System

Abstract

A system is provided.SOLUTION: A system comprising: means for loading historical data; means for pre-processing the loaded data; means for training a machine learning model using the pre-processed data; means for making predictions on a new data set using the trained model; and means for optimizing a waste collection route based on the prediction results.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] In Japanese cities, efficient waste recycling and disposal is a major challenge due to increasing environmental impacts and operational costs. Current waste management systems have difficulty operating efficiently due to inaccurate waste generation forecasts and insufficient optimization of collection routes. This situation has a negative impact on urban sustainability. This invention aims to provide a system that can accurately predict the amount of waste generated and optimize collection routes to make waste management more effective and sustainable. [Means for solving the problem]

[0005] This invention provides a system that includes a means for reading historical data, a means for preprocessing the read data, a means for training a machine learning model using the preprocessed data, a means for making predictions for a new dataset using the trained model, and a means for optimizing waste collection routes based on the prediction results. Specifically, after reading the historical data, missing values ​​are removed and the data is separated into features and a target variable. Next, a random forest regression model is used to train a model based on the features and target variable, and the trained model is used to make predictions for a new dataset. Then, collection routes are optimized based on the prediction results, making waste management efficient and sustainable.

[0006] "History data" refers to data recorded in the past, and in this invention refers to data including information such as the amount of waste generated, collection locations, and collection dates.

[0007] "Means of loading" refers to the process or function that takes data from outside and converts it into a format that can be used within the system.

[0008] "Preprocessing" refers to the process of preparing, cleaning, and transforming data before applying it to analytics or machine learning models.

[0009] A "machine learning model" refers to a mathematical model that learns patterns and regularities from large amounts of data and makes predictions based on future data.

[0010] "Method of making predictions" refers to the process of using a trained machine learning model to estimate outcomes based on new data.

[0011] "Means for optimizing collection routes" refers to the process of calculating the most efficient route for waste collection operations.

[0012] "Missing values" are values ​​that are missing from a dataset and may affect the overall quality of the data and the results of the analysis.

[0013] "Features" refer to the attributes and variables of input data that a machine learning model uses to make predictions.

[0014] The "target variable" refers to the variable that the machine learning model is trying to predict.

[0015] A "random forest regression model" is a type of machine learning algorithm that performs regression analysis using multiple decision trees, and refers to a model that makes more accurate predictions by averaging the predictions of individual decision trees. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. Hereinafter, an embodiment of the present invention will be described in detail.

[0038] System configuration

[0039] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[0040] Next, the server preprocesses the loaded historical data, which includes the following steps:

[0041] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0042] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[0043] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation patterns.

[0044] The server then uses the trained model to make predictions on a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[0045] Finally, the server optimizes waste collection routes based on the prediction results. The optimization algorithm sorts collection points in descending order of waste volume and calculates efficient collection routes.

[0046] Program processing flow

[0047] 1. Data loading:

[0048] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[0049] 2. Data Preprocessing:

[0050] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[0051] 3. Model training:

[0052] The server trains a random forest regression model using the preprocessed data.

[0053] 4. Prediction:

[0054] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[0055] 5. Collection route optimization:

[0056] The server optimizes collection routes based on the prediction results, which are determined by sorting collection points in descending order of waste volume.

[0057] Specific examples

[0058] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0059] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0060] The processing flow will be explained below.

[0061] Step 1:

[0062] The server reads historical data from an external database or a CSV file, such as waste_data.csv, which contains information such as the amount of waste generated, collection locations, collection dates, weather, and population density.

[0063] Step 2:

[0064] The server preprocesses the data it loads. First, it removes or imputes missing values. Then it reshapes the data to separate it into features (e.g., population density, weather) and target variables (waste generation volume). This results in a clean dataset for use in machine learning models.

[0065] Step 3:

[0066] The server trains a random forest regression model using the preprocessed data. Special settings, such as n_estimators=100 and random_state=42, are used to ensure model stability. Once trained, the model is used for future predictions.

[0067] Step 4:

[0068] The server makes predictions using a new dataset, which includes population density and weather data for days that have not yet been collected, and uses the trained random forest regression model to predict waste generation for these new data points.

[0069] Step 5:

[0070] The server optimizes collection routes based on the predicted waste generation volume. Specifically, it sorts collection points in descending order of waste volume based on the prediction results and calculates the most efficient collection route. This improves waste collection efficiency and reduces operation costs.

[0071] Step 6:

[0072] The system notifies the user of the results of the collection route. Based on the notified data, the user can appropriately deploy collection vehicles and begin collection work. The user can then efficiently collect waste by following the presented optimized route.

[0073] Through these steps, the system can improve the accuracy of waste generation predictions and optimize collection routes, thereby reducing environmental impact and operational costs.

[0074] Example 1

[0075] 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."

[0076] Conventional waste collection systems have difficulty predicting waste generation volumes, leading to insufficient optimization of collection routes. It is particularly difficult to effectively incorporate external factors, such as population density and weather, that affect waste generation volumes. As a result, efficient operation of collection vehicles is hindered, resulting in increased operating costs and adverse environmental impacts.

[0077] 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.

[0078] In this invention, the server further includes means for reading historical data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for the historical data including information such as waste generation amount, collection location, collection date, weather, and population density, means for deleting or completing missing values, and means for separating features and target variables. This enables highly accurate prediction of waste generation amount and calculation of efficient collection routes based on the predictions.

[0079] "Historical data" is data that includes information about past waste generation and collection, such as waste generation volume, collection locations, collection dates, weather, population density, and the like.

[0080] "Preprocessing" refers to the process of removing or complementing missing values ​​in a dataset and separating the features and objective variables required for prediction.

[0081] A "machine learning model" is a model that learns patterns and rules from input data and makes predictions about new data.

[0082] A "random forest regression model" is a machine learning technique that uses a large number of decision trees to improve prediction accuracy.

[0083] "Prediction" is the act of using a trained machine learning model to estimate waste generation for a new dataset.

[0084] "Waste collection route optimization" refers to the calculation of the optimal route for efficiently visiting collection points based on the predicted amount of waste generated.

[0085] A "new dataset" is data that includes information such as population density and weather for future collection dates.

[0086] "Missing values" are values ​​that are missing in a dataset and must be handled appropriately.

[0087] "Features" are attributes or variables of input data that a machine learning model uses to make predictions.

[0088] A "target variable" is an attribute or variable of the output data that a machine learning model is intended to predict.

[0089] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. Specifically, it consists of the following steps:

[0090] First, the server loads historical data on waste generation and collection from an external database or a CSV file. This historical data includes information such as the amount of waste generated, collection locations, collection dates, weather, and population density. For example, data can be loaded from a file called "waste_data.csv."

[0091] Next, the historical data loaded by the server is preprocessed. Preprocessing includes removing missing values ​​and separating features and target variables. Missing values ​​are values ​​that are missing from a dataset, and are handled appropriately using methods such as mean value imputation and data deletion. Features refer to the attributes and variables of data required for prediction, such as population density and weather, while target variables refer to the data to be predicted, such as waste generation volume.

[0092] Using the preprocessed data, the server trains a random forest regression model, a machine learning technique that uses a large number of decision trees to improve prediction accuracy. The server uses this random forest model to learn from past data and generate a model capable of predicting future waste generation patterns.

[0093] The trained model is then used to make predictions on a new dataset, which includes population density and weather data for a specific collection day. The user enters this information, and the server uses it to predict future waste generation. For example, to create a collection plan for a new week, the generative AI model can be given the following prompt:

[0094] "Predict waste generation based on population density and weather data for the new week, and optimize waste collection routes accordingly."

[0095] Finally, the server optimizes waste collection routes based on the prediction results. Using an optimization algorithm, collection points are sorted in descending order of waste volume, and an efficient collection route is calculated. This collection route information is output to a terminal and can be checked by the user. It is also reflected in collection vehicle schedules, improving operational efficiency.

[0096] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0097] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0098] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0099] Step 1:

[0100] Data loading

[0101] The server reads historical data from an external database or a CSV file (e.g., "waste_data.csv"). The input data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. The server stores this data in memory and prepares it for subsequent preprocessing steps. Specifically, the server specifies the file path and opens it to read the data.

[0102] Step 2:

[0103] Data preprocessing (missing value handling)

[0104] The server detects missing values ​​in the dataset loaded and performs deletion or imputation processing. Historical data is given as input. For example, if there are missing values ​​in weather data, they are imputed using historical average weather data. The output is a clean dataset with the missing values ​​processed. Specifically, it scans the dataset and processes missing values ​​in each column using the appropriate method.

[0105] Step 3:

[0106] Data preprocessing (separation of features and target variables)

[0107] The server uses a clean dataset to separate the features and target variables required for prediction. The input is a dataset with missing values ​​processed. The features include population density and weather data, and the target variables include waste generation volume. The output is a dataset separated into the features and target variables. Specifically, the data is separated based on each column.

[0108] Step 4:

[0109] Model training

[0110] The server trains a random forest regression model using the preprocessed data. The input is the separated features and the target variable. The server applies the random forest regression algorithm to generate and combine multiple decision trees. The output is a trained random forest model. Specifically, the data is shuffled to separate it into a training set and a validation set, and the model is optimized through cross-validation.

[0111] Step 5:

[0112] Entering New Data

[0113] The user inputs population density and weather data for a new collection date. The input is weather data and demographic information for a new week. The output is data formatted in a way that the server can use for predictions. Specifically, the user inputs new data through a web interface, and the server receives and preprocesses the data.

[0114] Step 6:

[0115] Model prediction

[0116] The server uses the new dataset to make predictions using the trained model. The input is the feature data for the new collection date. The output is a prediction of future waste generation volume. Specifically, the server inputs the preprocessed new data into the model and calculates the predicted value.

[0117] Step 7:

[0118] Collection route optimization

[0119] The server optimizes waste collection routes based on the prediction results. The input is the predicted waste generation data. The server uses an optimization algorithm to sort collection points in descending order of waste volume and calculates an efficient collection route. The output is optimized collection route information. Specifically, the algorithm is run based on the prediction results, and the collection route is calculated and visualized.

[0120] Step 8:

[0121] Output of results

[0122] The server outputs the optimized collection route to the terminal and displays it to the user. The optimized route information is given as input. The output is collection route information in a display format that can be confirmed by the user. Specifically, the calculation results are generated in table or map format and sent to the terminal.

