refers to a module that displays information directly on the terminal, or a communication system that sends notifications to the user's smartphone.

[0300] "Emotion data" is digital information that indicates the user's emotional state, obtained from the user's facial expression, tone of voice, etc.

[0301] The "analysis means" is a software module that recognizes the user's emotions based on the collected emotional data and determines actions based on the results.

[0302] A "personalization means" is an algorithm or system that uses acquired emotional data from a user to provide sustainable practice information optimized for that user.

[0303] This invention is a system that improves the efficiency of waste management in stores and provides sustainable practice information based on user sentiment. The system operates as follows.

[0304] First, the device is equipped with a camera and various sensors to collect waste characteristic information and the user's emotional data. Waste characteristic information includes the type of material, color, and weight, while the user's emotional data includes facial expressions and tone of voice. Each piece of data is digitized in real time and sent to a cloud server via the Internet or a local network.

[0305] The server stores the received data and first cleans and formats it. Then, it uses the data generated on the server to train a specific classification model. This classification model is trained using a machine learning algorithm (using TensorFlow as an example) to classify waste as recyclable or not.

[0306] The server is also equipped with an emotion engine that analyzes the user's emotion data. Based on the results obtained from the emotion engine, personalized sustainable practice information (e.g., recycling instructions and positive messages to reduce stress) is created in real time and delivered to the device. Examples of this sustainable practice information include easy-to-understand recycling instruction videos and positive messages.

[0307] As a concrete example, when an employee scans waste at a store's waste disposal station with their smartphone, the user's (employee's) facial expression is captured by the camera. If the employee shows any signs of anxiety or stress, that data is collected. The collected data is sent to a server and analyzed by an emotion engine. An easy-to-understand video on recycling methods is then displayed on the device to the employee.

[0308] This will enable users to properly separate waste without stress, and the system as a whole will promote efficient waste management and recycling.

[0309] Examples of prompts include:

[0310] "Generate a scenario where a user scans a waste item. The user feels anxious. Suggest a flowchart to provide appropriate support to this user."

[0311] This invention can significantly reduce a store's environmental impact through efficient waste sorting, optimised collection routes and the provision of personalized sustainable practice information based on user sentiment.

[0312] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0313] Step 1:

[0314] The device uses sensors and cameras to collect characteristic information about the waste (such as type of material, color, and weight) and the user's emotional data (such as facial expressions and tone of voice).

[0315] Input: Waste characteristics, user emotion data

[0316] Data processing: Converting collected data into a digital format

[0317] Output: Digitized waste information and emotion data

[0318] Step 2:

[0319] The terminal transmits the digitized waste information and emotion data to a server via the Internet or a local network.

[0320] Input: Digitized waste information and emotion data

[0321] Data processing: Packetizing data according to network protocols

[0322] Output: Data packet sent to the server

[0323] Step 3:

[0324] The server stores the received data in a database and cleans and formats it.

[0325] Input: Data packet (waste information and emotion data)

[0326] Data processing: Data cleaning (insertion of missing values, removal of invalid data) and formatting

[0327] Output: Preprocessed data

[0328] Step 4:

[0329] The server splits the preprocessed data into features and labels and trains a machine learning model.

[0330] Input: Preprocessed data

[0331] Data Processing: Feature Engineering and Labeling for Machine Learning

[0332] Output: A trained classification model

[0333] Step 5:

[0334] The server uses a trained classification model to classify waste as recyclable or not.

[0335] Input: Real-time collected waste data

[0336] Data processing: Analyzing and classifying data using classification models

[0337] Output: Classification result as recyclable or not

[0338] Step 6:

[0339] The server designs the optimal collection route based on the classification results and provides feedback to the contractor.

[0340] Input: Classification results

[0341] Data processing: Applying collection route optimization algorithms

[0342] Output: Optimized collection route

[0343] Step 7:

[0344] The server analyzes the collected emotion data and generates sustainable practice information based on the user's emotions.

[0345] Input: Emotion data

[0346] Data processing: Analysis and information generation using emotion engines

[0347] Output: Personalized sustainable practice information

[0348] Step 8:

[0349] The server distributes the generated sustainable practice information to the terminal.

[0350] Input: Personalized sustainable practice information

[0351] Data processing: Sending information to the user's device

[0352] Output: Sustainable practice information displayed on a terminal

[0353] This series of processing steps enables efficient classification and management of waste and provides information based on the user's emotions.

[0354] 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.

[0355] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0356] 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.

[0357] [Second embodiment]

[0358] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0359] 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.

[0360] 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).

[0361] 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.

[0362] 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.

[0363] 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).

[0364] 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. 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.

[0365] 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.

[0366] 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.

[0367] 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.

[0368] 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.

[0369] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0370] This invention is a system that aims to reduce waste and promote recycling by using generative AI. Specifically, terminals installed in each supermarket and community send collected waste data to a server in real time, and the server analyzes and optimizes this data to achieve efficient waste management.

[0371] System Overview

[0372] The sensors in the device collect characteristic information about the waste (e.g., material type, color, weight, etc.) and convert it into digital format as generated data. The collected data is then sent to a server via the Internet or a local network.

[0373] The server receives the data, stores it, and pre-processes it, which includes cleaning and formatting it. The server then splits the pre-processed data into features and labels to train a machine learning model, a specific classification model, that classifies waste as recyclable or not.

[0374] System Operation

[0375] After training the classification model, the server analyzes newly collected data in real time and classifies the waste. Based on the classification results, it designs optimal collection routes and delivers relevant sustainable practice information to users via their devices.

[0376] Specific examples

[0377] For example, smart trash cans installed around town use various sensors to collect information about the characteristics of waste. This information is sent to a server via the user's smartphone or dedicated device. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[0378] The server then uses the classification results to plan optimal collection routes and provide feedback to waste collection companies. It also delivers educational content about non-recyclable waste to users' devices and provides information on sustainable practices, helping users take appropriate recycling actions.

[0379] In this way, the system contributes to efficient waste management, promoting recycling, and realizing a sustainable society through education.

[0380] The processing flow will be explained below.

[0381] Step 1:

[0382] The terminal uses various sensors to collect information about the characteristics of the waste (such as the type of material, color, weight, etc.), which is then converted into a digital format in real time.

[0383] Step 2:

[0384] The terminal transmits the collected data to a server via the Internet or a local network. The transmitted data includes information about the characteristics of the waste.

[0385] Step 3:

[0386] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[0387] Step 4:

[0388] The server separates the preprocessed data into features (e.g., material type, color, weight) and labels (e.g., recyclable / non-recyclable), forming the dataset needed to train the machine learning model.

[0389] Step 5:

[0390] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[0391] Step 6:

[0392] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[0393] Step 7:

[0394] The server then designs the optimal garbage collection route based on the classification results, enabling efficient collection and reducing costs and environmental impact.

[0395] Step 8:

[0396] The server provides feedback to waste collection companies on the designed collection routes and recycling information, allowing them to carry out collection according to the optimal plan.

[0397] Step 9:

[0398] The server generates educational content about non-recyclable waste and delivers sustainable practice information to users via their devices, allowing them to learn about correct recycling methods and environmentally friendly behavior.

[0399] Step 10:

[0400] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[0401] Example 1

[0402] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0403] To efficiently manage waste and promote recycling, it is necessary to classify waste, optimize collection routes, and provide information on sustainable practices. However, conventional systems lack the ability to adequately control the quality of collected data, train classification models, and analyze data in real time, making appropriate waste management difficult. Furthermore, they are also inadequate in providing users with the information they need on sustainable practices, limiting their contribution to a sustainable society.

[0404] 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.

[0405] In this invention, the server includes means for the terminal to collect waste characteristic information using sensors, means for converting the collected data into digital form and sending it to the server, means for storing, cleaning, and formatting the received data on the server, means for dividing the preprocessed data into features and labels and training a machine learning model, means for analyzing the new data in real time and classifying the waste as recyclable or not, means for designing optimal waste collection routes based on the classification results, and means for delivering sustainable practice information to the user through the terminal, thereby enabling efficient waste management and promotion of recycling, as well as providing users with prompt and appropriate information.

[0406] The "terminal" is a device equipped with a sensor to collect characteristic information about waste, convert the collected data into digital format, and send it to a server.

[0407] "Server" is a computer system for receiving, storing, and pre-processing collected data and for training and running machine learning models.

[0408] A "sensor" is a measuring device installed in a terminal to detect and collect characteristic information such as the material, color, and weight of waste.

[0409] "Characteristic information" is data that indicates the material, color, weight, etc. of the waste, and is collected by sensors.

[0410] "Digital format" refers to data that has been converted from analog data so that it can be processed electronically.

[0411] "Data cleaning" is the process of removing noise and missing values ​​from collected data, leaving only the necessary information.

[0412] "Formatting" refers to the conversion of data of different formats or structures into a unified format or structure.

[0413] A "feature" is a value that indicates a specific attribute or characteristic of the data used to train a machine learning model.

[0414] A "label" is a value that indicates the output or classification result corresponding to a feature in a machine learning model.

[0415] A "machine learning model" is an algorithm that learns patterns and relationships from given data and makes predictions and classifications for new data.

[0416] "Classification" is the process of grouping received data based on predefined categories or labels.

[0417] A "waste collection route" is a route used by a waste collection vehicle to efficiently collect waste.

[0418] "Sustainable practice information" is information on actions aimed at realizing a sustainable society, such as reducing environmental impact and promoting recycling.

[0419] "User" means any person or organization that uses this system and receives guidance on how to properly dispose of and recycle waste.

[0420] This invention is a system that aims to reduce waste and promote recycling by utilizing generative AI models. In this system, terminals installed in each supermarket and community send waste data in real time to a server, which then analyzes and optimizes this data to achieve efficient waste management.

[0421] System configuration

[0422] 1. The terminal is equipped with multiple sensors. These sensors collect characteristic information about the waste (such as material type, color, and weight) and convert it into digital form. Specific examples of sensors include material detection sensors, color-identification cameras, and weight sensors.

[0423] 2. The collected data is sent to a server via the Internet or a local network. The device transmits this data in real time, ensuring that it arrives at the server without delay.

[0424] 3. The server first stores the received data, then cleans and formats it. The software used includes database management systems and data cleaning tools, such as PostgreSQL and pandas (a Python library).

[0425] 4. The server splits the preprocessed data into features and labels to train a machine learning model. This model is implemented using the Python machine learning library scikit-learn. In particular, it uses a random forest classifier to classify whether the waste is recyclable or not.

[0426] 5. The server uses the trained model to analyze newly collected data in real time and classify the waste. Based on this classification, it designs optimal waste collection routes. Collection route planning is done using Geographic Information System (GIS) software.

[0427] 6. The server sends the sorting results and collection route information to the device and also provides the user with sustainable practice information, such as educational content about non-recyclable waste and the location of specific recycling facilities.

[0428] Specific examples

[0429] For example, a smart trash can installed in a city uses various sensors to collect information about the characteristics of waste. When a user throws a plastic bottle into the trash can, a material detection sensor detects the plastic, a color-identification camera identifies the color blue, and a weight sensor measures 200g. This data is converted into digital format and sent to a server in real time.

[0430] The server stores the received data, cleans and formats it using pandas, then splits the data ("plastic, blue, 200g") into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[0431] The server analyzes newly received data in real time and classifies it as "recyclable." Based on this classification, the GIS software designs optimal garbage collection routes and provides them to collection companies. At the same time, the user receives a notification via their device saying, "This waste is recyclable. Please separate it appropriately."

[0432] Below is an example of a prompt sentence:

[0433] "What is the recycling status of the waste I put in yesterday?"

[0434] In this way, the present invention can efficiently manage waste and promote recycling, and further contributes to the realization of a sustainable society through user education.

[0435] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0436] Step 1:

[0437] The terminal uses sensors to collect information about the characteristics of the waste. The terminal is equipped with a material detection sensor, a color recognition camera, a weight sensor, etc. The data obtained by these sensors is information about the material, color, and weight of the waste.

[0438] Input: Waste

[0439] Specific operation: The sensor detects plastic waste, the color-detection camera identifies the color blue, and the weight sensor measures 200g.

[0440] Output: Feature information: "Plastic, Blue, 200g"

[0441] Step 2:

[0442] The device converts the collected data into a digital format and transmits it to a server via the internet or a local network. The device transmits this data in real time, so it arrives at the server without delay.

[0443] Input: Feature information "Plastic, Blue, 200g"

[0444] Specific operation: The device converts the analog data into digital format and sends the data to the specified address on the server.

[0445] Output: Digital characteristic information (data sent to the server)

[0446] Step 3:

[0447] The server stores, cleans, and formats the data it receives using a database management system (e.g., PostgreSQL) and a data cleaning tool (e.g., pandas).

[0448] Input: Digital feature information

[0449] Specific operation: The server stores the received data in a database and performs cleaning such as filling in missing values ​​and removing outliers. After that, the data format is unified.

[0450] Output: Preprocessed feature information

[0451] Step 4:

[0452] The server splits the preprocessed data into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[0453] Input: Preprocessed feature information

[0454] Specific operation: The server splits the data "plastic, blue, 200g" into "features (plastic, blue, 200g)" and "label (recyclable)" and supplies them to the random forest classifier as training data.

[0455] Output: A trained machine learning model

[0456] Step 5:

[0457] The server analyzes newly received data in real time and uses machine learning models to classify the waste.

[0458] Input: Newly received data (e.g. "glass, blue, 300g")

[0459] What it does: Use the trained model to analyze new data: "glass, blue, 300g" and classify it as "recyclable."

[0460] Output: Classification result (e.g. "Recyclable")

[0461] Step 6:

[0462] The server uses Geographic Information System (GIS) software to design optimal waste collection routes based on the classification results and provide feedback to collection companies.

[0463] Input: Classification results and permitted information (e.g. waste type and collection amount)

[0464] How it works: The server analyzes all waste data, calculates efficient collection routes based on the amount and type of waste, and provides collection schedules and efficient route guidance to collection companies.

[0465] Output: Optimized collection route information

[0466] Step 7:

[0467] The server delivers sustainable practice information to users through their devices, including, for example, educational content about non-recyclable waste and the locations of specific recycling facilities.

[0468] Input: Classification results and sustainable practice information

[0469] What it does: Generates an educational video on "How to dispose of non-recyclable plastics" and sends a notification to the user's smartphone.

[0470] Output: User notification and educational content

[0471] The above is the processing flow of this system.

[0472] (Application example 1)

[0473] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0474] In modern society, the increase in waste and the need for recycling have created a demand for efficient waste management. However, proper sorting and promotion of recycling are often insufficient. Brick-and-mortar stores in particular generate large amounts of waste, and a system is needed to efficiently manage this waste and promote sustainable practices.

[0475] 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.

[0476] In this invention, the server includes means for collecting the generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for providing the waste classification results in real time through the user's information terminal, and means for delivering educational content for promoting recycling to the user's information terminal, thereby enabling efficient waste management, promotion of recycling, and realization of a sustainable society.

[0477] "Generated data" means waste characteristics collected by a waste management system and converted into digital form.

[0478] The "server" is a central control unit that receives collected data and performs analysis and training of classification models.

[0479] A "classification model" is a model that uses machine learning algorithms to classify waste as recyclable or not.

[0480] "Collection route optimization" is a method for designing optimal waste collection routes based on classification results, thereby achieving efficient collection activities.

[0481] "Sustainable Practice Information" is educational content and guidelines for recycling and waste reduction provided to users.

[0482] "User's information terminal" refers to a device such as a smartphone or tablet that is used to receive and display information in real time.

[0483] "Means for providing waste classification results in real time" refers to a function that displays the classification results on the user's information terminal the moment the waste is collected.

[0484] "Educational content to promote recycling" is content that provides users with information on the correct methods of waste disposal and recycling.

[0485] As an embodiment of this invention, the following system can be constructed. This system improves the efficiency of waste management in physical stores, promotes recycling, and supports sustainable practices. The system is mainly composed of a server, terminals, and users.

[0486] The server has the following functions: First, it collects data generated by sensors and transmits it via the Internet or local network. This digitally aggregates waste characteristic information (such as material type, color, and weight) on the server. Next, the server analyzes the received data and performs preprocessing (data cleaning, formatting, etc.). It uses the preprocessed data to train a specific classification model, which is then used to classify waste as recyclable or not. It designs optimal collection routes based on the classification results and delivers this information to users' devices in real time. It also provides users with educational content to promote recycling as needed.

[0487] The device has the function of receiving and displaying sustainable practice information and waste classification results sent from the server, allowing users to check important information on how to correctly classify waste and recycling in real time. Specifically, this applies to users' information devices such as smartphones and tablets.

[0488] Users receive information from the server via their devices and implement appropriate waste management practices, which promotes efficient waste disposal and recycling. Furthermore, educational content helps users deepen their own understanding of sustainable practices.

[0489] For example, supermarket staff can send data obtained from waste management devices (smart trash cans) to a server via a smartphone app. This data is analyzed on the server, and real-time information on whether the waste can be recycled is provided to staff. Based on this information, staff can properly classify the waste. The app also provides educational content about non-recyclable waste, promoting sustainable behavior.

[0490] Example prompt sentence:

[0491] Write a program that uses data obtained from waste management terminals by supermarket staff to determine in real time whether waste can be recycled and displays the results on the staff's smartphone app, "Eco Shopper." Data obtained from the sensors includes material type, color, and weight. Based on the results, provide staff with information about recyclability.

[0492] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0493] Step 1:

[0494] The server collects the data generated by the sensors. Specifically, the sensors in the waste management terminal collect characteristic information such as the type, color, and weight of the waste material, converts this data into a digital format, and transmits it to the server via the Internet or a local network.

[0495] Input: Waste characteristics (type of material, color, weight)

[0496] Output: Digital feature information data

[0497] Step 2:

[0498] The server preprocesses the received data. Specifically, the server performs data cleaning and formatting to generate preprocessed data.

[0499] Input: Digital feature data

[0500] Output: Preprocessed data

[0501] Step 3:

[0502] The server uses the preprocessed data to train a classification model. Specifically, it splits the preprocessed data into features and labels, and uses them as input to train a machine learning model (e.g., random forest).

[0503] Input: Preprocessed data

[0504] Output: A trained classification model

[0505] Step 4:

[0506] The server classifies newly collected waste data by analyzing the newly received data using a trained classification model and classifying the waste as recyclable or not in real time.

[0507] Input: Newly collected waste data

[0508] Output: Waste classification results

[0509] Step 5:

[0510] The server optimizes collection routes based on the classification results, specifically by designing optimal collection routes depending on the type and amount of waste, and preparing feedback for collection companies.

[0511] Input: Waste classification results

[0512] Output: Optimized collection route information

[0513] Step 6:

[0514] The server delivers sustainable practice information and classification results to the device, specifically, information on recyclable waste and educational content to promote recycling to the user's smartphone or tablet.

[0515] Input: Sustainable practice information and classification results

[0516] Output: Information displayed on the user's terminal

[0517] Step 7:

[0518] The user's device displays the received information. Specifically, the smartphone app displays the classification results and sustainable practice information in real time, and guides the user to appropriate waste disposal methods.

[0519] Input: Information from the server (classification results, sustainable practice information)

[0520] Output: Information displayed on the terminal screen

[0521] 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.

[0522] This invention combines a system that uses generative AI to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[0523] System Overview

[0524] The device's sensors collect waste characteristics (e.g., material type, color, weight, etc.) and simultaneously acquire the user's emotional data. For example, the device can recognize the user's emotions by analyzing their facial expressions and tone of voice as they sort the waste. The collected information is converted into digital form in real time and sent to a server via the Internet or a local network.

[0525] System Operation

[0526] The server stores the received data and pre-processes it. This includes cleaning and formatting the data. The server then splits the pre-processed data into features and labels to train a machine learning model. This model is a specific classification model that classifies waste as recyclable or not.

[0527] Furthermore, the server is equipped with an emotion engine that analyzes and classifies the user's emotion data. Based on the emotions recognized by the emotion engine, the server personalizes sustainable practice information and delivers it to the user via their device.

[0528] Specific examples

[0529] For example, smart trash cans installed in urban areas use various sensors to collect information on the characteristics of waste and user emotional data. The sensors detect anxiety and stress felt by users when throwing away trash, and this information is sent to a server. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[0530] At the same time, the emotion engine analyzes the user's emotions, and if it determines that the user is experiencing high levels of anxiety or stress, for example, easy-to-understand recycling instruction videos and positive messages are delivered to the user via the device.

[0531] The server also uses the classification results to design optimal collection routes and provides feedback to waste collection companies, allowing them to carry out collection work according to the optimal plan.For non-recyclable waste, the server notifies users of sustainable practices, promoting environmental education and raising awareness.

[0532] This will enable efficient waste management, promote recycling, and use emotional data to personalize sustainable practices.

[0533] The processing flow will be explained below.

[0534] Step 1:

[0535] The device uses various sensors to collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and simultaneously acquires the user's emotional data (e.g., facial expression, tone of voice, etc.). The collected information is converted into digital form in real time.

[0536] Step 2:

[0537] The terminal transmits the collected data (waste characteristic information and user emotion data) to a server via the Internet or a local network.

[0538] Step 3:

[0539] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[0540] Step 4:

[0541] The server divides the preprocessed data into features (characteristic information about the waste) and labels (whether it can be recycled or not), forming the dataset needed for training the machine learning model.

[0542] Step 5:

[0543] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[0544] Step 6:

[0545] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[0546] Step 7:

[0547] The server runs an emotion engine to analyze the user's emotion data, for example, by using facial recognition or voice analysis to estimate the user's emotion and record the results in a database.

[0548] Step 8:

[0549] The server personalizes sustainable practice information based on the user's emotions recognized by the emotion engine. For example, if the user is feeling anxious, it will recommend an easy-to-understand video on recycling.

[0550] Step 9:

[0551] The server sends the classification results and personalized sustainable practice information to the terminal, where the user can receive the information.

[0552] Step 10:

[0553] The device displays the information received from the server to the user, allowing them to learn how to properly classify waste, how to recycle, and other emotionally relevant practices.

[0554] Step 11:

[0555] The server then designs optimal garbage collection routes based on the classification results and provides feedback to garbage collection companies, enabling them to collect waste efficiently and along optimal routes.

[0556] Step 12:

[0557] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[0558] Example 2

[0559] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0560] Conventional waste management systems not only have difficulty collecting information on waste characteristics and classifying them appropriately, but also lack the ability to provide sustainable practice information that takes users' emotions into account. This makes it difficult to maintain users' interest in recycling and manage waste efficiently. Furthermore, there is a lack of means to provide appropriate support to users to alleviate the anxiety and stress they may feel during the waste sorting process.

[0561] 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.

[0562] In this invention, the server includes means for collecting waste characteristic information, means for collecting user emotion data, means for transmitting the collected data to the server, means for learning a specific classification model using data generated on the server, means for classifying waste using the classification model, means for analyzing the emotion data and recognizing the user's emotion, means for generating personalized sustainable practice information based on the classification results and the emotion analysis results, means for transmitting the generated sustainable practice information to the terminal, and means for delivering the sustainable practice information to the terminal, thereby enabling efficient waste classification and management and personalization of sustainable practice information based on the user's emotion.

[0563] "Waste characteristic information" refers to the physical and chemical attributes of waste materials, such as type, color, and weight.

[0564] "User emotion data" is data related to emotions obtained by analyzing the user's facial expressions, tone of voice, body movements, etc. when sorting waste.

[0565] "Server" refers to the central computer system for storing, analyzing, and processing collected data.

[0566] "Specific classification model" refers to a machine learning algorithm for classifying waste as recyclable or not.

[0567] "Emotion engine" refers to an algorithm for analyzing a user's emotion data and recognizing and classifying the emotion.

[0568] "Personalized sustainable practice information" refers to information and guidance on sustainable waste management and recycling that is individually optimized based on the user's emotional data.

[0569] "Terminal" refers to a device for collecting waste characteristic information and user emotion data, and for delivering sustainable practice information from a server to users.

[0570] "Emotion analysis results" refers to the results of the user's emotion data analyzed by the emotion engine, and refers to information indicating what emotions the user is feeling.

[0571] This invention combines a system that uses generative AI models to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[0572] The sensors on the device collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and also use a camera and microphone to simultaneously acquire the user's emotional data (facial expression, tone of voice, etc.). Specific examples of sensors include image sensors and weight sensors, and it is desirable for the camera to have high resolution. The collected information is converted into digital format in real time and sent to a server via the Internet or local network. The server requires a high-performance processor and large amount of memory, and it is recommended to use a database system such as MongoDB or MySQL.

[0573] The server stores the received data and performs preprocessing. Preprocessing includes data cleaning (removing noise and missing values) and formatting. Python scripts and the Pandas library can be used for data preprocessing. The server then splits the preprocessed data into features and labels and trains a machine learning model. The server uses machine learning libraries such as TensorFlow and PyTorch to build and train a classification model.