[0123] (Application example 1)

[0124] 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."

[0125] Conventional waste collection systems lack the means to predict waste generation patterns and design efficient routes, resulting in poor collection efficiency. Similarly, in the supply of parts within factories, there is a lack of a means to design efficient supply routes based on past consumption data, resulting in wasted time and money in parts supply. To address these issues, the present invention aims to optimize waste collection and in-factory parts supply routes by applying preprocessing of historical data and machine learning models.

[0126] 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.

[0127] In this invention, the server includes a means for reading historical data, a means for preprocessing the read data, a means for training a machine learning model using the preprocessed data, a means for making predictions for a new data set using the trained model, a means for optimizing waste collection routes based on the prediction results, and a means for optimizing parts supply routes within the factory by applying a similar system, thereby enabling the efficiency of not only waste collection but also parts supply within the factory.

[0128] "Historical data" refers to recorded data about past events or activities, and in this context refers to waste generation and parts consumption data.

[0129] "Preprocessing" refers to the process of correcting or supplementing incomplete data and preparing it in an appropriate format prior to data analysis or training a machine learning model.

[0130] A "machine learning model" is a set of mathematical algorithms that learn patterns from past data and make predictions or classifications for new data.

[0131] "Training" is the process of using historical data to train a machine learning model and improve its accuracy.

[0132] "Prediction" is the act of using existing data and a trained machine learning model to estimate the value of future data points or events.

[0133] "Optimization" is the process of finding the most efficient use of available resources based on specific objectives and constraints.

[0134] "Waste collection route" refers to the specific route or sequence used to collect waste.

[0135] "Parts supply route" refers to the route and order for efficiently delivering parts within a factory to each workplace and process.

[0136] "Dataset" refers to a collection of data used for a particular calculation or analysis.

[0137] A "random forest regression model" is a type of machine learning algorithm that uses multiple decision trees to perform regression analysis.

[0138] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection and in-factory parts supply routes. Hereinafter, an embodiment of the present invention will be described in detail.

[0139] System configuration

[0140] First, the server loads data on past waste generation and collection (historical data) from an external database or CSV file. Similarly, it loads data on in-factory parts consumption. This historical data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. In-factory parts consumption data includes the amount of parts consumed, consumption location, consumption date, and work progress status in the factory.

[0141] Next, the server preprocesses the loaded history data, which includes the following steps:

[0142] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0143] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather, factory work progress) from the target variables (waste generation volume, parts consumption volume).

[0144] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation and component consumption patterns.

[0145] The server then uses the trained model to make predictions on a new dataset, which includes population density, weather data, and factory work progress data for a specific collection or consumption date, and predicts waste generation and parts consumption based on these inputs.

[0146] Finally, the server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. The optimization algorithm sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes.

[0147] Hardware and Software Configuration

[0148] This system uses the following hardware and software:

[0149] Hardware: Servers, factory robots, smart glasses

[0150] Software: Python, Pandas, Scikit-learn

[0151] The server loads historical data and performs data preprocessing using Pandas. It then uses Scikit-learn to train a random forest regression model to make predictions based on new data sets. The optimized collection and delivery routes are displayed or instructed to factory robots or smart glasses.

[0152] Specific examples

[0153] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0154] Within the factory, parts consumption data from the past year is used to predict parts consumption based on the new week's data (population density and factory work progress).Based on this prediction, robots within the factory operate optimal supply routes, aiming to achieve efficient parts supply.

[0155] Example prompt sentence:

[0156] "Generate Python code that uses the past year's in-factory parts consumption data (e.g., CSV file) to predict parts consumption based on the new week's data (population density and weather). Include code that optimizes parts supply routes within the factory based on the predictions."

[0157] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0158] Step 1:

[0159] The server reads the historical data. Specifically, the server retrieves data on past waste generation and parts consumption from an external database or a CSV file (e.g., "factory_parts_data.csv"). The input is the historical data, and the output is the retrieved dataset.

[0160] Step 2:

[0161] The data loaded by the server is preprocessed. Specifically, missing values ​​are removed and the features required for prediction (e.g., population density, weather, factory work progress) are separated from the target variables (waste generation volume, parts consumption volume). The input is the acquired dataset, and the output is the preprocessed dataset.

[0162] Step 3:

[0163] The server uses the preprocessed data to train a machine learning model. Specifically, it uses a random forest regression model to learn patterns of waste generation and component consumption from historical data. The input is the preprocessed dataset, and the output is the trained model.

[0164] Step 4:

[0165] The server uses the trained model to make predictions for a new data set. Specifically, it predicts the amount of waste generated and parts consumed using inputs such as population density, weather data, and factory work progress data for a new collection or consumption date. The input is the new data set, and the output is the prediction result.

[0166] Step 5:

[0167] The server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. Specifically, it sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes. The input is the prediction results, and the output is optimized route information.

[0168] Step 6:

[0169] The server sends the optimized waste collection route and parts supply route to related terminals and devices (e.g., factory robots, smart glasses). Specifically, the route information is displayed or instructed on a display device or control system. The input is the optimized route information, and the output is a confirmation of transmission to the terminal.

[0170] 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.

[0171] The present invention is a system in which a server reads historical data, preprocesses the data, and trains a machine learning model to predict waste generation patterns and optimize waste collection routes. Furthermore, by combining it with an emotion engine that recognizes user emotions, the user experience can be improved. The following describes embodiments of the present invention in detail.

[0172] System configuration

[0173] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[0174] Next, the server preprocesses the loaded historical data, which includes the following steps:

[0175] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0176] 2. Separation of features and target variables: The data is shaped to separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[0177] The server uses the preprocessed data to train a random forest regression model, which is capable of learning waste generation patterns from past data and predicting future waste generation.

[0178] The server then uses the trained model to make waste generation predictions for a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[0179] Furthermore, we use an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the collected results and system suggestions.

[0180] The server optimizes collection routes based on the waste generation prediction results and user emotion data. Based on the information obtained from the emotion engine, adjustments are made taking complaints and feedback into consideration, and a collection route that satisfies the user is calculated.

[0181] Program processing flow

[0182] 1. Data loading:

[0183] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[0184] 2. Data Preprocessing:

[0185] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[0186] 3. Model training:

[0187] The server trains a random forest regression model using the preprocessed data.

[0188] 4. Prediction:

[0189] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[0190] 5. Collecting Emotional Data:

[0191] The server uses an emotion engine to collect user emotion data, measure how users react to the system, and collect positive or negative feedback.

[0192] 6. Collection route optimization:

[0193] The server optimizes collection routes based on waste generation prediction results and emotion data, thereby calculating efficient routes that will increase user satisfaction.

[0194] Specific examples

[0195] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation. It then uses an emotion engine to collect user feedback and takes this into account when optimizing collection routes.

[0196] In this way, the introduction of the emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[0197] The processing flow will be explained below.

[0198] Step 1:

[0199] The server reads historical data from an external database or a CSV file. For example, it retrieves information such as the amount of waste generated, collection location, collection date, weather, and population density from a file called waste_data.csv.

[0200] Step 2:

[0201] The server preprocesses the data it has loaded. Specifically, it removes or imputes missing values ​​and separates the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). This process results in a clean dataset.

[0202] Step 3:

[0203] The server trains a random forest regression model using the preprocessed data. It uses specific parameter settings (e.g., n_estimators=100, random_state=42) to train the model stably and efficiently.

[0204] Step 4:

[0205] The server uses the trained model to make predictions on a new dataset, which includes population density and weather data for days that haven't been collected. Based on these inputs, the server predicts new waste generation volumes.

[0206] Step 5:

[0207] The server uses an emotion engine to collect user emotion data. Specifically, it identifies situations where users provide feedback to the system, identifies positive or negative reactions, and records them in a database. This information is collected using methods such as text analysis and speech analysis.

[0208] Step 6:

[0209] The server analyzes the collected emotion data and optimizes the collection route based on the prediction results. Based on the emotion data, the collection route is adjusted to satisfy the user, balancing efficiency and user experience. For example, if a user expresses dissatisfaction with a previous collection route, improvements can be made based on that information.

[0210] Step 7:

[0211] The server notifies the user of the optimized collection route, including specific collection dates and times and route information. Based on this information, the user can operate the collection vehicle efficiently.

[0212] The above is a detailed explanation of each processing step for the server, terminal, and user, which allows the waste management system to achieve high prediction accuracy and user experience, while reducing environmental impact and operational costs.

[0213] Example 2

[0214] 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."

[0215] Conventional waste collection systems have low accuracy in predicting waste generation, which makes it difficult to optimize collection routes. Furthermore, adjustments were not made taking into account user feedback or emotions, which prevented improvements in user satisfaction. Therefore, there was a need for efficient collection route calculations and adjustments that reflected user emotions.

[0216] 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.

[0217] In this invention, the server includes means for reading historical data, means for preprocessing the read data, means for training a predictive model using the preprocessed data, means for making predictions for a new data set using the trained predictive model, means for collecting emotion data, means for optimizing waste collection routes based on the prediction results, and means for adjusting the optimization results using the emotion data. This not only enables highly accurate prediction of waste generation patterns, but also realizes optimization of collection routes taking user emotions into consideration, enabling efficient collection operations and improved user satisfaction.

[0218] "Historical data" refers to data that includes information such as the amount of waste generated in the past, collection locations, collection dates, weather, and population density.

[0219] "Preprocessing" refers to the process of deleting missing values ​​from the loaded data and separating it into features and target variables.

[0220] A "predictive model" is a machine learning algorithm model that is trained using pre-processed data and used to predict future waste generation.

[0221] "Emotional data" is data that measures a user's emotions and collects negative or positive feedback.

[0222] "Waste collection route optimization" is the process of efficiently designing collection routes based on the results of forecasting waste generation amounts.

[0223] "Adjustment" refers to the process of rearranging collection routes optimized using emotion data to reflect user feedback.

[0224] A "regression algorithm" is a machine learning technique used to predict continuous values, and in the present system may include a random forest regression model.

[0225] A "new dataset" is data, including population density and weather data for future collection dates, that is used in the predictive model.