[0574] The machine learning model classifies waste as recyclable or not. This classification result contributes to more efficient waste management. In addition, the server is equipped with an emotion engine that analyzes and classifies users' emotional data. The emotion engine uses TensorFlow to analyze the user's facial expression data and recognizes emotions such as "anxiety" and "stress." Based on these emotions, the engine personalizes sustainable practice information and sends the generated information to the device. For example, if the emotion engine recognizes high levels of "anxiety" or "stress," it will provide the user with easy-to-understand recycling instruction videos and positive messages.

[0575] The device receives personalized sustainable practice information sent from the server and delivers it to the user via a display screen and audio output device, helping users deepen their understanding of recycling and providing information that helps reduce stress.

[0576] Examples:

[0577] Smart trash cans installed in urban areas use various sensors (image sensors, weight sensors, etc.) to collect information on the characteristics of waste. When a user throws away a plastic bottle, their facial expressions and tone of voice are collected by a camera and microphone, and this information is sent to a server. The server's emotion engine analyzes the user's emotional data, and if it determines that the user is experiencing high levels of "anxiety" or "stress," it will provide the user with recycling instruction videos and positive messages via their device.

[0578] Example prompt sentence:

[0579] "Analyze facial and vocal data as users sort waste, and suggest guidance and messages to provide if they feel anxious or stressed."

[0580] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0581] Step 1:

[0582] The terminal collects waste characteristic information and user emotion data.

[0583] Specific operation: The device is equipped with an image sensor, weight sensor, camera, and microphone to collect characteristic information such as the type, color, and weight of the waste material. The camera and microphone also collect the user's facial expression and tone of voice to obtain emotional data.

[0584] Input: waste (bottles, paper, etc.), user's face, voice

[0585] Output: Waste characteristic information (material, color, weight, etc.), user emotion data (facial expression, tone of voice)

[0586] Step 2:

[0587] The terminal transmits the collected data to the server.

[0588] What it does: Converts collected data into digital form in real time and sends it to a server over the Internet or local network using HTTP or HTTPS protocols.

[0589] Input: waste characteristics, user emotion data

[0590] Output: Raw data sent to the server

[0591] Step 3:

[0592] The server stores the received data and performs preprocessing.

[0593] Specific operations: The received data is stored in a database (e.g., MongoDB, MySQL), and the data is cleaned (removed noise and missing values) and formatted using Python scripts and the Pandas library.

[0594] Input: Raw data (waste characteristics, user emotion data)

[0595] Output: Preprocessed data

[0596] Step 4:

[0597] The server uses the preprocessed data to train a machine learning model.

[0598] Specific operation: Split the preprocessed data into features and labels, and train a neural network model using TensorFlow or PyTorch.

[0599] Input: Preprocessed data (features, labels)

[0600] Output: A trained classification model

[0601] Step 5:

[0602] The server classifies the waste as recyclable or not.

[0603] How it works: Using the trained classification model, newly received waste data is classified as recyclable or not in real time.

[0604] Input: New waste data

[0605] Output: Waste classification results (recyclable or not)

[0606] Step 6:

[0607] The server analyzes the user's emotional data using an emotion engine and generates personalized sustainable practice information.

[0608] Specific operation: Analyzes user emotion data using TensorFlow and recognizes emotions such as "anxiety" and "stress." Based on the recognition results, generates personalized sustainable practice information using Jinja2 templates.

[0609] Input: User emotion data

[0610] Output: Personalized sustainable practice information

[0611] Step 7:

[0612] The server sends the generated information to the terminal, which then distributes it to the user.

[0613] Specific operation: The generated sustainable practice information is sent to the terminal, which receives it and delivers it to the user using a display screen or audio output device.

[0614] Input: Personalized sustainable practice information

[0615] Output: Providing information to users (e.g., recycling instruction videos, positive messages)

[0616] This details how each processing step collects, transmits, analyzes, generates, and distributes data.

[0617] (Application example 2)

[0618] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0619] Conventional waste management systems lack the means to streamline waste sorting and collection, and are unable to provide sustainable practice information that takes into account the emotions of users. Especially in the case of waste disposal in stores, employees often feel stressed and anxious, which can lead to delays in proper waste sorting and recycling.

[0620] 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.

[0621] In this invention, the server includes means for collecting generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for collecting user emotion data, and means for analyzing the emotion data to personalize the sustainable practice information, thereby enabling efficient waste classification and optimization of collection routes, and further enabling the provision of personalized sustainable practice information based on the user's emotions.

[0622] "Generated data" refers to all digital information collected from devices, including emotional data and waste information.

[0623] The "collection means" refers to devices such as sensors and cameras for acquiring waste characteristic information and user emotional data.

[0624] The "means for transmitting to the server" is a communication module or protocol for transferring collected data to a server on the cloud via the Internet or a local network.

[0625] A "specific classification model" is a model trained using a machine learning algorithm to classify waste as recyclable or non-recyclable.

[0626] The "classification means" is a component that inputs collected data into a specific classification model and processes it to classify waste as recyclable or not.

[0627] The "means for optimizing collection routes" refers to algorithms or systems that design optimal collection routes for waste collection companies based on the classification results and enable efficient waste collection.

[0628] "Sustainable practice information" is information that helps users take sustainable actions, such as recycling methods and knowledge about environmental protection.

[0629] The "means of distribution" refers to a module that displays information directly on the terminal, or a communication system that sends notifications to the user's smartphone.

[0630] "Emotion data" is digital information that indicates the user's emotional state, obtained from the user's facial expression, tone of voice, etc.

[0631] The "analysis means" is a software module that recognizes the user's emotions based on the collected emotional data and determines actions based on the results.

[0632] A "personalization means" is an algorithm or system that uses acquired emotional data from a user to provide sustainable practice information optimized for that user.

[0633] This invention is a system that improves the efficiency of waste management in stores and provides sustainable practice information based on user sentiment. The system operates as follows.

[0634] First, the device is equipped with a camera and various sensors to collect waste characteristic information and the user's emotional data. Waste characteristic information includes the type of material, color, and weight, while the user's emotional data includes facial expressions and tone of voice. Each piece of data is digitized in real time and sent to a cloud server via the Internet or a local network.

[0635] The server stores the received data and first cleans and formats it. Then, it uses the data generated on the server to train a specific classification model. This classification model is trained using a machine learning algorithm (using TensorFlow as an example) to classify waste as recyclable or not.

[0636] The server is also equipped with an emotion engine that analyzes the user's emotion data. Based on the results obtained from the emotion engine, personalized sustainable practice information (e.g., recycling instructions and positive messages to reduce stress) is created in real time and delivered to the device. Examples of this sustainable practice information include easy-to-understand recycling instruction videos and positive messages.

[0637] As a concrete example, when an employee scans waste at a store's waste disposal station with their smartphone, the user's (employee's) facial expression is captured by the camera. If the employee shows any signs of anxiety or stress, that data is collected. The collected data is sent to a server and analyzed by an emotion engine. An easy-to-understand video on recycling methods is then displayed on the device to the employee.

[0638] This will enable users to properly separate waste without stress, and the system as a whole will promote efficient waste management and recycling.

[0639] Examples of prompts include:

[0640] "Generate a scenario where a user scans a waste item. The user feels anxious. Suggest a flowchart to provide appropriate support to this user."

[0641] This invention can significantly reduce a store's environmental impact through efficient waste sorting, optimised collection routes and the provision of personalized sustainable practice information based on user sentiment.

[0642] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0643] Step 1:

[0644] The device uses sensors and cameras to collect characteristic information about the waste (such as type of material, color, and weight) and the user's emotional data (such as facial expressions and tone of voice).

[0645] Input: Waste characteristics, user emotion data

[0646] Data processing: Converting collected data into a digital format

[0647] Output: Digitized waste information and emotion data

[0648] Step 2:

[0649] The terminal transmits the digitized waste information and emotion data to a server via the Internet or a local network.

[0650] Input: Digitized waste information and emotion data

[0651] Data processing: Packetizing data according to network protocols

[0652] Output: Data packet sent to the server

[0653] Step 3:

[0654] The server stores the received data in a database and cleans and formats it.

[0655] Input: Data packet (waste information and emotion data)

[0656] Data processing: Data cleaning (insertion of missing values, removal of invalid data) and formatting

[0657] Output: Preprocessed data

[0658] Step 4:

[0659] The server splits the preprocessed data into features and labels and trains a machine learning model.

[0660] Input: Preprocessed data

[0661] Data Processing: Feature Engineering and Labeling for Machine Learning

[0662] Output: A trained classification model

[0663] Step 5:

[0664] The server uses a trained classification model to classify waste as recyclable or not.

[0665] Input: Real-time collected waste data

[0666] Data processing: Analyzing and classifying data using classification models

[0667] Output: Classification result as recyclable or not

[0668] Step 6:

[0669] The server designs the optimal collection route based on the classification results and provides feedback to the contractor.

[0670] Input: Classification results

[0671] Data processing: Applying collection route optimization algorithms

[0672] Output: Optimized collection route

[0673] Step 7:

[0674] The server analyzes the collected emotion data and generates sustainable practice information based on the user's emotions.

[0675] Input: Emotion data

[0676] Data processing: Analysis and information generation using emotion engines

[0677] Output: Personalized sustainable practice information

[0678] Step 8:

[0679] The server distributes the generated sustainable practice information to the terminal.

[0680] Input: Personalized sustainable practice information

[0681] Data processing: Sending information to the user's device

[0682] Output: Sustainable practice information displayed on a terminal

[0683] This series of processing steps enables efficient classification and management of waste and provides information based on the user's emotions.

[0684] 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.

[0685] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0686] 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.

[0687] [Third embodiment]

[0688] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0689] 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.

[0690] 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).

[0691] 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.

[0692] 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.

[0693] 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).

[0694] 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. 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.

[0695] 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.

[0696] 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.

[0697] 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.

[0698] 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.

[0699] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0700] This invention is a system that aims to reduce waste and promote recycling by using generative AI. Specifically, terminals installed in each supermarket and community send collected waste data to a server in real time, and the server analyzes and optimizes this data to achieve efficient waste management.

[0701] System Overview

[0702] The sensors in the device collect characteristic information about the waste (e.g., material type, color, weight, etc.) and convert it into digital format as generated data. The collected data is then sent to a server via the Internet or a local network.

[0703] The server receives the data, stores it, and pre-processes it, which includes cleaning and formatting it. The server then splits the pre-processed data into features and labels to train a machine learning model, a specific classification model, that classifies waste as recyclable or not.

[0704] System Operation

[0705] After training the classification model, the server analyzes newly collected data in real time and classifies the waste. Based on the classification results, it designs optimal collection routes and delivers relevant sustainable practice information to users via their devices.

[0706] Specific examples

[0707] For example, smart trash cans installed around town use various sensors to collect information about the characteristics of waste. This information is sent to a server via the user's smartphone or dedicated device. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[0708] The server then uses the classification results to plan optimal collection routes and provide feedback to waste collection companies. It also delivers educational content about non-recyclable waste to users' devices and provides information on sustainable practices, helping users take appropriate recycling actions.

[0709] In this way, the system contributes to efficient waste management, promoting recycling, and realizing a sustainable society through education.

[0710] The processing flow will be explained below.

[0711] Step 1:

[0712] The terminal uses various sensors to collect information about the characteristics of the waste (such as the type of material, color, weight, etc.), which is then converted into a digital format in real time.

[0713] Step 2:

[0714] The terminal transmits the collected data to a server via the Internet or a local network. The transmitted data includes information about the characteristics of the waste.

[0715] Step 3:

[0716] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[0717] Step 4:

[0718] The server separates the preprocessed data into features (e.g., material type, color, weight) and labels (e.g., recyclable / non-recyclable), forming the dataset needed to train the machine learning model.

[0719] Step 5:

[0720] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[0721] Step 6:

[0722] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[0723] Step 7:

[0724] The server then designs the optimal garbage collection route based on the classification results, enabling efficient collection and reducing costs and environmental impact.

[0725] Step 8:

[0726] The server provides feedback to waste collection companies on the designed collection routes and recycling information, allowing them to carry out collection according to the optimal plan.

[0727] Step 9:

[0728] The server generates educational content about non-recyclable waste and delivers sustainable practice information to users via their devices, allowing them to learn about correct recycling methods and environmentally friendly behavior.

[0729] Step 10:

[0730] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[0731] Example 1

[0732] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0733] To efficiently manage waste and promote recycling, it is necessary to classify waste, optimize collection routes, and provide information on sustainable practices. However, conventional systems lack the ability to adequately control the quality of collected data, train classification models, and analyze data in real time, making appropriate waste management difficult. Furthermore, they are also inadequate in providing users with the information they need on sustainable practices, limiting their contribution to a sustainable society.

[0734] 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.

[0735] In this invention, the server includes means for the terminal to collect waste characteristic information using sensors, means for converting the collected data into digital form and sending it to the server, means for storing, cleaning, and formatting the received data on the server, means for dividing the preprocessed data into features and labels and training a machine learning model, means for analyzing the new data in real time and classifying the waste as recyclable or not, means for designing optimal waste collection routes based on the classification results, and means for delivering sustainable practice information to the user through the terminal, thereby enabling efficient waste management and promotion of recycling, as well as providing users with prompt and appropriate information.

[0736] The "terminal" is a device equipped with a sensor to collect characteristic information about waste, convert the collected data into digital format, and send it to a server.

[0737] "Server" is a computer system for receiving, storing, and pre-processing collected data and for training and running machine learning models.

[0738] A "sensor" is a measuring device installed in a terminal to detect and collect characteristic information such as the material, color, and weight of waste.

[0739] "Characteristic information" is data that indicates the material, color, weight, etc. of the waste, and is collected by sensors.

[0740] "Digital format" refers to data that has been converted from analog data so that it can be processed electronically.

[0741] "Data cleaning" is the process of removing noise and missing values ​​from collected data, leaving only the necessary information.

[0742] "Formatting" refers to the conversion of data of different formats or structures into a unified format or structure.

[0743] A "feature" is a value that indicates a specific attribute or characteristic of the data used to train a machine learning model.

[0744] A "label" is a value that indicates the output or classification result corresponding to a feature in a machine learning model.

[0745] A "machine learning model" is an algorithm that learns patterns and relationships from given data and makes predictions and classifications for new data.

[0746] "Classification" is the process of grouping received data based on predefined categories or labels.

[0747] A "waste collection route" is a route used by a waste collection vehicle to efficiently collect waste.

[0748] "Sustainable practice information" is information on actions aimed at realizing a sustainable society, such as reducing environmental impact and promoting recycling.

[0749] "User" means any person or organization that uses this system and receives guidance on how to properly dispose of and recycle waste.

[0750] This invention is a system that aims to reduce waste and promote recycling by utilizing generative AI models. In this system, terminals installed in each supermarket and community send waste data in real time to a server, which then analyzes and optimizes this data to achieve efficient waste management.

[0751] System configuration

[0752] 1. The terminal is equipped with multiple sensors. These sensors collect characteristic information about the waste (such as material type, color, and weight) and convert it into digital form. Specific examples of sensors include material detection sensors, color-identification cameras, and weight sensors.

[0753] 2. The collected data is sent to a server via the Internet or a local network. The device transmits this data in real time, ensuring that it arrives at the server without delay.

[0754] 3. The server first stores the received data, then cleans and formats it. The software used includes database management systems and data cleaning tools, such as PostgreSQL and pandas (a Python library).

[0755] 4. The server splits the preprocessed data into features and labels to train a machine learning model. This model is implemented using the Python machine learning library scikit-learn. In particular, it uses a random forest classifier to classify whether the waste is recyclable or not.

[0756] 5. The server uses the trained model to analyze newly collected data in real time and classify the waste. Based on this classification, it designs optimal waste collection routes. Collection route planning is done using Geographic Information System (GIS) software.

[0757] 6. The server sends the sorting results and collection route information to the device and also provides the user with sustainable practice information, such as educational content about non-recyclable waste and the location of specific recycling facilities.

[0758] Specific examples

[0759] For example, a smart trash can installed in a city uses various sensors to collect information about the characteristics of waste. When a user throws a plastic bottle into the trash can, a material detection sensor detects the plastic, a color-identification camera identifies the color blue, and a weight sensor measures 200g. This data is converted into digital format and sent to a server in real time.

[0760] The server stores the received data, cleans and formats it using pandas, then splits the data ("plastic, blue, 200g") into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[0761] The server analyzes newly received data in real time and classifies it as "recyclable." Based on this classification, the GIS software designs optimal garbage collection routes and provides them to collection companies. At the same time, the user receives a notification via their device saying, "This waste is recyclable. Please separate it appropriately."

[0762] Below is an example of a prompt sentence:

[0763] "What is the recycling status of the waste I put in yesterday?"

[0764] In this way, the present invention can efficiently manage waste and promote recycling, and further contributes to the realization of a sustainable society through user education.

[0765] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0766] Step 1:

[0767] The terminal uses sensors to collect information about the characteristics of the waste. The terminal is equipped with a material detection sensor, a color recognition camera, a weight sensor, etc. The data obtained by these sensors is information about the material, color, and weight of the waste.

[0768] Input: Waste

[0769] Specific operation: The sensor detects plastic waste, the color-detection camera identifies the color blue, and the weight sensor measures 200g.

[0770] Output: Feature information: "Plastic, Blue, 200g"

[0771] Step 2:

[0772] The device converts the collected data into a digital format and transmits it to a server via the internet or a local network. The device transmits this data in real time, so it arrives at the server without delay.

[0773] Input: Feature information "Plastic, Blue, 200g"

[0774] Specific operation: The device converts the analog data into digital format and sends the data to the specified address on the server.

[0775] Output: Digital characteristic information (data sent to the server)

[0776] Step 3:

[0777] The server stores, cleans, and formats the data it receives using a database management system (e.g., PostgreSQL) and a data cleaning tool (e.g., pandas).

[0778] Input: Digital feature information

[0779] Specific operation: The server stores the received data in a database and performs cleaning such as filling in missing values ​​and removing outliers. After that, the data format is unified.

[0780] Output: Preprocessed feature information

[0781] Step 4:

[0782] The server splits the preprocessed data into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[0783] Input: Preprocessed feature information

[0784] Specific operation: The server splits the data "plastic, blue, 200g" into "features (plastic, blue, 200g)" and "label (recyclable)" and supplies them to the random forest classifier as training data.

[0785] Output: A trained machine learning model

[0786] Step 5:

[0787] The server analyzes newly received data in real time and uses machine learning models to classify the waste.

[0788] Input: Newly received data (e.g. "glass, blue, 300g")

[0789] What it does: Use the trained model to analyze new data: "glass, blue, 300g" and classify it as "recyclable."

[0790] Output: Classification result (e.g. "Recyclable")

[0791] Step 6:

[0792] The server uses Geographic Information System (GIS) software to design optimal waste collection routes based on the classification results and provide feedback to collection companies.

[0793] Input: Classification results and permitted information (e.g. waste type and collection amount)

[0794] How it works: The server analyzes all waste data, calculates efficient collection routes based on the amount and type of waste, and provides collection schedules and efficient route guidance to collection companies.

[0795] Output: Optimized collection route information

[0796] Step 7:

[0797] The server delivers sustainable practice information to users through their devices, including, for example, educational content about non-recyclable waste and the locations of specific recycling facilities.

[0798] Input: Classification results and sustainable practice information

[0799] What it does: Generates an educational video on "How to dispose of non-recyclable plastics" and sends a notification to the user's smartphone.

[0800] Output: User notification and educational content

[0801] The above is the processing flow of this system.

[0802] (Application example 1)

[0803] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0804] In modern society, the increase in waste and the need for recycling have created a demand for efficient waste management. However, proper sorting and promotion of recycling are often insufficient. Brick-and-mortar stores in particular generate large amounts of waste, and a system is needed to efficiently manage this waste and promote sustainable practices.

[0805] 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.

[0806] In this invention, the server includes means for collecting the generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for providing the waste classification results in real time through the user's information terminal, and means for delivering educational content for promoting recycling to the user's information terminal, thereby enabling efficient waste management, promotion of recycling, and realization of a sustainable society.

[0807] "Generated data" means waste characteristics collected by a waste management system and converted into digital form.

[0808] The "server" is a central control unit that receives collected data and performs analysis and training of classification models.

[0809] A "classification model" is a model that uses machine learning algorithms to classify waste as recyclable or not.

[0810] "Collection route optimization" is a method for designing optimal waste collection routes based on classification results, thereby achieving efficient collection activities.

[0811] "Sustainable Practice Information" is educational content and guidelines for recycling and waste reduction provided to users.

[0812] "User's information terminal" refers to a device such as a smartphone or tablet that is used to receive and display information in real time.

[0813] "Means for providing waste classification results in real time" refers to a function that displays the classification results on the user's information terminal the moment the waste is collected.

[0814] "Educational content to promote recycling" is content that provides users with information on the correct methods of waste disposal and recycling.

[0815] As an embodiment of this invention, the following system can be constructed. This system improves the efficiency of waste management in physical stores, promotes recycling, and supports sustainable practices. The system is mainly composed of a server, terminals, and users.

[0816] The server has the following functions: First, it collects data generated by sensors and transmits it via the Internet or local network. This digitally aggregates waste characteristic information (such as material type, color, and weight) on the server. Next, the server analyzes the received data and performs preprocessing (data cleaning, formatting, etc.). It uses the preprocessed data to train a specific classification model, which is then used to classify waste as recyclable or not. It designs optimal collection routes based on the classification results and delivers this information to users' devices in real time. It also provides users with educational content to promote recycling as needed.

[0817] The device has the function of receiving and displaying sustainable practice information and waste classification results sent from the server, allowing users to check important information on how to correctly classify waste and recycling in real time. Specifically, this applies to users' information devices such as smartphones and tablets.

[0818] Users receive information from the server via their devices and implement appropriate waste management practices, which promotes efficient waste disposal and recycling. Furthermore, educational content helps users deepen their own understanding of sustainable practices.

[0819] For example, supermarket staff can send data obtained from waste management devices (smart trash cans) to a server via a smartphone app. This data is analyzed on the server, and real-time information on whether the waste can be recycled is provided to staff. Based on this information, staff can properly classify the waste. The app also provides educational content about non-recyclable waste, promoting sustainable behavior.

[0820] Example prompt sentence:

[0821] Write a program that uses data obtained from waste management terminals by supermarket staff to determine in real time whether waste can be recycled and displays the results on the staff's smartphone app, "Eco Shopper." Data obtained from the sensors includes material type, color, and weight. Based on the results, provide staff with information about recyclability.

[0822] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0823] Step 1:

[0824] The server collects the data generated by the sensors. Specifically, the sensors in the waste management terminal collect characteristic information such as the type, color, and weight of the waste material, converts this data into a digital format, and transmits it to the server via the Internet or a local network.

[0825] Input: Waste characteristics (type of material, color, weight)

[0826] Output: Digital feature information data

[0827] Step 2:

[0828] The server preprocesses the received data. Specifically, the server performs data cleaning and formatting to generate preprocessed data.

[0829] Input: Digital feature data

[0830] Output: Preprocessed data

[0831] Step 3:

[0832] The server uses the preprocessed data to train a classification model. Specifically, it splits the preprocessed data into features and labels, and uses them as input to train a machine learning model (e.g., random forest).

[0833] Input: Preprocessed data

[0834] Output: A trained classification model

[0835] Step 4:

[0836] The server classifies newly collected waste data by analyzing the newly received data using a trained classification model and classifying the waste as recyclable or not in real time.

[0837] Input: Newly collected waste data

[0838] Output: Waste classification results

[0839] Step 5:

[0840] The server optimizes collection routes based on the classification results, specifically by designing optimal collection routes depending on the type and amount of waste, and preparing feedback for collection companies.

[0841] Input: Waste classification results

[0842] Output: Optimized collection route information

[0843] Step 6:

[0844] The server delivers sustainable practice information and classification results to the device, specifically, information on recyclable waste and educational content to promote recycling to the user's smartphone or tablet.

[0845] Input: Sustainable practice information and classification results

[0846] Output: Information displayed on the user's terminal

[0847] Step 7:

[0848] The user's device displays the received information. Specifically, the smartphone app displays the classification results and sustainable practice information in real time, and guides the user to appropriate waste disposal methods.

[0849] Input: Information from the server (classification results, sustainable practice information)

[0850] Output: Information displayed on the terminal screen

[0851] 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.

[0852] This invention combines a system that uses generative AI to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[0853] System Overview

[0854] The device's sensors collect waste characteristics (e.g., material type, color, weight, etc.) and simultaneously acquire the user's emotional data. For example, the device can recognize the user's emotions by analyzing their facial expressions and tone of voice as they sort the waste. The collected information is converted into digital form in real time and sent to a server via the Internet or a local network.

[0855] System Operation

[0856] The server stores the received data and pre-processes it. This includes cleaning and formatting the data. The server then splits the pre-processed data into features and labels to train a machine learning model. This model is a specific classification model that classifies waste as recyclable or not.