[0226] The present invention is a system in which a server reads historical data, pre-processes the data, and trains a predictive model to predict waste generation patterns and optimize waste collection routes. Furthermore, the server can collect sentiment data and make adjustments based on user feedback to improve the user experience. The following describes an embodiment of the present invention in detail.

[0227] First, the server reads historical data on past waste generation and collection from an external database or CSV file. This historical data includes information such as waste generation volume, collection location, collection date, weather, population density, etc. The server obtains the data from, for example, a database management system or file system.

[0228] Next, the historical data loaded by the server is preprocessed. Preprocessing includes removing or completing missing values, and separating the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). A data analysis library (e.g., Pandas) is used for data preprocessing.

[0229] The server uses the preprocessed data to train a predictive model. This predictive model is capable of learning waste generation patterns from past data and predicting future waste generation. For training, a regression algorithm from a machine learning library (e.g., Scikit-learn) is used.

[0230] The server then uses the trained model to predict waste generation for a new dataset, which includes population density and weather data for the next collection day or week, and predicts waste generation based on these inputs.

[0231] Furthermore, the server uses an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the waste collection system. The server performs emotion analysis of user feedback using, for example, a text analysis API.

[0232] The server optimizes waste collection routes based on the prediction results and emotion data. Optimization algorithms such as linear programming and genetic algorithms are used to optimize collection routes. The emotion data is also used to adjust collection routes to reflect user feedback and complaints.

[0233] Specific examples

[0234] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains a predictive model. After training, it inputs population density and weather data for the next week to predict future waste generation. It then uses an emotion engine to collect feedback from users and optimizes collection routes accordingly.

[0235] Examples of prompts include:

[0236] "Predict next week's waste generation volume based on the past year's waste generation volume data, collection locations, collection dates, weather, and population density information. Additionally, suggest optimal collection routes based on user feedback."

[0237] In this way, the introduction of the emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[0238] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0239] Step 1:

[0240] The server reads historical data. Specifically, the server retrieves data such as past waste generation volume, collection locations, collection dates, weather, and population density from an external database or the "waste_data.csv" file. The input is the "waste_data.csv" file, and the output is a dataset in Pandas DataFrame format.

[0241] Step 2:

[0242] The server preprocesses the historical data. Specifically, the server removes or imputes missing values ​​and separates the data into the features and target variables required for prediction. The input is a DataFrame of historical data, and the output is the preprocessed feature data and target variable data in DataFrame format.

[0243] Step 3:

[0244] The server trains a predictive model. Specifically, the server trains a machine learning model using the preprocessed data. It uses Scikit-learn's random forest regression model. The input is the preprocessed feature data and the objective variable data, and the output is a trained predictive model.

[0245] Step 4:

[0246] The server makes a prediction based on the new dataset. Specifically, the server uses a prediction model to predict waste generation volume using population density and weather data for the next collection day as input. The input is the new dataset, and the output is the predicted waste generation volume.

[0247] Step 5:

[0248] The server collects emotion data. Specifically, it collects feedback from users via their devices and performs emotion analysis using an emotion engine. The input is the user's feedback (in text format), and the output is the emotion analysis result (positive, negative, etc.).

[0249] Step 6:

[0250] The server optimizes collection routes using emotion data. Specifically, the server calculates the optimal waste collection route based on the prediction results and emotion data. Linear programming and genetic algorithms are used for optimization. The inputs are the predicted waste generation volume and emotion data, and the output is the optimized collection route.

[0251] The above are the specific processing steps of this system's program. We have explained the input and output at each step, as well as the specific data processing and data calculation, to clarify the overall flow of the system.

[0252] (Application example 2)

[0253] 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."

[0254] In modern factories, efficient waste management and the development of optimal collection routes are important challenges. In particular, irregular waste generation patterns and low user (factory manager) satisfaction with collection services pose concerns about increased operational costs and environmental impact. Furthermore, adjusting collection routes based on user feedback is likely to contribute to further efficiency and improved satisfaction. Given this background, existing waste management systems are required to be equipped with highly accurate waste generation predictions and the ability to collect and reflect user sentiment data.

[0255] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for reading history data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for recognizing and collecting user emotion data, and means for adjusting collection routes based on the user emotion data. This not only improves the accuracy of waste generation predictions and provides optimized collection routes, but also enables efficient waste management with high user satisfaction by adjusting the routes based on user emotions.

[0256] "Historical data" means information relating to past waste generation and collection, including data such as waste generation volume, collection locations, collection dates, weather, and population density.

[0257] "Preprocessing" is a data processing method that removes missing values ​​and separates features and target variables in order to prepare the data in a format suitable for machine learning model training.

[0258] A "machine learning model" is an algorithm that uses given data to learn patterns and make predictions or classifications for future data.

[0259] A "random forest regression model" is a machine learning model that makes predictions by combining multiple decision trees, and is particularly effective when applied to regression problems, enabling highly accurate predictions.

[0260] An "emotion engine" is a system that recognizes users' emotions, collects and analyzes that data, and is used to understand users' feedback and emotional state.

[0261] "Collection route optimization" is a means of minimizing costs and time by adjusting collection routes based on prediction results and user emotion data in order to achieve efficient waste collection.

[0262] "User emotion data" refers to data collected by the emotion engine regarding the user's emotional state, including positive or negative feedback.

[0263] A "new dataset" is new input data, such as population density or weather on an uncollected day, that the trained model will use to make predictions.

[0264] "Missing values" are incomplete or missing data in a dataset that should be appropriately handled to improve the accuracy of the model.

[0265] This invention is a system for streamlining waste management within factories. The system utilizes historical data and machine learning models to predict waste generation patterns and optimize collection routes. Furthermore, by collecting user sentiment data, the system can be applied to adjusting collection routes.

[0266] Hardware and software used

[0267] Hardware: Factory robots, servers

[0268] Software: Python, TensorFlow, OpenCV (for emotion recognition), pandas (data preprocessing), scikit-learn (machine learning)

[0269] System program processing explanation

[0270] 1. Data loading

[0271] The server reads data on historical waste generation and collection from an external database or a CSV file, including information on waste generation volume, collection location, collection date, weather, population density, etc. For example, it retrieves data from a file called "waste_data.csv."

[0272] 2. Data Preprocessing

[0273] The server performs data preprocessing. In this step, missing values ​​are removed and feature values ​​and target variables are separated. Specifically, feature values ​​such as population density and weather are separated from target variables such as waste generation volume in order to input them into the machine learning model. The Python pandas library is used to format the data and process missing values.

[0274] 3. Model training

[0275] The server uses the preprocessed data to train a random forest regression model. Using the scikit-learn library, the model is built using the training data, enabling highly accurate predictions of waste generation.

[0276] 4. Prediction

[0277] The server uses the trained model to predict waste generation for a new dataset, such as population density or weather data for a specific day, and uses Python and a scikit-learn random forest regression model to make predictions based on this data.

[0278] 5. Collecting Emotional Data

[0279] The emotion engine collects emotional data from users such as factory managers. Specifically, it combines a camera using OpenCV with a Python script to recognize emotions from the user's facial expressions and analyzes positive or negative feedback.

[0280] 6. Optimizing collection routes

[0281] The server optimizes collection routes based on waste generation prediction results and user emotion data. This optimization process takes into account information obtained from emotion data and makes adjustments to increase user satisfaction. For example, if a user expresses negative emotions, the server takes measures such as changing the collection route or collection frequency.

[0282] Examples of concrete examples and prompts

[0283] For example, at a factory in a certain city, the server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. It then uses an emotion engine to collect feedback from the factory manager and takes this into consideration to optimize collection routes. An example of a prompt would be, "Predict the amount of waste generated within the factory and calculate the optimal collection route based on the user (manager) feedback."

[0284] In this way, the present invention simultaneously realizes efficient waste management and improved user satisfaction.

[0285] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0286] Step 1:

[0287] The server reads historical data about past waste generation and collection from a central database or CSV files. This input data includes waste generation volume, collection location, collection date, weather, population density, and other parameters. The obtained data frame is generated as output.

[0288] Step 2:

[0289] The server preprocesses the loaded data. Specifically, it removes missing values ​​and separates features (e.g., population density, weather) from the target variable (waste generation volume). It uses the pandas library to format the data and process missing values, and outputs the preprocessed dataset.

[0290] Step 3:

[0291] The server trains a random forest regression model using the preprocessed dataset. Using the Scikit-learn library, it builds a model from the training data and gives it the ability to predict waste generation. The output is a trained machine learning model.

[0292] Step 4:

[0293] The server uses the trained model to predict waste generation for a new dataset, which may include population density and weather information for a specific day. By inputting this new data into the trained model, a waste generation prediction result is obtained. The output is a predicted value data.

[0294] Step 5:

[0295] The device uses the emotion engine to collect user emotion data. Specifically, it uses a camera and the OpenCV library to analyze the user's face and recognize their emotional status, such as positive or negative. The detection results are output as emotion data.

[0296] Step 6:

[0297] The server optimizes waste collection routes based on the prediction results and the user's emotional data. If the collected emotional data indicates negative feedback, it adjusts the collection route or collection frequency. For example, it recalculates the collection route to increase user satisfaction or changes the collection frequency. This optimized collection route information is output.

[0298] The above are the specific processing steps of the system that realizes the application example. The specific operations performed at each step realize efficient waste management and improved user satisfaction.

[0299] 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.

[0300] 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.

[0301] 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.

[0302] [Second embodiment]

[0303] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0304] 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.

[0305] 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).

[0306] 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.

[0307] 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.

[0308] 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).

[0309] 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.

[0310] 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.

[0311] 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.

[0312] 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.

[0313] 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.

[0314] 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."

[0315] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. Hereinafter, an embodiment of the present invention will be described in detail.

[0316] System configuration

[0317] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[0318] Next, the server preprocesses the loaded historical data, which includes the following steps:

[0319] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0320] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[0321] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation patterns.

[0322] The server then uses the trained model to make predictions on a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[0323] Finally, the server optimizes waste collection routes based on the prediction results. The optimization algorithm sorts collection points in descending order of waste volume and calculates efficient collection routes.

[0324] Program processing flow

[0325] 1. Data loading:

[0326] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[0327] 2. Data Preprocessing:

[0328] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[0329] 3. Model training:

[0330] The server trains a random forest regression model using the preprocessed data.