[0857] Furthermore, the server is equipped with an emotion engine that analyzes and classifies the user's emotion data. Based on the emotions recognized by the emotion engine, the server personalizes sustainable practice information and delivers it to the user via their device.

[0858] Specific examples

[0859] For example, smart trash cans installed in urban areas use various sensors to collect information on the characteristics of waste and user emotional data. The sensors detect anxiety and stress felt by users when throwing away trash, and this information is sent to a server. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[0860] At the same time, the emotion engine analyzes the user's emotions, and if it determines that the user is experiencing high levels of anxiety or stress, for example, easy-to-understand recycling instruction videos and positive messages are delivered to the user via the device.

[0861] The server also uses the classification results to design optimal collection routes and provides feedback to waste collection companies, allowing them to carry out collection work according to the optimal plan.For non-recyclable waste, the server notifies users of sustainable practices, promoting environmental education and raising awareness.

[0862] This will enable efficient waste management, promote recycling, and use emotional data to personalize sustainable practices.

[0863] The processing flow will be explained below.

[0864] Step 1:

[0865] The device uses various sensors to collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and simultaneously acquires the user's emotional data (e.g., facial expression, tone of voice, etc.). The collected information is converted into digital form in real time.

[0866] Step 2:

[0867] The terminal transmits the collected data (waste characteristic information and user emotion data) to a server via the Internet or a local network.

[0868] Step 3:

[0869] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[0870] Step 4:

[0871] The server divides the preprocessed data into features (characteristic information about the waste) and labels (whether it can be recycled or not), forming the dataset needed for training the machine learning model.

[0872] Step 5:

[0873] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[0874] Step 6:

[0875] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[0876] Step 7:

[0877] The server runs an emotion engine to analyze the user's emotion data, for example, by using facial recognition or voice analysis to estimate the user's emotion and record the results in a database.

[0878] Step 8:

[0879] The server personalizes sustainable practice information based on the user's emotions recognized by the emotion engine. For example, if the user is feeling anxious, it will recommend an easy-to-understand video on recycling.

[0880] Step 9:

[0881] The server sends the classification results and personalized sustainable practice information to the terminal, where the user can receive the information.

[0882] Step 10:

[0883] The device displays the information received from the server to the user, allowing them to learn how to properly classify waste, how to recycle, and other emotionally relevant practices.

[0884] Step 11:

[0885] The server then designs optimal garbage collection routes based on the classification results and provides feedback to garbage collection companies, enabling them to collect waste efficiently and along optimal routes.

[0886] Step 12:

[0887] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[0888] Example 2

[0889] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0890] Conventional waste management systems not only have difficulty collecting information on waste characteristics and classifying them appropriately, but also lack the ability to provide sustainable practice information that takes users' emotions into account. This makes it difficult to maintain users' interest in recycling and manage waste efficiently. Furthermore, there is a lack of means to provide appropriate support to users to alleviate the anxiety and stress they may feel during the waste sorting process.

[0891] 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.

[0892] In this invention, the server includes means for collecting waste characteristic information, means for collecting user emotion data, means for transmitting the collected data to the server, means for learning a specific classification model using data generated on the server, means for classifying waste using the classification model, means for analyzing the emotion data and recognizing the user's emotion, means for generating personalized sustainable practice information based on the classification results and the emotion analysis results, means for transmitting the generated sustainable practice information to the terminal, and means for delivering the sustainable practice information to the terminal, thereby enabling efficient waste classification and management and personalization of sustainable practice information based on the user's emotion.

[0893] "Waste characteristic information" refers to the physical and chemical attributes of waste materials, such as type, color, and weight.

[0894] "User emotion data" is data related to emotions obtained by analyzing the user's facial expressions, tone of voice, body movements, etc. when sorting waste.

[0895] "Server" refers to the central computer system for storing, analyzing, and processing collected data.

[0896] "Specific classification model" refers to a machine learning algorithm for classifying waste as recyclable or not.

[0897] "Emotion engine" refers to an algorithm for analyzing a user's emotion data and recognizing and classifying the emotion.

[0898] "Personalized sustainable practice information" refers to information and guidance on sustainable waste management and recycling that is individually optimized based on the user's emotional data.

[0899] "Terminal" refers to a device for collecting waste characteristic information and user emotion data, and for delivering sustainable practice information from a server to users.

[0900] "Emotion analysis results" refers to the results of the user's emotion data analyzed by the emotion engine, and refers to information indicating what emotions the user is feeling.

[0901] This invention combines a system that uses generative AI models to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[0902] The sensors on the device collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and also use a camera and microphone to simultaneously acquire the user's emotional data (facial expression, tone of voice, etc.). Specific examples of sensors include image sensors and weight sensors, and it is desirable for the camera to have high resolution. The collected information is converted into digital format in real time and sent to a server via the Internet or local network. The server requires a high-performance processor and large amount of memory, and it is recommended to use a database system such as MongoDB or MySQL.

[0903] The server stores the received data and performs preprocessing. Preprocessing includes data cleaning (removing noise and missing values) and formatting. Python scripts and the Pandas library can be used for data preprocessing. The server then splits the preprocessed data into features and labels and trains a machine learning model. The server uses machine learning libraries such as TensorFlow and PyTorch to build and train a classification model.

[0904] The machine learning model classifies waste as recyclable or not. This classification result contributes to more efficient waste management. In addition, the server is equipped with an emotion engine that analyzes and classifies users' emotional data. The emotion engine uses TensorFlow to analyze the user's facial expression data and recognizes emotions such as "anxiety" and "stress." Based on these emotions, the engine personalizes sustainable practice information and sends the generated information to the device. For example, if the emotion engine recognizes high levels of "anxiety" or "stress," it will provide the user with easy-to-understand recycling instruction videos and positive messages.

[0905] The device receives personalized sustainable practice information sent from the server and delivers it to the user via a display screen and audio output device, helping users deepen their understanding of recycling and providing information that helps reduce stress.

[0906] Examples:

[0907] Smart trash cans installed in urban areas use various sensors (image sensors, weight sensors, etc.) to collect information on the characteristics of waste. When a user throws away a plastic bottle, their facial expressions and tone of voice are collected by a camera and microphone, and this information is sent to a server. The server's emotion engine analyzes the user's emotional data, and if it determines that the user is experiencing high levels of "anxiety" or "stress," it will provide the user with recycling instruction videos and positive messages via their device.

[0908] Example prompt sentence:

[0909] "Analyze facial and vocal data as users sort waste, and suggest guidance and messages to provide if they feel anxious or stressed."

[0910] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0911] Step 1:

[0912] The terminal collects waste characteristic information and user emotion data.

[0913] Specific operation: The device is equipped with an image sensor, weight sensor, camera, and microphone to collect characteristic information such as the type, color, and weight of the waste material. The camera and microphone also collect the user's facial expression and tone of voice to obtain emotional data.

[0914] Input: waste (bottles, paper, etc.), user's face, voice

[0915] Output: Waste characteristic information (material, color, weight, etc.), user emotion data (facial expression, tone of voice)

[0916] Step 2:

[0917] The terminal transmits the collected data to the server.

[0918] What it does: Converts collected data into digital form in real time and sends it to a server over the Internet or local network using HTTP or HTTPS protocols.

[0919] Input: waste characteristics, user emotion data

[0920] Output: Raw data sent to the server

[0921] Step 3:

[0922] The server stores the received data and performs preprocessing.

[0923] Specific operations: The received data is stored in a database (e.g., MongoDB, MySQL), and the data is cleaned (removed noise and missing values) and formatted using Python scripts and the Pandas library.

[0924] Input: Raw data (waste characteristics, user emotion data)

[0925] Output: Preprocessed data

[0926] Step 4:

[0927] The server uses the preprocessed data to train a machine learning model.

[0928] Specific operation: Split the preprocessed data into features and labels, and train a neural network model using TensorFlow or PyTorch.

[0929] Input: Preprocessed data (features, labels)

[0930] Output: A trained classification model

[0931] Step 5:

[0932] The server classifies the waste as recyclable or not.

[0933] How it works: Using the trained classification model, newly received waste data is classified as recyclable or not in real time.

[0934] Input: New waste data

[0935] Output: Waste classification results (recyclable or not)

[0936] Step 6:

[0937] The server analyzes the user's emotional data using an emotion engine and generates personalized sustainable practice information.

[0938] Specific operation: Analyzes user emotion data using TensorFlow and recognizes emotions such as "anxiety" and "stress." Based on the recognition results, generates personalized sustainable practice information using Jinja2 templates.

[0939] Input: User emotion data

[0940] Output: Personalized sustainable practice information

[0941] Step 7:

[0942] The server sends the generated information to the terminal, which then distributes it to the user.

[0943] Specific operation: The generated sustainable practice information is sent to the terminal, which receives it and delivers it to the user using a display screen or audio output device.

[0944] Input: Personalized sustainable practice information

[0945] Output: Providing information to users (e.g., recycling instruction videos, positive messages)

[0946] This details how each processing step collects, transmits, analyzes, generates, and distributes data.

[0947] (Application example 2)

[0948] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0949] Conventional waste management systems lack the means to streamline waste sorting and collection, and are unable to provide sustainable practice information that takes into account the emotions of users. Especially in the case of waste disposal in stores, employees often feel stressed and anxious, which can lead to delays in proper waste sorting and recycling.

[0950] 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.

[0951] In this invention, the server includes means for collecting generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for collecting user emotion data, and means for analyzing the emotion data to personalize the sustainable practice information, thereby enabling efficient waste classification and optimization of collection routes, and further enabling the provision of personalized sustainable practice information based on the user's emotions.

[0952] "Generated data" refers to all digital information collected from devices, including emotional data and waste information.

[0953] The "collection means" refers to devices such as sensors and cameras for acquiring waste characteristic information and user emotional data.

[0954] The "means for transmitting to the server" is a communication module or protocol for transferring collected data to a server on the cloud via the Internet or a local network.

[0955] A "specific classification model" is a model trained using a machine learning algorithm to classify waste as recyclable or non-recyclable.

[0956] The "classification means" is a component that inputs collected data into a specific classification model and processes it to classify waste as recyclable or not.

[0957] The "means for optimizing collection routes" refers to algorithms or systems that design optimal collection routes for waste collection companies based on the classification results and enable efficient waste collection.

[0958] "Sustainable practice information" is information that helps users take sustainable actions, such as recycling methods and knowledge about environmental protection.

[0959] The "means of distribution" refers to a module that displays information directly on the terminal, or a communication system that sends notifications to the user's smartphone.

[0960] "Emotion data" is digital information that indicates the user's emotional state, obtained from the user's facial expression, tone of voice, etc.

[0961] The "analysis means" is a software module that recognizes the user's emotions based on the collected emotional data and determines actions based on the results.

[0962] A "personalization means" is an algorithm or system that uses acquired emotional data from a user to provide sustainable practice information optimized for that user.

[0963] This invention is a system that improves the efficiency of waste management in stores and provides sustainable practice information based on user sentiment. The system operates as follows.

[0964] First, the device is equipped with a camera and various sensors to collect waste characteristic information and the user's emotional data. Waste characteristic information includes the type of material, color, and weight, while the user's emotional data includes facial expressions and tone of voice. Each piece of data is digitized in real time and sent to a cloud server via the Internet or a local network.

[0965] The server stores the received data and first cleans and formats it. Then, it uses the data generated on the server to train a specific classification model. This classification model is trained using a machine learning algorithm (using TensorFlow as an example) to classify waste as recyclable or not.

[0966] The server is also equipped with an emotion engine that analyzes the user's emotion data. Based on the results obtained from the emotion engine, personalized sustainable practice information (e.g., recycling instructions and positive messages to reduce stress) is created in real time and delivered to the device. Examples of this sustainable practice information include easy-to-understand recycling instruction videos and positive messages.

[0967] As a concrete example, when an employee scans waste at a store's waste disposal station with their smartphone, the user's (employee's) facial expression is captured by the camera. If the employee shows any signs of anxiety or stress, that data is collected. The collected data is sent to a server and analyzed by an emotion engine. An easy-to-understand video on recycling methods is then displayed on the device to the employee.

[0968] This will enable users to properly separate waste without stress, and the system as a whole will promote efficient waste management and recycling.

[0969] Examples of prompts include:

[0970] "Generate a scenario where a user scans a waste item. The user feels anxious. Suggest a flowchart to provide appropriate support to this user."

[0971] This invention can significantly reduce a store's environmental impact through efficient waste sorting, optimised collection routes and the provision of personalized sustainable practice information based on user sentiment.

[0972] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0973] Step 1:

[0974] The device uses sensors and cameras to collect characteristic information about the waste (such as type of material, color, and weight) and the user's emotional data (such as facial expressions and tone of voice).

[0975] Input: Waste characteristics, user emotion data

[0976] Data processing: Converting collected data into a digital format

[0977] Output: Digitized waste information and emotion data

[0978] Step 2:

[0979] The terminal transmits the digitized waste information and emotion data to a server via the Internet or a local network.

[0980] Input: Digitized waste information and emotion data

[0981] Data processing: Packetizing data according to network protocols

[0982] Output: Data packet sent to the server

[0983] Step 3:

[0984] The server stores the received data in a database and cleans and formats it.

[0985] Input: Data packet (waste information and emotion data)

[0986] Data processing: Data cleaning (insertion of missing values, removal of invalid data) and formatting

[0987] Output: Preprocessed data

[0988] Step 4:

[0989] The server splits the preprocessed data into features and labels and trains a machine learning model.

[0990] Input: Preprocessed data

[0991] Data Processing: Feature Engineering and Labeling for Machine Learning

[0992] Output: A trained classification model

[0993] Step 5:

[0994] The server uses a trained classification model to classify waste as recyclable or not.

[0995] Input: Real-time collected waste data

[0996] Data processing: Analyzing and classifying data using classification models

[0997] Output: Classification result as recyclable or not

[0998] Step 6:

[0999] The server designs the optimal collection route based on the classification results and provides feedback to the contractor.

[1000] Input: Classification results

[1001] Data processing: Applying collection route optimization algorithms

[1002] Output: Optimized collection route

[1003] Step 7:

[1004] The server analyzes the collected emotion data and generates sustainable practice information based on the user's emotions.

[1005] Input: Emotion data

[1006] Data processing: Analysis and information generation using emotion engines

[1007] Output: Personalized sustainable practice information

[1008] Step 8:

[1009] The server distributes the generated sustainable practice information to the terminal.

[1010] Input: Personalized sustainable practice information

[1011] Data processing: Sending information to the user's device

[1012] Output: Sustainable practice information displayed on a terminal

[1013] This series of processing steps enables efficient classification and management of waste and provides information based on the user's emotions.

[1014] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1015] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1016] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1017] [Fourth embodiment]

[1018] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1019] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1020] 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).

[1021] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1022] 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.

[1023] 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).

[1024] 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. 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.

[1025] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1026] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1027] 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.

[1028] 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.

[1029] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1030] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1031] This invention is a system that aims to reduce waste and promote recycling by using generative AI. Specifically, terminals installed in each supermarket and community send collected waste data to a server in real time, and the server analyzes and optimizes this data to achieve efficient waste management.

[1032] System Overview

[1033] The sensors in the device collect characteristic information about the waste (e.g., material type, color, weight, etc.) and convert it into digital format as generated data. The collected data is then sent to a server via the Internet or a local network.

[1034] The server receives the data, stores it, and pre-processes it, which includes cleaning and formatting it. The server then splits the pre-processed data into features and labels to train a machine learning model, a specific classification model, that classifies waste as recyclable or not.

[1035] System Operation

[1036] After training the classification model, the server analyzes newly collected data in real time and classifies the waste. Based on the classification results, it designs optimal collection routes and delivers relevant sustainable practice information to users via their devices.

[1037] Specific examples

[1038] For example, smart trash cans installed around town use various sensors to collect information about the characteristics of waste. This information is sent to a server via the user's smartphone or dedicated device. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[1039] The server then uses the classification results to plan optimal collection routes and provide feedback to waste collection companies. It also delivers educational content about non-recyclable waste to users' devices and provides information on sustainable practices, helping users take appropriate recycling actions.

[1040] In this way, the system contributes to efficient waste management, promoting recycling, and realizing a sustainable society through education.

[1041] The processing flow will be explained below.

[1042] Step 1:

[1043] The terminal uses various sensors to collect information about the characteristics of the waste (such as the type of material, color, weight, etc.), which is then converted into a digital format in real time.

[1044] Step 2:

[1045] The terminal transmits the collected data to a server via the Internet or a local network. The transmitted data includes information about the characteristics of the waste.

[1046] Step 3:

[1047] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[1048] Step 4:

[1049] The server separates the preprocessed data into features (e.g., material type, color, weight) and labels (e.g., recyclable / non-recyclable), forming the dataset needed to train the machine learning model.

[1050] Step 5:

[1051] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[1052] Step 6:

[1053] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[1054] Step 7:

[1055] The server then designs the optimal garbage collection route based on the classification results, enabling efficient collection and reducing costs and environmental impact.

[1056] Step 8:

[1057] The server provides feedback to waste collection companies on the designed collection routes and recycling information, allowing them to carry out collection according to the optimal plan.

[1058] Step 9:

[1059] The server generates educational content about non-recyclable waste and delivers sustainable practice information to users via their devices, allowing them to learn about correct recycling methods and environmentally friendly behavior.

[1060] Step 10:

[1061] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[1062] Example 1

[1063] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1064] To efficiently manage waste and promote recycling, it is necessary to classify waste, optimize collection routes, and provide information on sustainable practices. However, conventional systems lack the ability to adequately control the quality of collected data, train classification models, and analyze data in real time, making appropriate waste management difficult. Furthermore, they are also inadequate in providing users with the information they need on sustainable practices, limiting their contribution to a sustainable society.

[1065] 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.

[1066] In this invention, the server includes means for the terminal to collect waste characteristic information using sensors, means for converting the collected data into digital form and sending it to the server, means for storing, cleaning, and formatting the received data on the server, means for dividing the preprocessed data into features and labels and training a machine learning model, means for analyzing the new data in real time and classifying the waste as recyclable or not, means for designing optimal waste collection routes based on the classification results, and means for delivering sustainable practice information to the user through the terminal, thereby enabling efficient waste management and promotion of recycling, as well as providing users with prompt and appropriate information.

[1067] The "terminal" is a device equipped with a sensor to collect characteristic information about waste, convert the collected data into digital format, and send it to a server.

[1068] "Server" is a computer system for receiving, storing, and pre-processing collected data and for training and running machine learning models.

[1069] A "sensor" is a measuring device installed in a terminal to detect and collect characteristic information such as the material, color, and weight of waste.

[1070] "Characteristic information" is data that indicates the material, color, weight, etc. of the waste, and is collected by sensors.

[1071] "Digital format" refers to data that has been converted from analog data so that it can be processed electronically.

[1072] "Data cleaning" is the process of removing noise and missing values ​​from collected data, leaving only the necessary information.

[1073] "Formatting" refers to the conversion of data of different formats or structures into a unified format or structure.

[1074] A "feature" is a value that indicates a specific attribute or characteristic of the data used to train a machine learning model.

[1075] A "label" is a value that indicates the output or classification result corresponding to a feature in a machine learning model.

[1076] A "machine learning model" is an algorithm that learns patterns and relationships from given data and makes predictions and classifications for new data.

[1077] "Classification" is the process of grouping received data based on predefined categories or labels.

[1078] A "waste collection route" is a route used by a waste collection vehicle to efficiently collect waste.

[1079] "Sustainable practice information" is information on actions aimed at realizing a sustainable society, such as reducing environmental impact and promoting recycling.

[1080] "User" means any person or organization that uses this system and receives guidance on how to properly dispose of and recycle waste.

[1081] This invention is a system that aims to reduce waste and promote recycling by utilizing generative AI models. In this system, terminals installed in each supermarket and community send waste data in real time to a server, which then analyzes and optimizes this data to achieve efficient waste management.

[1082] System configuration

[1083] 1. The terminal is equipped with multiple sensors. These sensors collect characteristic information about the waste (such as material type, color, and weight) and convert it into digital form. Specific examples of sensors include material detection sensors, color-identification cameras, and weight sensors.

[1084] 2. The collected data is sent to a server via the Internet or a local network. The device transmits this data in real time, ensuring that it arrives at the server without delay.

[1085] 3. The server first stores the received data, then cleans and formats it. The software used includes database management systems and data cleaning tools, such as PostgreSQL and pandas (a Python library).

[1086] 4. The server splits the preprocessed data into features and labels to train a machine learning model. This model is implemented using the Python machine learning library scikit-learn. In particular, it uses a random forest classifier to classify whether the waste is recyclable or not.

[1087] 5. The server uses the trained model to analyze newly collected data in real time and classify the waste. Based on this classification, it designs optimal waste collection routes. Collection route planning is done using Geographic Information System (GIS) software.

[1088] 6. The server sends the sorting results and collection route information to the device and also provides the user with sustainable practice information, such as educational content about non-recyclable waste and the location of specific recycling facilities.

[1089] Specific examples

[1090] For example, a smart trash can installed in a city uses various sensors to collect information about the characteristics of waste. When a user throws a plastic bottle into the trash can, a material detection sensor detects the plastic, a color-identification camera identifies the color blue, and a weight sensor measures 200g. This data is converted into digital format and sent to a server in real time.

[1091] The server stores the received data, cleans and formats it using pandas, then splits the data ("plastic, blue, 200g") into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[1092] The server analyzes newly received data in real time and classifies it as "recyclable." Based on this classification, the GIS software designs optimal garbage collection routes and provides them to collection companies. At the same time, the user receives a notification via their device saying, "This waste is recyclable. Please separate it appropriately."

[1093] Below is an example of a prompt sentence:

[1094] "What is the recycling status of the waste I put in yesterday?"

[1095] In this way, the present invention can efficiently manage waste and promote recycling, and further contributes to the realization of a sustainable society through user education.

[1096] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1097] Step 1:

[1098] The terminal uses sensors to collect information about the characteristics of the waste. The terminal is equipped with a material detection sensor, a color recognition camera, a weight sensor, etc. The data obtained by these sensors is information about the material, color, and weight of the waste.

[1099] Input: Waste

[1100] Specific operation: The sensor detects plastic waste, the color-detection camera identifies the color blue, and the weight sensor measures 200g.

[1101] Output: Feature information: "Plastic, Blue, 200g"

[1102] Step 2:

[1103] The device converts the collected data into a digital format and transmits it to a server via the internet or a local network. The device transmits this data in real time, so it arrives at the server without delay.

[1104] Input: Feature information "Plastic, Blue, 200g"

[1105] Specific operation: The device converts the analog data into digital format and sends the data to the specified address on the server.

[1106] Output: Digital characteristic information (data sent to the server)

[1107] Step 3:

[1108] The server stores, cleans, and formats the data it receives using a database management system (e.g., PostgreSQL) and a data cleaning tool (e.g., pandas).

[1109] Input: Digital feature information

[1110] Specific operation: The server stores the received data in a database and performs cleaning such as filling in missing values ​​and removing outliers. After that, the data format is unified.

[1111] Output: Preprocessed feature information

[1112] Step 4:

[1113] The server splits the preprocessed data into features and labels, and trains a machine learning model using scikit-learn's random forest classifier.

[1114] Input: Preprocessed feature information

[1115] Specific operation: The server splits the data "plastic, blue, 200g" into "features (plastic, blue, 200g)" and "label (recyclable)" and supplies them to the random forest classifier as training data.

[1116] Output: A trained machine learning model

[1117] Step 5:

[1118] The server analyzes newly received data in real time and uses machine learning models to classify the waste.

[1119] Input: Newly received data (e.g. "glass, blue, 300g")

[1120] What it does: Use the trained model to analyze new data: "glass, blue, 300g" and classify it as "recyclable."

[1121] Output: Classification result (e.g. "Recyclable")

[1122] Step 6:

[1123] The server uses Geographic Information System (GIS) software to design optimal waste collection routes based on the classification results and provide feedback to collection companies.

[1124] Input: Classification results and permitted information (e.g. waste type and collection amount)

[1125] How it works: The server analyzes all waste data, calculates efficient collection routes based on the amount and type of waste, and provides collection schedules and efficient route guidance to collection companies.

[1126] Output: Optimized collection route information

[1127] Step 7:

[1128] The server delivers sustainable practice information to users through their devices, including, for example, educational content about non-recyclable waste and the locations of specific recycling facilities.

[1129] Input: Classification results and sustainable practice information

[1130] What it does: Generates an educational video on "How to dispose of non-recyclable plastics" and sends a notification to the user's smartphone.

[1131] Output: User notification and educational content

[1132] The above is the processing flow of this system.

[1133] (Application example 1)

[1134] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1135] In modern society, the increase in waste and the need for recycling have created a demand for efficient waste management. However, proper sorting and promotion of recycling are often insufficient. Brick-and-mortar stores in particular generate large amounts of waste, and a system is needed to efficiently manage this waste and promote sustainable practices.

[1136] 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.