[0331] 4. Prediction:

[0332] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[0333] 5. Collection route optimization:

[0334] The server optimizes collection routes based on the prediction results, which are determined by sorting collection points in descending order of waste volume.

[0335] Specific examples

[0336] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0337] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0338] The processing flow will be explained below.

[0339] Step 1:

[0340] The server reads historical data from an external database or a CSV file, such as waste_data.csv, which contains information such as the amount of waste generated, collection locations, collection dates, weather, and population density.

[0341] Step 2:

[0342] The server preprocesses the data it loads. First, it removes or imputes missing values. Then it reshapes the data to separate it into features (e.g., population density, weather) and target variables (waste generation volume). This results in a clean dataset for use in machine learning models.

[0343] Step 3:

[0344] The server trains a random forest regression model using the preprocessed data. Special settings, such as n_estimators=100 and random_state=42, are used to ensure model stability. Once trained, the model is used for future predictions.

[0345] Step 4:

[0346] The server makes predictions using a new dataset, which includes population density and weather data for days that have not yet been collected, and uses the trained random forest regression model to predict waste generation for these new data points.

[0347] Step 5:

[0348] The server optimizes collection routes based on the predicted waste generation volume. Specifically, it sorts collection points in descending order of waste volume based on the prediction results and calculates the most efficient collection route. This improves waste collection efficiency and reduces operation costs.

[0349] Step 6:

[0350] The system notifies the user of the results of the collection route. Based on the notified data, the user can appropriately deploy collection vehicles and begin collection work. The user can then efficiently collect waste by following the presented optimized route.

[0351] Through these steps, the system can improve the accuracy of waste generation predictions and optimize collection routes, thereby reducing environmental impact and operational costs.

[0352] Example 1

[0353] 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."

[0354] Conventional waste collection systems have difficulty predicting waste generation volumes, leading to insufficient optimization of collection routes. It is particularly difficult to effectively incorporate external factors, such as population density and weather, that affect waste generation volumes. As a result, efficient operation of collection vehicles is hindered, resulting in increased operating costs and adverse environmental impacts.

[0355] 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.

[0356] In this invention, the server further includes means for reading historical data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for the historical data including information such as waste generation amount, collection location, collection date, weather, and population density, means for deleting or completing missing values, and means for separating features and target variables. This enables highly accurate prediction of waste generation amount and calculation of efficient collection routes based on the predictions.

[0357] "Historical data" is data that includes information about past waste generation and collection, such as waste generation volume, collection locations, collection dates, weather, population density, and the like.

[0358] "Preprocessing" refers to the process of removing or complementing missing values ​​in a dataset and separating the features and objective variables required for prediction.

[0359] A "machine learning model" is a model that learns patterns and rules from input data and makes predictions about new data.

[0360] A "random forest regression model" is a machine learning technique that uses a large number of decision trees to improve prediction accuracy.

[0361] "Prediction" is the act of using a trained machine learning model to estimate waste generation for a new dataset.

[0362] "Waste collection route optimization" refers to the calculation of the optimal route for efficiently visiting collection points based on the predicted amount of waste generated.

[0363] A "new dataset" is data that includes information such as population density and weather for future collection dates.

[0364] "Missing values" are values ​​that are missing in a dataset and must be handled appropriately.

[0365] "Features" are attributes or variables of input data that a machine learning model uses to make predictions.

[0366] A "target variable" is an attribute or variable of the output data that a machine learning model is intended to predict.

[0367] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. Specifically, it consists of the following steps:

[0368] First, the server loads historical data on waste generation and collection from an external database or a CSV file. This historical data includes information such as the amount of waste generated, collection locations, collection dates, weather, and population density. For example, data can be loaded from a file called "waste_data.csv."

[0369] Next, the historical data loaded by the server is preprocessed. Preprocessing includes removing missing values ​​and separating features and target variables. Missing values ​​are values ​​that are missing from a dataset, and are handled appropriately using methods such as mean value imputation and data deletion. Features refer to the attributes and variables of data required for prediction, such as population density and weather, while target variables refer to the data to be predicted, such as waste generation volume.

[0370] Using the preprocessed data, the server trains a random forest regression model, a machine learning technique that uses a large number of decision trees to improve prediction accuracy. The server uses this random forest model to learn from past data and generate a model capable of predicting future waste generation patterns.

[0371] The trained model is then used to make predictions on a new dataset, which includes population density and weather data for a specific collection day. The user enters this information, and the server uses it to predict future waste generation. For example, to create a collection plan for a new week, the generative AI model can be given the following prompt:

[0372] "Predict waste generation based on population density and weather data for the new week, and optimize waste collection routes accordingly."

[0373] Finally, the server optimizes waste collection routes based on the prediction results. Using an optimization algorithm, collection points are sorted in descending order of waste volume, and an efficient collection route is calculated. This collection route information is output to a terminal and can be checked by the user. It is also reflected in collection vehicle schedules, improving operational efficiency.

[0374] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0375] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0376] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0377] Step 1:

[0378] Data loading

[0379] The server reads historical data from an external database or a CSV file (e.g., "waste_data.csv"). The input data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. The server stores this data in memory and prepares it for subsequent preprocessing steps. Specifically, the server specifies the file path and opens it to read the data.

[0380] Step 2:

[0381] Data preprocessing (missing value handling)

[0382] The server detects missing values ​​in the dataset loaded and performs deletion or imputation processing. Historical data is given as input. For example, if there are missing values ​​in weather data, they are imputed using historical average weather data. The output is a clean dataset with the missing values ​​processed. Specifically, it scans the dataset and processes missing values ​​in each column using the appropriate method.

[0383] Step 3:

[0384] Data preprocessing (separation of features and target variables)

[0385] The server uses a clean dataset to separate the features and target variables required for prediction. The input is a dataset with missing values ​​processed. The features include population density and weather data, and the target variables include waste generation volume. The output is a dataset separated into the features and target variables. Specifically, the data is separated based on each column.

[0386] Step 4:

[0387] Model training

[0388] The server trains a random forest regression model using the preprocessed data. The input is the separated features and the target variable. The server applies the random forest regression algorithm to generate and combine multiple decision trees. The output is a trained random forest model. Specifically, the data is shuffled to separate it into a training set and a validation set, and the model is optimized through cross-validation.

[0389] Step 5:

[0390] Entering New Data

[0391] The user inputs population density and weather data for a new collection date. The input is weather data and demographic information for a new week. The output is data formatted in a way that the server can use for predictions. Specifically, the user inputs new data through a web interface, and the server receives and preprocesses the data.

[0392] Step 6:

[0393] Model prediction

[0394] The server uses the new dataset to make predictions using the trained model. The input is the feature data for the new collection date. The output is a prediction of future waste generation volume. Specifically, the server inputs the preprocessed new data into the model and calculates the predicted value.

[0395] Step 7:

[0396] Collection route optimization

[0397] The server optimizes waste collection routes based on the prediction results. The input is the predicted waste generation data. The server uses an optimization algorithm to sort collection points in descending order of waste volume and calculates an efficient collection route. The output is optimized collection route information. Specifically, the algorithm is run based on the prediction results, and the collection route is calculated and visualized.

[0398] Step 8:

[0399] Output of results

[0400] The server outputs the optimized collection route to the terminal and displays it to the user. The optimized route information is given as input. The output is collection route information in a display format that can be confirmed by the user. Specifically, the calculation results are generated in table or map format and sent to the terminal.

[0401] (Application example 1)

[0402] 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."

[0403] Conventional waste collection systems lack the means to predict waste generation patterns and design efficient routes, resulting in poor collection efficiency. Similarly, in the supply of parts within factories, there is a lack of a means to design efficient supply routes based on past consumption data, resulting in wasted time and money in parts supply. To address these issues, the present invention aims to optimize waste collection and in-factory parts supply routes by applying preprocessing of historical data and machine learning models.

[0404] 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.

[0405] In this invention, the server includes a means for reading historical data, a means for preprocessing the read data, a means for training a machine learning model using the preprocessed data, a means for making predictions for a new data set using the trained model, a means for optimizing waste collection routes based on the prediction results, and a means for optimizing parts supply routes within the factory by applying a similar system, thereby enabling the efficiency of not only waste collection but also parts supply within the factory.

[0406] "Historical data" refers to recorded data about past events or activities, and in this context refers to waste generation and parts consumption data.

[0407] "Preprocessing" refers to the process of correcting or supplementing incomplete data and preparing it in an appropriate format prior to data analysis or training a machine learning model.

[0408] A "machine learning model" is a set of mathematical algorithms that learn patterns from past data and make predictions or classifications for new data.

[0409] "Training" is the process of using historical data to train a machine learning model and improve its accuracy.

[0410] "Prediction" is the act of using existing data and a trained machine learning model to estimate the value of future data points or events.

[0411] "Optimization" is the process of finding the most efficient use of available resources based on specific objectives and constraints.

[0412] "Waste collection route" refers to the specific route or sequence used to collect waste.

[0413] "Parts supply route" refers to the route and order for efficiently delivering parts within a factory to each workplace and process.

[0414] "Dataset" refers to a collection of data used for a particular calculation or analysis.

[0415] A "random forest regression model" is a type of machine learning algorithm that uses multiple decision trees to perform regression analysis.

[0416] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection and in-factory parts supply routes. Hereinafter, an embodiment of the present invention will be described in detail.

[0417] System configuration

[0418] First, the server loads data on past waste generation and collection (historical data) from an external database or CSV file. Similarly, it loads data on in-factory parts consumption. This historical data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. In-factory parts consumption data includes the amount of parts consumed, consumption location, consumption date, and work progress status in the factory.

[0419] Next, the server preprocesses the loaded history data, which includes the following steps:

[0420] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0421] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather, factory work progress) from the target variables (waste generation volume, parts consumption volume).

[0422] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation and component consumption patterns.

[0423] The server then uses the trained model to make predictions on a new dataset, which includes population density, weather data, and factory work progress data for a specific collection or consumption date, and predicts waste generation and parts consumption based on these inputs.

[0424] Finally, the server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. The optimization algorithm sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes.

[0425] Hardware and Software Configuration

[0426] This system uses the following hardware and software:

[0427] Hardware: Servers, factory robots, smart glasses

[0428] Software: Python, Pandas, Scikit-learn

[0429] The server loads historical data and performs data preprocessing using Pandas. It then uses Scikit-learn to train a random forest regression model to make predictions based on new data sets. The optimized collection and delivery routes are displayed or instructed to factory robots or smart glasses.