[1137] In this invention, the server includes means for collecting the generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for providing the waste classification results in real time through the user's information terminal, and means for delivering educational content for promoting recycling to the user's information terminal, thereby enabling efficient waste management, promotion of recycling, and realization of a sustainable society.

[1138] "Generated data" means waste characteristics collected by a waste management system and converted into digital form.

[1139] The "server" is a central control unit that receives collected data and performs analysis and training of classification models.

[1140] A "classification model" is a model that uses machine learning algorithms to classify waste as recyclable or not.

[1141] "Collection route optimization" is a method for designing optimal waste collection routes based on classification results, thereby achieving efficient collection activities.

[1142] "Sustainable Practice Information" is educational content and guidelines for recycling and waste reduction provided to users.

[1143] "User's information terminal" refers to a device such as a smartphone or tablet that is used to receive and display information in real time.

[1144] "Means for providing waste classification results in real time" refers to a function that displays the classification results on the user's information terminal the moment the waste is collected.

[1145] "Educational content to promote recycling" is content that provides users with information on the correct methods of waste disposal and recycling.

[1146] As an embodiment of this invention, the following system can be constructed. This system improves the efficiency of waste management in physical stores, promotes recycling, and supports sustainable practices. The system is mainly composed of a server, terminals, and users.

[1147] The server has the following functions: First, it collects data generated by sensors and transmits it via the Internet or local network. This digitally aggregates waste characteristic information (such as material type, color, and weight) on the server. Next, the server analyzes the received data and performs preprocessing (data cleaning, formatting, etc.). It uses the preprocessed data to train a specific classification model, which is then used to classify waste as recyclable or not. It designs optimal collection routes based on the classification results and delivers this information to users' devices in real time. It also provides users with educational content to promote recycling as needed.

[1148] The device has the function of receiving and displaying sustainable practice information and waste classification results sent from the server, allowing users to check important information on how to correctly classify waste and recycling in real time. Specifically, this applies to users' information devices such as smartphones and tablets.

[1149] Users receive information from the server via their devices and implement appropriate waste management practices, which promotes efficient waste disposal and recycling. Furthermore, educational content helps users deepen their own understanding of sustainable practices.

[1150] For example, supermarket staff can send data obtained from waste management devices (smart trash cans) to a server via a smartphone app. This data is analyzed on the server, and real-time information on whether the waste can be recycled is provided to staff. Based on this information, staff can properly classify the waste. The app also provides educational content about non-recyclable waste, promoting sustainable behavior.

[1151] Example prompt sentence:

[1152] Write a program that uses data obtained from waste management terminals by supermarket staff to determine in real time whether waste can be recycled and displays the results on the staff's smartphone app, "Eco Shopper." Data obtained from the sensors includes material type, color, and weight. Based on the results, provide staff with information about recyclability.

[1153] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1154] Step 1:

[1155] The server collects the data generated by the sensors. Specifically, the sensors in the waste management terminal collect characteristic information such as the type, color, and weight of the waste material, converts this data into a digital format, and transmits it to the server via the Internet or a local network.

[1156] Input: Waste characteristics (type of material, color, weight)

[1157] Output: Digital feature information data

[1158] Step 2:

[1159] The server preprocesses the received data. Specifically, the server performs data cleaning and formatting to generate preprocessed data.

[1160] Input: Digital feature data

[1161] Output: Preprocessed data

[1162] Step 3:

[1163] The server uses the preprocessed data to train a classification model. Specifically, it splits the preprocessed data into features and labels, and uses them as input to train a machine learning model (e.g., random forest).

[1164] Input: Preprocessed data

[1165] Output: A trained classification model

[1166] Step 4:

[1167] The server classifies newly collected waste data by analyzing the newly received data using a trained classification model and classifying the waste as recyclable or not in real time.

[1168] Input: Newly collected waste data

[1169] Output: Waste classification results

[1170] Step 5:

[1171] The server optimizes collection routes based on the classification results, specifically by designing optimal collection routes depending on the type and amount of waste, and preparing feedback for collection companies.

[1172] Input: Waste classification results

[1173] Output: Optimized collection route information

[1174] Step 6:

[1175] The server delivers sustainable practice information and classification results to the device, specifically, information on recyclable waste and educational content to promote recycling to the user's smartphone or tablet.

[1176] Input: Sustainable practice information and classification results

[1177] Output: Information displayed on the user's terminal

[1178] Step 7:

[1179] The user's device displays the received information. Specifically, the smartphone app displays the classification results and sustainable practice information in real time, and guides the user to appropriate waste disposal methods.

[1180] Input: Information from the server (classification results, sustainable practice information)

[1181] Output: Information displayed on the terminal screen

[1182] 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.

[1183] This invention combines a system that uses generative AI to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[1184] System Overview

[1185] The device's sensors collect waste characteristics (e.g., material type, color, weight, etc.) and simultaneously acquire the user's emotional data. For example, the device can recognize the user's emotions by analyzing their facial expressions and tone of voice as they sort the waste. The collected information is converted into digital form in real time and sent to a server via the Internet or a local network.

[1186] System Operation

[1187] The server stores the received data and pre-processes it. This includes cleaning and formatting the data. The server then splits the pre-processed data into features and labels to train a machine learning model. This model is a specific classification model that classifies waste as recyclable or not.

[1188] Furthermore, the server is equipped with an emotion engine that analyzes and classifies the user's emotion data. Based on the emotions recognized by the emotion engine, the server personalizes sustainable practice information and delivers it to the user via their device.

[1189] Specific examples

[1190] For example, smart trash cans installed in urban areas use various sensors to collect information on the characteristics of waste and user emotional data. The sensors detect anxiety and stress felt by users when throwing away trash, and this information is sent to a server. The server analyzes the received data and uses machine learning models to classify the waste as recyclable or not.

[1191] At the same time, the emotion engine analyzes the user's emotions, and if it determines that the user is experiencing high levels of anxiety or stress, for example, easy-to-understand recycling instruction videos and positive messages are delivered to the user via the device.

[1192] The server also uses the classification results to design optimal collection routes and provides feedback to waste collection companies, allowing them to carry out collection work according to the optimal plan.For non-recyclable waste, the server notifies users of sustainable practices, promoting environmental education and raising awareness.

[1193] This will enable efficient waste management, promote recycling, and use emotional data to personalize sustainable practices.

[1194] The processing flow will be explained below.

[1195] Step 1:

[1196] The device uses various sensors to collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and simultaneously acquires the user's emotional data (e.g., facial expression, tone of voice, etc.). The collected information is converted into digital form in real time.

[1197] Step 2:

[1198] The terminal transmits the collected data (waste characteristic information and user emotion data) to a server via the Internet or a local network.

[1199] Step 3:

[1200] The server stores and pre-processes the received data, which includes cleaning and formatting the data, for example, imputing missing values ​​and removing outliers.

[1201] Step 4:

[1202] The server divides the preprocessed data into features (characteristic information about the waste) and labels (whether it can be recycled or not), forming the dataset needed for training the machine learning model.

[1203] Step 5:

[1204] The server uses the features and labels to train a machine learning model (e.g., RandomForestClassifier), splits the data into training and test data, and performs model training and performance evaluation.

[1205] Step 6:

[1206] The server analyzes newly collected data in real time based on the learned classification model and classifies waste as recyclable or not.

[1207] Step 7:

[1208] The server runs an emotion engine to analyze the user's emotion data, for example, by using facial recognition or voice analysis to estimate the user's emotion and record the results in a database.

[1209] Step 8:

[1210] The server personalizes sustainable practice information based on the user's emotions recognized by the emotion engine. For example, if the user is feeling anxious, it will recommend an easy-to-understand video on recycling.

[1211] Step 9:

[1212] The server sends the classification results and personalized sustainable practice information to the terminal, where the user can receive the information.

[1213] Step 10:

[1214] The device displays the information received from the server to the user, allowing them to learn how to properly classify waste, how to recycle, and other emotionally relevant practices.

[1215] Step 11:

[1216] The server then designs optimal garbage collection routes based on the classification results and provides feedback to garbage collection companies, enabling them to collect waste efficiently and along optimal routes.

[1217] Step 12:

[1218] Users receive notifications from their devices and practice appropriate recycling behavior, which improves waste management in cities and helps create a sustainable society.

[1219] Example 2

[1220] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1221] Conventional waste management systems not only have difficulty collecting information on waste characteristics and classifying them appropriately, but also lack the ability to provide sustainable practice information that takes users' emotions into account. This makes it difficult to maintain users' interest in recycling and manage waste efficiently. Furthermore, there is a lack of means to provide appropriate support to users to alleviate the anxiety and stress they may feel during the waste sorting process.

[1222] 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.

[1223] In this invention, the server includes means for collecting waste characteristic information, means for collecting user emotion data, means for transmitting the collected data to the server, means for learning a specific classification model using data generated on the server, means for classifying waste using the classification model, means for analyzing the emotion data and recognizing the user's emotion, means for generating personalized sustainable practice information based on the classification results and the emotion analysis results, means for transmitting the generated sustainable practice information to the terminal, and means for delivering the sustainable practice information to the terminal, thereby enabling efficient waste classification and management and personalization of sustainable practice information based on the user's emotion.

[1224] "Waste characteristic information" refers to the physical and chemical attributes of waste materials, such as type, color, and weight.

[1225] "User emotion data" is data related to emotions obtained by analyzing the user's facial expressions, tone of voice, body movements, etc. when sorting waste.

[1226] "Server" refers to the central computer system for storing, analyzing, and processing collected data.

[1227] "Specific classification model" refers to a machine learning algorithm for classifying waste as recyclable or not.

[1228] "Emotion engine" refers to an algorithm for analyzing a user's emotion data and recognizing and classifying the emotion.

[1229] "Personalized sustainable practice information" refers to information and guidance on sustainable waste management and recycling that is individually optimized based on the user's emotional data.

[1230] "Terminal" refers to a device for collecting waste characteristic information and user emotion data, and for delivering sustainable practice information from a server to users.

[1231] "Emotion analysis results" refers to the results of the user's emotion data analyzed by the emotion engine, and refers to information indicating what emotions the user is feeling.

[1232] This invention combines a system that uses generative AI models to reduce waste and promote recycling with an emotion engine that recognizes user emotions. Specifically, terminals installed in each community collect waste data and user emotion data, and a server analyzes and optimizes the data to achieve efficient waste management and promote sustainable practices.

[1233] The sensors on the device collect characteristic information about the waste (e.g., type of material, color, weight, etc.), and also use a camera and microphone to simultaneously acquire the user's emotional data (facial expression, tone of voice, etc.). Specific examples of sensors include image sensors and weight sensors, and it is desirable for the camera to have high resolution. The collected information is converted into digital format in real time and sent to a server via the Internet or local network. The server requires a high-performance processor and large amount of memory, and it is recommended to use a database system such as MongoDB or MySQL.

[1234] The server stores the received data and performs preprocessing. Preprocessing includes data cleaning (removing noise and missing values) and formatting. Python scripts and the Pandas library can be used for data preprocessing. The server then splits the preprocessed data into features and labels and trains a machine learning model. The server uses machine learning libraries such as TensorFlow and PyTorch to build and train a classification model.

[1235] The machine learning model classifies waste as recyclable or not. This classification result contributes to more efficient waste management. In addition, the server is equipped with an emotion engine that analyzes and classifies users' emotional data. The emotion engine uses TensorFlow to analyze the user's facial expression data and recognizes emotions such as "anxiety" and "stress." Based on these emotions, the engine personalizes sustainable practice information and sends the generated information to the device. For example, if the emotion engine recognizes high levels of "anxiety" or "stress," it will provide the user with easy-to-understand recycling instruction videos and positive messages.

[1236] The device receives personalized sustainable practice information sent from the server and delivers it to the user via a display screen and audio output device, helping users deepen their understanding of recycling and providing information that helps reduce stress.

[1237] Examples:

[1238] Smart trash cans installed in urban areas use various sensors (image sensors, weight sensors, etc.) to collect information on the characteristics of waste. When a user throws away a plastic bottle, their facial expressions and tone of voice are collected by a camera and microphone, and this information is sent to a server. The server's emotion engine analyzes the user's emotional data, and if it determines that the user is experiencing high levels of "anxiety" or "stress," it will provide the user with recycling instruction videos and positive messages via their device.

[1239] Example prompt sentence:

[1240] "Analyze facial and vocal data as users sort waste, and suggest guidance and messages to provide if they feel anxious or stressed."

[1241] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1242] Step 1:

[1243] The terminal collects waste characteristic information and user emotion data.

[1244] Specific operation: The device is equipped with an image sensor, weight sensor, camera, and microphone to collect characteristic information such as the type, color, and weight of the waste material. The camera and microphone also collect the user's facial expression and tone of voice to obtain emotional data.

[1245] Input: waste (bottles, paper, etc.), user's face, voice

[1246] Output: Waste characteristic information (material, color, weight, etc.), user emotion data (facial expression, tone of voice)

[1247] Step 2:

[1248] The terminal transmits the collected data to the server.

[1249] What it does: Converts collected data into digital form in real time and sends it to a server over the Internet or local network using HTTP or HTTPS protocols.

[1250] Input: waste characteristics, user emotion data

[1251] Output: Raw data sent to the server

[1252] Step 3:

[1253] The server stores the received data and performs preprocessing.

[1254] Specific operations: The received data is stored in a database (e.g., MongoDB, MySQL), and the data is cleaned (removed noise and missing values) and formatted using Python scripts and the Pandas library.

[1255] Input: Raw data (waste characteristics, user emotion data)

[1256] Output: Preprocessed data

[1257] Step 4:

[1258] The server uses the preprocessed data to train a machine learning model.

[1259] Specific operation: Split the preprocessed data into features and labels, and train a neural network model using TensorFlow or PyTorch.

[1260] Input: Preprocessed data (features, labels)

[1261] Output: A trained classification model

[1262] Step 5:

[1263] The server classifies the waste as recyclable or not.

[1264] How it works: Using the trained classification model, newly received waste data is classified as recyclable or not in real time.

[1265] Input: New waste data

[1266] Output: Waste classification results (recyclable or not)

[1267] Step 6:

[1268] The server analyzes the user's emotional data using an emotion engine and generates personalized sustainable practice information.

[1269] Specific operation: Analyzes user emotion data using TensorFlow and recognizes emotions such as "anxiety" and "stress." Based on the recognition results, generates personalized sustainable practice information using Jinja2 templates.

[1270] Input: User emotion data

[1271] Output: Personalized sustainable practice information

[1272] Step 7:

[1273] The server sends the generated information to the terminal, which then distributes it to the user.

[1274] Specific operation: The generated sustainable practice information is sent to the terminal, which receives it and delivers it to the user using a display screen or audio output device.

[1275] Input: Personalized sustainable practice information

[1276] Output: Providing information to users (e.g., recycling instruction videos, positive messages)

[1277] This details how each processing step collects, transmits, analyzes, generates, and distributes data.

[1278] (Application example 2)

[1279] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1280] Conventional waste management systems lack the means to streamline waste sorting and collection, and are unable to provide sustainable practice information that takes into account the emotions of users. Especially in the case of waste disposal in stores, employees often feel stressed and anxious, which can lead to delays in proper waste sorting and recycling.

[1281] 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.

[1282] In this invention, the server includes means for collecting generated data, means for transmitting the collected data to the server, means for learning a specific classification model using the data generated on the server, means for classifying waste using the classification model, means for optimizing collection routes based on the classification results, means for delivering sustainable practice information to the terminal, means for collecting user emotion data, and means for analyzing the emotion data to personalize the sustainable practice information, thereby enabling efficient waste classification and optimization of collection routes, and further enabling the provision of personalized sustainable practice information based on the user's emotions.

[1283] "Generated data" refers to all digital information collected from devices, including emotional data and waste information.

[1284] The "collection means" refers to devices such as sensors and cameras for acquiring waste characteristic information and user emotional data.

[1285] The "means for transmitting to the server" is a communication module or protocol for transferring collected data to a server on the cloud via the Internet or a local network.

[1286] A "specific classification model" is a model trained using a machine learning algorithm to classify waste as recyclable or non-recyclable.

[1287] The "classification means" is a component that inputs collected data into a specific classification model and processes it to classify waste as recyclable or not.

[1288] The "means for optimizing collection routes" refers to algorithms or systems that design optimal collection routes for waste collection companies based on the classification results and enable efficient waste collection.

[1289] "Sustainable practice information" is information that helps users take sustainable actions, such as recycling methods and knowledge about environmental protection.

[1290] The "means of distribution" refers to a module that displays information directly on the terminal, or a communication system that sends notifications to the user's smartphone.

[1291] "Emotion data" is digital information that indicates the user's emotional state, obtained from the user's facial expression, tone of voice, etc.

[1292] The "analysis means" is a software module that recognizes the user's emotions based on the collected emotional data and determines actions based on the results.

[1293] A "personalization means" is an algorithm or system that uses acquired emotional data from a user to provide sustainable practice information optimized for that user.

[1294] This invention is a system that improves the efficiency of waste management in stores and provides sustainable practice information based on user sentiment. The system operates as follows.

[1295] First, the device is equipped with a camera and various sensors to collect waste characteristic information and the user's emotional data. Waste characteristic information includes the type of material, color, and weight, while the user's emotional data includes facial expressions and tone of voice. Each piece of data is digitized in real time and sent to a cloud server via the Internet or a local network.

[1296] The server stores the received data and first cleans and formats it. Then, it uses the data generated on the server to train a specific classification model. This classification model is trained using a machine learning algorithm (using TensorFlow as an example) to classify waste as recyclable or not.

[1297] The server is also equipped with an emotion engine that analyzes the user's emotion data. Based on the results obtained from the emotion engine, personalized sustainable practice information (e.g., recycling instructions and positive messages to reduce stress) is created in real time and delivered to the device. Examples of this sustainable practice information include easy-to-understand recycling instruction videos and positive messages.

[1298] As a concrete example, when an employee scans waste at a store's waste disposal station with their smartphone, the user's (employee's) facial expression is captured by the camera. If the employee shows any signs of anxiety or stress, that data is collected. The collected data is sent to a server and analyzed by an emotion engine. An easy-to-understand video on recycling methods is then displayed on the device to the employee.

[1299] This will enable users to properly separate waste without stress, and the system as a whole will promote efficient waste management and recycling.

[1300] Examples of prompts include:

[1301] "Generate a scenario where a user scans a waste item. The user feels anxious. Suggest a flowchart to provide appropriate support to this user."

[1302] This invention can significantly reduce a store's environmental impact through efficient waste sorting, optimised collection routes and the provision of personalized sustainable practice information based on user sentiment.

[1303] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1304] Step 1:

[1305] The device uses sensors and cameras to collect characteristic information about the waste (such as type of material, color, and weight) and the user's emotional data (such as facial expressions and tone of voice).

[1306] Input: Waste characteristics, user emotion data

[1307] Data processing: Converting collected data into a digital format

[1308] Output: Digitized waste information and emotion data

[1309] Step 2:

[1310] The terminal transmits the digitized waste information and emotion data to a server via the Internet or a local network.

[1311] Input: Digitized waste information and emotion data

[1312] Data processing: Packetizing data according to network protocols

[1313] Output: Data packet sent to the server

[1314] Step 3:

[1315] The server stores the received data in a database and cleans and formats it.

[1316] Input: Data packet (waste information and emotion data)

[1317] Data processing: Data cleaning (insertion of missing values, removal of invalid data) and formatting

[1318] Output: Preprocessed data

[1319] Step 4:

[1320] The server splits the preprocessed data into features and labels and trains a machine learning model.

[1321] Input: Preprocessed data

[1322] Data Processing: Feature Engineering and Labeling for Machine Learning

[1323] Output: A trained classification model

[1324] Step 5:

[1325] The server uses a trained classification model to classify waste as recyclable or not.

[1326] Input: Real-time collected waste data

[1327] Data processing: Analyzing and classifying data using classification models

[1328] Output: Classification result as recyclable or not

[1329] Step 6:

[1330] The server designs the optimal collection route based on the classification results and provides feedback to the contractor.

[1331] Input: Classification results

[1332] Data processing: Applying collection route optimization algorithms

[1333] Output: Optimized collection route

[1334] Step 7:

[1335] The server analyzes the collected emotion data and generates sustainable practice information based on the user's emotions.

[1336] Input: Emotion data

[1337] Data processing: Analysis and information generation using emotion engines

[1338] Output: Personalized sustainable practice information

[1339] Step 8:

[1340] The server distributes the generated sustainable practice information to the terminal.

[1341] Input: Personalized sustainable practice information

[1342] Data processing: Sending information to the user's device

[1343] Output: Sustainable practice information displayed on a terminal

[1344] This series of processing steps enables efficient classification and management of waste and provides information based on the user's emotions.

[1345] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1346] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1347] 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 robot 414.

[1348] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1349] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1350] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1351] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1352] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, motorcycles, and other devices, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1353] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1354] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1355] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1356] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1357] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1358] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1359] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1360] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1361] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1362] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1363] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1364] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1365] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1366] The following is further disclosed regarding the above embodiment.

[1367] (Claim 1)

[1368] a means for collecting the generated data;

[1369] means for transmitting the collected data to a server;

[1370] means for training a particular classification model using data generated on the server;

[1371] a means for classifying waste using a classification model;

[1372] a means for optimizing collection routes based on the classification results;

[1373] A system including means for delivering sustainable practice information to a terminal.

[1374] (Claim 2)

[1375] 10. The system of claim 1, further comprising means for dividing the collected data into features and labels.

[1376] (Claim 3)

[1377] 10. The system of claim 1, further comprising means for evaluating performance of the classification model.

[1378] "Example 1"

[1379] (Claim 1)

[1380] A means for the terminal to collect characteristic information of the waste using a sensor;

[1381] means for converting the collected data into a digital format and transmitting it to a server;

[1382] means for storing, cleaning and formatting the received data on a server;

[1383] A means to split the preprocessed data into features and labels and train a machine learning model.

[1384] A means to analyze new data in real time and classify waste as recyclable or not;

[1385] A means for designing optimal waste collection routes based on the classification results;

[1386] A system including means for delivering sustainable practice information to a user through a terminal.

[1387] (Claim 2)

[1388] 10. The system of claim 1, further comprising means for storing, cleaning and formatting the received data.

[1389] (Claim 3)

[1390] 10. The system of claim 1, further comprising means for evaluating performance of the classification model and providing feedback.

[1391] "Application Example 1"

[1392] (Claim 1)

[1393] a means for collecting the generated data;

[1394] means for transmitting the collected data to a server;

[1395] means for training a particular classification model using data generated on the server;

[1396] a means for classifying waste using a classification model;

[1397] a means for optimizing collection routes based on the classification results;

[1398] a means for delivering sustainable practice information to the device;

[1399] A means of providing real-time waste classification results through the user's information terminal;

[1400] A means to deliver educational content to users' information terminals to promote recycling

[1401] Including system.

[1402] (Claim 2)

[1403] 10. The system of claim 1, further comprising means for dividing the collected data into features and labels.

[1404] (Claim 3)

[1405] 10. The system of claim 1, further comprising means for evaluating performance of the classification model.

[1406] "Example 2: Combining Emotion Engines"

[1407] (Claim 1)

[1408] a means of collecting waste characterization information;

[1409] means for collecting user emotion data;

[1410] means for transmitting the collected data to a server;

[1411] means for training a particular classification model using data generated on the server;

[1412] a means for classifying waste using a classification model;

[1413] means for analyzing emotion data to recognize the emotion of a user;

[1414] a means for generating personalized sustainable practice information based on the classification results and the sentiment analysis results;

[1415] means for transmitting the generated sustainable practice information to a terminal;

[1416] A system including means for delivering sustainable practice information to a terminal.

[1417] (Claim 2)

[1418] 10. The system of claim 1, further comprising means for dividing the collected data into features and labels.

[1419] (Claim 3)

[1420] 10. The system of claim 1, further comprising means for evaluating performance of the classification model and the emotion engine.

[1421] "Application example 2 when combining emotion engines"

[1422] (Claim 1)

[1423] a means for collecting the generated data;

[1424] means for transmitting the collected data to a server;

[1425] means for training a particular classification model using data generated on the server;

[1426] a means for classifying waste using a classification model;

[1427] a means for optimizing collection routes based on the classification results;

[1428] a means for delivering sustainable practice information to the device;

[1429] means for collecting user emotion data;

[1430] A system that includes a means of analyzing emotional data to personalize sustainable practice information.

[1431] (Claim 2)

[1432] 10. The system of claim 1, further comprising means for dividing the collected data into features and labels.

[1433] (Claim 3)

[1434] 10. The system of claim 1, further comprising means for evaluating performance of the classification model. [Explanation of symbols]

[1435] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for collecting the generated data; means for transmitting the collected data to a server; means for training a particular classification model using data generated on the server; a means for classifying waste using a classification model; a means for optimizing collection routes based on the classification results; A system including means for delivering sustainable practice information to a terminal.

2. The system of claim 1 further comprising means for dividing the collected data into features and labels.

3. The system of claim 1 further comprising means for evaluating the performance of the classification model.

Images