[0430] Specific examples

[0431] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0432] Within the factory, parts consumption data from the past year is used to predict parts consumption based on the new week's data (population density and factory work progress).Based on this prediction, robots within the factory operate optimal supply routes, aiming to achieve efficient parts supply.

[0433] Example prompt sentence:

[0434] "Generate Python code that uses the past year's in-factory parts consumption data (e.g., CSV file) to predict parts consumption based on the new week's data (population density and weather). Include code that optimizes parts supply routes within the factory based on the predictions."

[0435] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0436] Step 1:

[0437] The server reads the historical data. Specifically, the server retrieves data on past waste generation and parts consumption from an external database or a CSV file (e.g., "factory_parts_data.csv"). The input is the historical data, and the output is the retrieved dataset.

[0438] Step 2:

[0439] The data loaded by the server is preprocessed. Specifically, missing values ​​are removed and the features required for prediction (e.g., population density, weather, factory work progress) are separated from the target variables (waste generation volume, parts consumption volume). The input is the acquired dataset, and the output is the preprocessed dataset.

[0440] Step 3:

[0441] The server uses the preprocessed data to train a machine learning model. Specifically, it uses a random forest regression model to learn patterns of waste generation and component consumption from historical data. The input is the preprocessed dataset, and the output is the trained model.

[0442] Step 4:

[0443] The server uses the trained model to make predictions for a new data set. Specifically, it predicts the amount of waste generated and parts consumed using inputs such as population density, weather data, and factory work progress data for a new collection or consumption date. The input is the new data set, and the output is the prediction result.

[0444] Step 5:

[0445] The server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. Specifically, it sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes. The input is the prediction results, and the output is optimized route information.

[0446] Step 6:

[0447] The server sends the optimized waste collection route and parts supply route to related terminals and devices (e.g., factory robots, smart glasses). Specifically, the route information is displayed or instructed on a display device or control system. The input is the optimized route information, and the output is a confirmation of transmission to the terminal.

[0448] 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.

[0449] The present invention is a system in which a server reads historical data, preprocesses the data, and trains a machine learning model to predict waste generation patterns and optimize waste collection routes. Furthermore, by combining it with an emotion engine that recognizes user emotions, the user experience can be improved. The following describes embodiments of the present invention in detail.

[0450] System configuration

[0451] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[0452] Next, the server preprocesses the loaded historical data, which includes the following steps:

[0453] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0454] 2. Separation of features and target variables: The data is shaped to separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[0455] The server uses the preprocessed data to train a random forest regression model, which is capable of learning waste generation patterns from past data and predicting future waste generation.

[0456] The server then uses the trained model to make waste generation predictions for a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[0457] Furthermore, we use an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the collected results and system suggestions.

[0458] The server optimizes collection routes based on the waste generation prediction results and user emotion data. Based on the information obtained from the emotion engine, adjustments are made taking complaints and feedback into consideration, and a collection route that satisfies the user is calculated.

[0459] Program processing flow

[0460] 1. Data loading:

[0461] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[0462] 2. Data Preprocessing:

[0463] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[0464] 3. Model training:

[0465] The server trains a random forest regression model using the preprocessed data.

[0466] 4. Prediction:

[0467] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[0468] 5. Collecting Emotional Data:

[0469] The server uses an emotion engine to collect user emotion data, measure how users react to the system, and collect positive or negative feedback.

[0470] 6. Collection route optimization:

[0471] The server optimizes collection routes based on waste generation prediction results and emotion data, thereby calculating efficient routes that will increase user satisfaction.

[0472] Specific examples

[0473] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation. It then uses an emotion engine to collect user feedback and takes this into account when optimizing collection routes.

[0474] In this way, the introduction of an emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[0475] The processing flow will be explained below.

[0476] Step 1:

[0477] The server reads historical data from an external database or a CSV file. For example, it retrieves information such as the amount of waste generated, collection location, collection date, weather, and population density from a file called waste_data.csv.

[0478] Step 2:

[0479] The server preprocesses the data it has loaded. Specifically, it removes or imputes missing values ​​and separates the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). This process results in a clean dataset.

[0480] Step 3:

[0481] The server trains a random forest regression model using the preprocessed data. It uses specific parameter settings (e.g., n_estimators=100, random_state=42) to train the model stably and efficiently.

[0482] Step 4:

[0483] The server uses the trained model to make predictions on a new dataset, which includes population density and weather data for days that haven't been collected. Based on these inputs, the server predicts new waste generation volumes.

[0484] Step 5:

[0485] The server uses an emotion engine to collect user emotion data. Specifically, it identifies situations where users provide feedback to the system, identifies positive or negative reactions, and records them in a database. This information is collected using methods such as text analysis and speech analysis.

[0486] Step 6:

[0487] The server analyzes the collected emotion data and optimizes the collection route based on the prediction results. Based on the emotion data, the collection route is adjusted to satisfy the user, balancing efficiency and user experience. For example, if a user expresses dissatisfaction with a previous collection route, improvements can be made based on that information.

[0488] Step 7:

[0489] The server notifies the user of the optimized collection route, including specific collection dates and times and route information. Based on this information, the user can operate the collection vehicle efficiently.

[0490] The above is a detailed explanation of each processing step for the server, terminal, and user, which allows the waste management system to achieve high prediction accuracy and user experience, while reducing environmental impact and operational costs.

[0491] Example 2

[0492] 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."

[0493] Conventional waste collection systems have low accuracy in predicting waste generation, which makes it difficult to optimize collection routes. Furthermore, adjustments were not made taking into account user feedback or emotions, preventing improvements in user satisfaction. Therefore, there was a need for efficient collection route calculations and adjustments that reflected user emotions.

[0494] 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.

[0495] In this invention, the server includes means for reading historical data, means for preprocessing the read data, means for training a predictive model using the preprocessed data, means for making predictions for a new data set using the trained predictive model, means for collecting emotion data, means for optimizing waste collection routes based on the prediction results, and means for adjusting the optimization results using the emotion data. This not only enables highly accurate prediction of waste generation patterns, but also realizes optimization of collection routes taking user emotions into consideration, enabling efficient collection operations and improved user satisfaction.

[0496] "Historical data" refers to data that includes information such as the amount of waste generated in the past, collection locations, collection dates, weather, and population density.

[0497] "Preprocessing" refers to the process of deleting missing values ​​from the loaded data and separating it into features and target variables.

[0498] A "predictive model" is a machine learning algorithm model that is trained using pre-processed data and used to predict future waste generation.

[0499] "Emotional data" is data that measures a user's emotions and collects negative or positive feedback.

[0500] "Waste collection route optimization" is the process of efficiently designing collection routes based on the results of forecasting waste generation amounts.

[0501] "Adjustment" refers to the process of rearranging collection routes optimized using emotion data to reflect user feedback.

[0502] A "regression algorithm" is a machine learning technique used to predict continuous values, and in the present system may include a random forest regression model.

[0503] A "new dataset" is data, including population density and weather data for future collection dates, that is used in the predictive model.

[0504] The present invention is a system in which a server reads historical data, pre-processes the data, and trains a predictive model to predict waste generation patterns and optimize waste collection routes. Furthermore, the server can collect sentiment data and make adjustments based on user feedback to improve the user experience. The following describes an embodiment of the present invention in detail.

[0505] First, the server reads historical data on past waste generation and collection from an external database or CSV file. This historical data includes information such as waste generation volume, collection location, collection date, weather, population density, etc. The server obtains the data from, for example, a database management system or file system.

[0506] Next, the historical data loaded by the server is preprocessed. Preprocessing includes removing or completing missing values, and separating the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). A data analysis library (e.g., Pandas) is used for data preprocessing.

[0507] The server uses the preprocessed data to train a predictive model. This predictive model is capable of learning waste generation patterns from past data and predicting future waste generation. For training, a regression algorithm from a machine learning library (e.g., Scikit-learn) is used.

[0508] The server then uses the trained model to predict waste generation for a new dataset, which includes population density and weather data for the next collection day or week, and predicts waste generation based on these inputs.

[0509] Furthermore, the server uses an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the waste collection system. The server performs emotion analysis of user feedback using, for example, a text analysis API.

[0510] The server optimizes waste collection routes based on the prediction results and emotion data. Optimization algorithms such as linear programming and genetic algorithms are used to optimize collection routes. The emotion data is also used to adjust collection routes to reflect user feedback and complaints.

[0511] Specific examples

[0512] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains a predictive model. After training, it inputs population density and weather data for the next week to predict future waste generation. It then uses an emotion engine to collect feedback from users and optimizes collection routes accordingly.

[0513] Examples of prompts include:

[0514] "Predict next week's waste generation volume based on the past year's waste generation volume data, collection locations, collection dates, weather, and population density information. Additionally, suggest optimal collection routes based on user feedback."

[0515] In this way, the introduction of an emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[0516] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0517] Step 1:

[0518] The server reads historical data. Specifically, the server retrieves data such as past waste generation volume, collection locations, collection dates, weather, and population density from an external database or the "waste_data.csv" file. The input is the "waste_data.csv" file, and the output is a dataset in Pandas DataFrame format.

[0519] Step 2:

[0520] The server preprocesses the historical data. Specifically, the server removes or imputes missing values ​​and separates the data into the features and target variables required for prediction. The input is a DataFrame of historical data, and the output is the preprocessed feature data and target variable data in DataFrame format.

[0521] Step 3:

[0522] The server trains a predictive model. Specifically, the server trains a machine learning model using the preprocessed data. It uses Scikit-learn's random forest regression model. The input is the preprocessed feature data and the objective variable data, and the output is a trained predictive model.

[0523] Step 4:

[0524] The server makes a prediction based on the new dataset. Specifically, the server uses a predictive model to predict waste generation volume using population density and weather data for the next collection day as input. The input is the new dataset, and the output is the predicted waste generation volume.

[0525] Step 5:

[0526] The server collects emotion data. Specifically, it collects feedback from users via their devices and performs emotion analysis using an emotion engine. The input is the user's feedback (in text format), and the output is the emotion analysis result (positive, negative, etc.).

[0527] Step 6:

[0528] The server optimizes collection routes using emotion data. Specifically, the server calculates the optimal waste collection route based on the prediction results and emotion data. Linear programming and genetic algorithms are used for optimization. The inputs are the predicted waste generation volume and emotion data, and the output is the optimized collection route.

[0529] The above are the specific processing steps of this system's program. The input and output at each step, as well as the specific data processing and data calculation, have been explained, and the overall flow of the system has been clarified.

[0530] (Application example 2)

[0531] 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."

[0532] In modern factories, efficient waste management and the development of optimal collection routes are important challenges. In particular, irregular waste generation patterns and low user (factory manager) satisfaction with collection services pose concerns about increased operational costs and environmental impact. Furthermore, adjusting collection routes based on user feedback is likely to contribute to further efficiency and improved satisfaction. Given this background, existing waste management systems are required to be equipped with highly accurate waste generation predictions and the ability to collect and reflect user sentiment data.

[0533] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for reading history data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for recognizing and collecting user emotion data, and means for adjusting collection routes based on the user emotion data. This not only improves the accuracy of waste generation predictions and provides optimized collection routes, but also enables efficient waste management with high user satisfaction by adjusting the routes based on user emotions.

[0534] "Historical data" means information relating to past waste generation and collection, including data such as waste generation volume, collection locations, collection dates, weather, and population density.

[0535] "Preprocessing" is a data processing method that removes missing values ​​and separates features and target variables in order to prepare the data in a format suitable for machine learning model training.

[0536] A "machine learning model" is an algorithm that uses given data to learn patterns and make predictions or classifications for future data.

[0537] A "random forest regression model" is a machine learning model that makes predictions by combining multiple decision trees, and is particularly effective when applied to regression problems, enabling highly accurate predictions.

[0538] An "emotion engine" is a system that recognizes users' emotions, collects and analyzes that data, and is used to understand users' feedback and emotional state.

[0539] "Collection route optimization" is a means of minimizing costs and time by adjusting collection routes based on prediction results and user emotion data in order to achieve efficient waste collection.

[0540] "User emotion data" refers to data collected by the emotion engine regarding the user's emotional state, including positive or negative feedback.

[0541] A "new dataset" is new input data, such as population density or weather on an uncollected day, that the trained model will use to make predictions.

[0542] "Missing values" are incomplete or missing data in a dataset that should be appropriately handled to improve the accuracy of the model.

[0543] This invention is a system for streamlining waste management within factories. The system utilizes historical data and machine learning models to predict waste generation patterns and optimize collection routes. Furthermore, by collecting user sentiment data, the system can be applied to adjusting collection routes.

[0544] Hardware and software used

[0545] Hardware: Factory robots, servers

[0546] Software: Python, TensorFlow, OpenCV (for emotion recognition), pandas (data preprocessing), scikit-learn (machine learning)

[0547] System program processing explanation

[0548] 1. Data loading

[0549] The server reads data on historical waste generation and collection from an external database or a CSV file, including information on waste generation volume, collection location, collection date, weather, population density, etc. For example, it retrieves data from a file called "waste_data.csv."

[0550] 2. Data Preprocessing

[0551] The server performs data preprocessing. In this step, missing values ​​are removed and features and target variables are separated. Specifically, features such as population density and weather are separated from target variables such as waste generation volume in order to input them into the machine learning model. The Python pandas library is used to format the data and process missing values.

[0552] 3. Model training

[0553] The server uses the preprocessed data to train a random forest regression model. Using the scikit-learn library, the model is built using the training data, enabling highly accurate predictions of waste generation.

[0554] 4. Prediction

[0555] The server uses the trained model to predict waste generation for a new dataset, such as population density or weather data for a specific day, and uses Python and a scikit-learn random forest regression model to make predictions based on this data.

[0556] 5. Collecting Emotional Data

[0557] The emotion engine collects emotional data from users such as factory managers. Specifically, it combines a camera using OpenCV with a Python script to recognize emotions from the user's facial expressions and analyzes positive or negative feedback.

[0558] 6. Optimizing collection routes

[0559] The server optimizes collection routes based on waste generation prediction results and user emotion data. This optimization process takes into account information obtained from emotion data and makes adjustments to increase user satisfaction. For example, if a user expresses negative emotions, the server takes measures such as changing the collection route or collection frequency.

[0560] Examples of concrete examples and prompts

[0561] For example, at a factory in a certain city, the server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. It then uses an emotion engine to collect feedback from the factory manager and takes this into consideration to optimize collection routes. An example of a prompt would be, "Predict the amount of waste generated within the factory and calculate the optimal collection route based on the user (manager) feedback."

[0562] In this way, the present invention simultaneously realizes efficient waste management and improved user satisfaction.

[0563] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0564] Step 1:

[0565] The server reads historical data about past waste generation and collection from a central database or CSV files. This input data includes waste generation volume, collection location, collection date, weather, population density, and other parameters. The obtained data frame is generated as output.

[0566] Step 2:

[0567] The server preprocesses the loaded data by removing missing values ​​and separating features (e.g., population density, weather) from the target variable (waste generation volume). The server uses the pandas library to format the data and process missing values, and outputs the preprocessed dataset.

[0568] Step 3:

[0569] The server trains a random forest regression model using the preprocessed dataset. Using the Scikit-learn library, it builds a model from the training data and gives it the ability to predict waste generation. The output is a trained machine learning model.

[0570] Step 4:

[0571] The server uses the trained model to predict waste generation for a new dataset, which may include population density and weather information for a specific day. By inputting this new data into the trained model, a waste generation prediction result is obtained. The output is a predicted value data.

[0572] Step 5:

[0573] Using the emotion engine, the device collects the user's emotion data. Specifically, it uses a camera and the OpenCV library to analyze the user's face and recognize their emotional status, such as positive or negative. The detection results are output as emotion data.

[0574] Step 6:

[0575] The server optimizes waste collection routes based on the prediction results and the user's emotional data. If the collected emotional data indicates negative feedback, it adjusts the collection route or collection frequency. For example, it recalculates the collection route to increase user satisfaction or changes the collection frequency. This optimized collection route information is output.

[0576] The above are the specific processing steps of the system that realizes the application example. The specific operations performed at each step realize efficient waste management and improved user satisfaction.

[0577] 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.

[0578] 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.

[0579] 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.

[0580] [Third embodiment]

[0581] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0582] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0583] 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).

[0584] 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.

[0585] 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.

[0586] 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).

[0587] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0588] 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.

[0589] 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.

[0590] 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.

[0591] 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.

[0592] 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."

[0593] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. The following describes in detail an embodiment of the present invention.

[0594] System configuration

[0595] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[0596] Next, the server preprocesses the loaded historical data, which includes the following steps:

[0597] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0598] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[0599] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation patterns.

[0600] The server then uses the trained model to make predictions on a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[0601] Finally, the server optimizes waste collection routes based on the prediction results. The optimization algorithm sorts collection points in descending order of waste volume and calculates efficient collection routes.

[0602] Program processing flow

[0603] 1. Data loading:

[0604] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[0605] 2. Data Preprocessing:

[0606] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[0607] 3. Model training:

[0608] The server trains a random forest regression model using the preprocessed data.

[0609] 4. Prediction:

[0610] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[0611] 5. Collection route optimization:

[0612] The server optimizes collection routes based on the prediction results, which are determined by sorting collection points in descending order of waste volume.

[0613] Specific examples

[0614] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0615] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0616] The processing flow will be explained below.

[0617] Step 1:

[0618] The server reads historical data from an external database or a CSV file, such as waste_data.csv, which contains information such as the amount of waste generated, collection locations, collection dates, weather, and population density.

[0619] Step 2:

[0620] The server preprocesses the data it loads. First, it removes or imputes missing values. Then it reshapes the data to separate it into features (e.g., population density, weather) and target variables (waste generation volume). This results in a clean dataset for use in machine learning models.

[0621] Step 3:

[0622] The server trains a random forest regression model using the preprocessed data. Special settings, such as n_estimators=100 and random_state=42, are used to ensure model stability. Once trained, the model is used for future predictions.

[0623] Step 4:

[0624] The server makes predictions using a new dataset, which includes population density and weather data for days that have not yet been collected, and uses the trained random forest regression model to predict waste generation for these new data points.

[0625] Step 5:

[0626] The server optimizes collection routes based on the predicted waste generation volume. Specifically, it sorts collection points in descending order of waste volume based on the prediction results and calculates the most efficient collection route. This improves waste collection efficiency and reduces operation costs.

[0627] Step 6:

[0628] The system notifies the user of the results of the collection route. Based on the notified data, the user can appropriately deploy collection vehicles and begin collection work. The user can then efficiently collect waste by following the presented optimized route.

[0629] Through these steps, the system can improve the accuracy of waste generation predictions and optimize collection routes, thereby reducing environmental impact and operational costs.

[0630] Example 1

[0631] 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."

[0632] Conventional waste collection systems have difficulty predicting waste generation volumes, leading to insufficient optimization of collection routes. It is particularly difficult to effectively incorporate external factors, such as population density and weather, that affect waste generation volumes. As a result, efficient operation of collection vehicles is hindered, resulting in increased operating costs and adverse environmental impacts.

[0633] 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.

[0634] In this invention, the server further includes means for reading historical data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for the historical data including information such as waste generation amount, collection location, collection date, weather, and population density, means for deleting or completing missing values, and means for separating features and target variables. This enables highly accurate prediction of waste generation amount and calculation of efficient collection routes based on the predictions.

[0635] "Historical data" is data that includes information about past waste generation and collection, such as waste generation volume, collection locations, collection dates, weather, population density, and the like.

[0636] "Preprocessing" refers to the process of removing or complementing missing values ​​in a dataset and separating the features and objective variables required for prediction.

[0637] A "machine learning model" is a model that learns patterns and rules from input data and makes predictions about new data.

[0638] A "random forest regression model" is a machine learning technique that uses a large number of decision trees to improve prediction accuracy.

[0639] "Prediction" is the act of using a trained machine learning model to estimate waste generation for a new dataset.

[0640] "Waste collection route optimization" refers to the calculation of the optimal route for efficiently visiting collection points based on the predicted amount of waste generated.

[0641] A "new dataset" is data that includes information such as population density and weather for future collection dates.

[0642] "Missing values" are values ​​that are missing in a dataset and must be handled appropriately.

[0643] "Features" are attributes or variables of input data that a machine learning model uses to make predictions.

[0644] A "target variable" is an attribute or variable of the output data that a machine learning model is intended to predict.

[0645] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. Specifically, it consists of the following steps:

[0646] First, the server loads historical data on waste generation and collection from an external database or a CSV file. This historical data includes information such as the amount of waste generated, collection locations, collection dates, weather, and population density. For example, data can be loaded from a file called "waste_data.csv."

[0647] Next, the historical data loaded by the server is preprocessed. Preprocessing includes removing missing values ​​and separating features and target variables. Missing values ​​are values ​​that are missing from a dataset, and are handled appropriately using methods such as mean value imputation and data deletion. Features refer to the attributes and variables of data required for prediction, such as population density and weather, while target variables refer to the data to be predicted, such as waste generation volume.

[0648] Using the preprocessed data, the server trains a random forest regression model, a machine learning technique that uses a large number of decision trees to improve prediction accuracy. The server uses this random forest model to learn from past data and generate a model capable of predicting future waste generation patterns.

[0649] The trained model is then used to make predictions on a new dataset, which includes population density and weather data for a specific collection day. The user enters this information, and the server uses it to predict future waste generation. For example, to create a collection plan for a new week, the generative AI model can be given the following prompt:

[0650] "Predict waste generation based on population density and weather data for the new week, and optimize waste collection routes accordingly."

[0651] Finally, the server optimizes waste collection routes based on the prediction results. Using an optimization algorithm, collection points are sorted in descending order of waste volume, and an efficient collection route is calculated. This collection route information is output to a terminal and can be checked by the user. It is also reflected in collection vehicle schedules, improving operational efficiency.

[0652] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0653] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0654] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0655] Step 1:

[0656] Data loading

[0657] The server reads historical data from an external database or a CSV file (e.g., "waste_data.csv"). The input data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. The server stores this data in memory and prepares it for subsequent preprocessing steps. Specifically, the server specifies the file path and opens it to read the data.

[0658] Step 2:

[0659] Data preprocessing (missing value handling)

[0660] The server detects missing values ​​in the dataset loaded and performs deletion or imputation processing. Historical data is given as input. For example, if there are missing values ​​in weather data, they are imputed using historical average weather data. The output is a clean dataset with the missing values ​​processed. Specifically, it scans the dataset and processes missing values ​​in each column using the appropriate method.

[0661] Step 3:

[0662] Data preprocessing (separation of features and target variables)

[0663] The server uses a clean dataset to separate the features and target variables required for prediction. The input is a dataset with missing values ​​processed. The features include population density and weather data, and the target variables include waste generation volume. The output is a dataset separated into the features and target variables. Specifically, the data is separated based on each column.

[0664] Step 4:

[0665] Model training

[0666] The server trains a random forest regression model using the preprocessed data. The input is the separated features and the target variable. The server applies the random forest regression algorithm to generate and combine multiple decision trees. The output is a trained random forest model. Specifically, the data is shuffled to separate it into a training set and a validation set, and the model is optimized through cross-validation.

[0667] Step 5:

[0668] Entering New Data

[0669] The user inputs population density and weather data for a new collection date. The input is weather data and demographic information for a new week. The output is data formatted in a way that the server can use for predictions. Specifically, the user inputs new data through a web interface, and the server receives and preprocesses the data.

[0670] Step 6:

[0671] Model prediction

[0672] The server uses the new dataset to make predictions using the trained model. The input is the feature data for the new collection date. The output is a prediction of future waste generation volume. Specifically, the server inputs the preprocessed new data into the model and calculates the predicted value.

[0673] Step 7:

[0674] Collection route optimization

[0675] The server optimizes waste collection routes based on the prediction results. The input is the predicted waste generation data. The server uses an optimization algorithm to sort collection points in descending order of waste volume and calculates an efficient collection route. The output is optimized collection route information. Specifically, the algorithm is run based on the prediction results, and the collection route is calculated and visualized.

[0676] Step 8:

[0677] Output of results

[0678] The server outputs the optimized collection route to the terminal and displays it to the user. The optimized route information is given as input. The output is collection route information in a display format that can be confirmed by the user. Specifically, the calculation results are generated in table or map format and sent to the terminal.

[0679] (Application example 1)

[0680] 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."

[0681] Conventional waste collection systems lack the means to predict waste generation patterns and design efficient routes, resulting in poor collection efficiency. Similarly, in the supply of parts within factories, there is a lack of a means to design efficient supply routes based on past consumption data, resulting in wasted time and money in parts supply. To address these issues, the present invention aims to optimize waste collection and in-factory parts supply routes by applying preprocessing of historical data and machine learning models.

[0682] 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.

[0683] In this invention, the server includes a means for reading historical data, a means for preprocessing the read data, a means for training a machine learning model using the preprocessed data, a means for making predictions for a new data set using the trained model, a means for optimizing waste collection routes based on the prediction results, and a means for optimizing parts supply routes within the factory by applying a similar system, thereby enabling the efficiency of not only waste collection but also parts supply within the factory.

[0684] "Historical data" refers to recorded data about past events or activities, and in this context refers to waste generation and parts consumption data.

[0685] "Preprocessing" refers to the process of correcting or supplementing incomplete data and preparing it in an appropriate format prior to data analysis or training a machine learning model.

[0686] A "machine learning model" is a set of mathematical algorithms that learn patterns from past data and make predictions or classifications for new data.

[0687] "Training" is the process of using historical data to train a machine learning model and improve its accuracy.

[0688] "Prediction" is the act of using existing data and a trained machine learning model to estimate the value of future data points or events.

[0689] "Optimization" is the process of finding the most efficient use of available resources based on specific objectives and constraints.

[0690] "Waste collection route" refers to the specific route or sequence used to collect waste.

[0691] "Parts supply route" refers to the route and order for efficiently delivering parts within a factory to each workplace and process.

[0692] "Dataset" refers to a collection of data used for a particular calculation or analysis.

[0693] A "random forest regression model" is a type of machine learning algorithm that uses multiple decision trees to perform regression analysis.

[0694] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection and in-factory parts supply routes. Hereinafter, an embodiment of the present invention will be described in detail.

[0695] System configuration

[0696] First, the server loads data on past waste generation and collection (historical data) from an external database or CSV file. Similarly, it loads data on in-factory parts consumption. This historical data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. In-factory parts consumption data includes the amount of parts consumed, consumption location, consumption date, and work progress status in the factory.

[0697] Next, the server preprocesses the loaded history data, which includes the following steps:

[0698] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0699] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather, factory work progress) from the target variables (waste generation volume, parts consumption volume).

[0700] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation and component consumption patterns.

[0701] The server then uses the trained model to make predictions on a new dataset, which includes population density, weather data, and factory work progress data for a specific collection or consumption date, and predicts waste generation and parts consumption based on these inputs.

[0702] Finally, the server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. The optimization algorithm sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes.

[0703] Hardware and Software Configuration

[0704] This system uses the following hardware and software:

[0705] Hardware: Servers, factory robots, smart glasses

[0706] Software: Python, Pandas, Scikit-learn

[0707] The server loads historical data and performs data preprocessing using Pandas. It then uses Scikit-learn to train a random forest regression model to make predictions based on new data sets. The optimized collection and delivery routes are displayed or instructed to factory robots or smart glasses.

[0708] Specific examples

[0709] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0710] Within the factory, parts consumption data from the past year is used to predict parts consumption based on the new week's data (population density and factory work progress).Based on this prediction, robots within the factory operate optimal supply routes, aiming to achieve efficient parts supply.

[0711] Example prompt sentence:

[0712] "Generate Python code that uses the past year's in-factory parts consumption data (e.g., CSV file) to predict parts consumption based on the new week's data (population density and weather). Include code that optimizes parts supply routes within the factory based on the predictions."

[0713] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0714] Step 1:

[0715] The server reads the historical data. Specifically, the server retrieves data on past waste generation and parts consumption from an external database or a CSV file (e.g., "factory_parts_data.csv"). The input is the historical data, and the output is the retrieved dataset.

[0716] Step 2:

[0717] The data loaded by the server is preprocessed. Specifically, missing values ​​are removed and the features required for prediction (e.g., population density, weather, factory work progress) are separated from the target variables (waste generation volume, parts consumption volume). The input is the acquired dataset, and the output is the preprocessed dataset.

[0718] Step 3:

[0719] The server uses the preprocessed data to train a machine learning model. Specifically, it uses a random forest regression model to learn patterns of waste generation and component consumption from historical data. The input is the preprocessed dataset, and the output is the trained model.

[0720] Step 4:

[0721] The server uses the trained model to make predictions for a new data set. Specifically, it predicts the amount of waste generated and parts consumed using inputs such as population density, weather data, and factory work progress data for a new collection or consumption date. The input is the new data set, and the output is the prediction result.

[0722] Step 5:

[0723] The server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. Specifically, it sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes. The input is the prediction results, and the output is optimized route information.

[0724] Step 6:

[0725] The server sends the optimized waste collection route and parts supply route to related terminals and devices (e.g., factory robots, smart glasses). Specifically, the route information is displayed or instructed on a display device or control system. The input is the optimized route information, and the output is a confirmation of transmission to the terminal.

[0726] 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.

[0727] The present invention is a system in which a server reads historical data, preprocesses the data, and trains a machine learning model to predict waste generation patterns and optimize waste collection routes. Furthermore, by combining it with an emotion engine that recognizes user emotions, the user experience can be improved. The following describes embodiments of the present invention in detail.

[0728] System configuration

[0729] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[0730] Next, the server preprocesses the loaded historical data, which includes the following steps:

[0731] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0732] 2. Separation of features and target variables: The data is shaped to separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[0733] The server uses the preprocessed data to train a random forest regression model, which is capable of learning waste generation patterns from past data and predicting future waste generation.

[0734] The server then uses the trained model to make waste generation predictions for a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[0735] Furthermore, we use an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the collected results and system suggestions.

[0736] The server optimizes collection routes based on the waste generation prediction results and user emotion data. Based on the information obtained from the emotion engine, adjustments are made taking complaints and feedback into consideration, and a collection route that satisfies the user is calculated.

[0737] Program processing flow

[0738] 1. Data loading:

[0739] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[0740] 2. Data Preprocessing:

[0741] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[0742] 3. Model training:

[0743] The server trains a random forest regression model using the preprocessed data.

[0744] 4. Prediction:

[0745] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[0746] 5. Collecting Emotional Data:

[0747] The server uses an emotion engine to collect user emotion data, measure how users react to the system, and collect positive or negative feedback.

[0748] 6. Collection route optimization:

[0749] The server optimizes collection routes based on waste generation prediction results and emotion data, thereby calculating efficient routes that will increase user satisfaction.

[0750] Specific examples

[0751] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation. It then uses an emotion engine to collect user feedback and takes this into account when optimizing collection routes.

[0752] In this way, the introduction of an emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[0753] The processing flow will be explained below.

[0754] Step 1:

[0755] The server reads historical data from an external database or a CSV file. For example, it retrieves information such as the amount of waste generated, collection location, collection date, weather, and population density from a file called waste_data.csv.

[0756] Step 2:

[0757] The server preprocesses the data it has loaded. Specifically, it removes or imputes missing values ​​and separates the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). This process results in a clean dataset.

[0758] Step 3:

[0759] The server trains a random forest regression model using the preprocessed data. It uses specific parameter settings (e.g., n_estimators=100, random_state=42) to train the model stably and efficiently.

[0760] Step 4:

[0761] The server uses the trained model to make predictions on a new dataset, which includes population density and weather data for days that haven't been collected. Based on these inputs, the server predicts new waste generation volumes.

[0762] Step 5:

[0763] The server uses an emotion engine to collect user emotion data. Specifically, it identifies situations where users provide feedback to the system, identifies positive or negative reactions, and records them in a database. This information is collected using methods such as text analysis and speech analysis.

[0764] Step 6:

[0765] The server analyzes the collected emotion data and optimizes the collection route based on the prediction results. Based on the emotion data, the collection route is adjusted to satisfy the user, balancing efficiency and user experience. For example, if a user expresses dissatisfaction with a previous collection route, improvements can be made based on that information.

[0766] Step 7:

[0767] The server notifies the user of the optimized collection route, including specific collection dates and times and route information. Based on this information, the user can operate the collection vehicle efficiently.

[0768] The above is a detailed explanation of each processing step for the server, terminal, and user, which allows the waste management system to achieve high prediction accuracy and user experience, while reducing environmental impact and operational costs.

[0769] Example 2

[0770] 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."

[0771] Conventional waste collection systems have low accuracy in predicting waste generation, which makes it difficult to optimize collection routes. Furthermore, adjustments were not made taking into account user feedback or emotions, preventing improvements in user satisfaction. Therefore, there was a need for efficient collection route calculations and adjustments that reflected user emotions.

[0772] 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.

[0773] In this invention, the server includes means for reading historical data, means for preprocessing the read data, means for training a predictive model using the preprocessed data, means for making predictions for a new data set using the trained predictive model, means for collecting emotion data, means for optimizing waste collection routes based on the prediction results, and means for adjusting the optimization results using the emotion data. This not only enables highly accurate prediction of waste generation patterns, but also realizes optimization of collection routes taking user emotions into consideration, enabling efficient collection operations and improved user satisfaction.

[0774] "Historical data" refers to data that includes information such as the amount of waste generated in the past, collection locations, collection dates, weather, and population density.

[0775] "Preprocessing" refers to the process of deleting missing values ​​from the loaded data and separating it into features and target variables.

[0776] A "predictive model" is a machine learning algorithm model that is trained using pre-processed data and used to predict future waste generation.

[0777] "Emotional data" is data that measures a user's emotions and collects negative or positive feedback.

[0778] "Waste collection route optimization" is the process of efficiently designing collection routes based on the results of forecasting waste generation amounts.

[0779] "Adjustment" refers to the process of rearranging collection routes optimized using emotion data to reflect user feedback.

[0780] A "regression algorithm" is a machine learning technique used to predict continuous values, and in the present system may include a random forest regression model.

[0781] A "new dataset" is data, including population density and weather data for future collection dates, that is used in the predictive model.

[0782] The present invention is a system in which a server reads historical data, pre-processes the data, and trains a predictive model to predict waste generation patterns and optimize waste collection routes. Furthermore, the server can collect sentiment data and make adjustments based on user feedback to improve the user experience. The following describes an embodiment of the present invention in detail.

[0783] First, the server reads historical data on past waste generation and collection from an external database or CSV file. This historical data includes information such as waste generation volume, collection location, collection date, weather, population density, etc. The server obtains the data from, for example, a database management system or file system.

[0784] Next, the historical data loaded by the server is preprocessed. Preprocessing includes removing or completing missing values, and separating the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). A data analysis library (e.g., Pandas) is used for data preprocessing.

[0785] The server uses the preprocessed data to train a predictive model. This predictive model is capable of learning waste generation patterns from past data and predicting future waste generation. For training, a regression algorithm from a machine learning library (e.g., Scikit-learn) is used.

[0786] The server then uses the trained model to predict waste generation for a new dataset, which includes population density and weather data for the next collection day or week, and predicts waste generation based on these inputs.

[0787] Furthermore, the server uses an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the waste collection system. The server performs emotion analysis of user feedback using, for example, a text analysis API.

[0788] The server optimizes waste collection routes based on the prediction results and emotion data. Optimization algorithms such as linear programming and genetic algorithms are used to optimize collection routes. The emotion data is also used to adjust collection routes to reflect user feedback and complaints.

[0789] Specific examples

[0790] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains a predictive model. After training, it inputs population density and weather data for the next week to predict future waste generation. It then uses an emotion engine to collect feedback from users and optimizes collection routes accordingly.

[0791] Examples of prompts include:

[0792] "Predict next week's waste generation volume based on the past year's waste generation volume data, collection locations, collection dates, weather, and population density information. Additionally, suggest optimal collection routes based on user feedback."

[0793] In this way, the introduction of an emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[0794] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0795] Step 1:

[0796] The server reads historical data. Specifically, the server retrieves data such as past waste generation volume, collection locations, collection dates, weather, and population density from an external database or the "waste_data.csv" file. The input is the "waste_data.csv" file, and the output is a dataset in Pandas DataFrame format.

[0797] Step 2:

[0798] The server preprocesses the historical data. Specifically, the server removes or imputes missing values ​​and separates the data into the features and target variables required for prediction. The input is a DataFrame of historical data, and the output is the preprocessed feature data and target variable data in DataFrame format.

[0799] Step 3:

[0800] The server trains a predictive model. Specifically, the server trains a machine learning model using the preprocessed data. It uses Scikit-learn's random forest regression model. The input is the preprocessed feature data and the objective variable data, and the output is a trained predictive model.

[0801] Step 4:

[0802] The server makes a prediction based on the new dataset. Specifically, the server uses a predictive model to predict waste generation volume using population density and weather data for the next collection day as input. The input is the new dataset, and the output is the predicted waste generation volume.

[0803] Step 5:

[0804] The server collects emotion data. Specifically, it collects feedback from users via their devices and performs emotion analysis using an emotion engine. The input is the user's feedback (in text format), and the output is the emotion analysis result (positive, negative, etc.).

[0805] Step 6:

[0806] The server optimizes collection routes using emotion data. Specifically, the server calculates the optimal waste collection route based on the prediction results and emotion data. Linear programming and genetic algorithms are used for optimization. The inputs are the predicted waste generation volume and emotion data, and the output is the optimized collection route.

[0807] The above are the specific processing steps of this system's program. The input and output at each step, as well as the specific data processing and data calculation, have been explained, and the overall flow of the system has been clarified.

[0808] (Application example 2)

[0809] 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."

[0810] In modern factories, efficient waste management and the development of optimal collection routes are important challenges. In particular, irregular waste generation patterns and low user (factory manager) satisfaction with collection services pose concerns about increased operational costs and environmental impact. Furthermore, adjusting collection routes based on user feedback is likely to contribute to further efficiency and improved satisfaction. Given this background, existing waste management systems are required to be equipped with highly accurate waste generation predictions and the ability to collect and reflect user sentiment data.

[0811] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for reading history data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for recognizing and collecting user emotion data, and means for adjusting collection routes based on the user emotion data. This not only improves the accuracy of waste generation predictions and provides optimized collection routes, but also enables efficient waste management with high user satisfaction by adjusting the routes based on user emotions.

[0812] "Historical data" means information relating to past waste generation and collection, including data such as waste generation volume, collection locations, collection dates, weather, and population density.

[0813] "Preprocessing" is a data processing method that removes missing values ​​and separates features and target variables in order to prepare the data in a format suitable for machine learning model training.

[0814] A "machine learning model" is an algorithm that uses given data to learn patterns and make predictions or classifications for future data.

[0815] A "random forest regression model" is a machine learning model that makes predictions by combining multiple decision trees, and is particularly effective when applied to regression problems, enabling highly accurate predictions.

[0816] An "emotion engine" is a system that recognizes users' emotions, collects and analyzes that data, and is used to understand users' feedback and emotional state.

[0817] "Collection route optimization" is a means of minimizing costs and time by adjusting collection routes based on prediction results and user emotion data in order to achieve efficient waste collection.

[0818] "User emotion data" refers to data collected by the emotion engine regarding the user's emotional state, including positive or negative feedback.

[0819] A "new dataset" is new input data, such as population density or weather on an uncollected day, that the trained model will use to make predictions.

[0820] "Missing values" are incomplete or missing data in a dataset that should be appropriately handled to improve the accuracy of the model.

[0821] This invention is a system for streamlining waste management within factories. The system utilizes historical data and machine learning models to predict waste generation patterns and optimize collection routes. Furthermore, by collecting user sentiment data, the system can be applied to adjusting collection routes.

[0822] Hardware and software used

[0823] Hardware: Factory robots, servers

[0824] Software: Python, TensorFlow, OpenCV (for emotion recognition), pandas (data preprocessing), scikit-learn (machine learning)

[0825] System program processing explanation

[0826] 1. Data loading

[0827] The server reads data on historical waste generation and collection from an external database or a CSV file, including information on waste generation volume, collection location, collection date, weather, population density, etc. For example, it retrieves data from a file called "waste_data.csv."

[0828] 2. Data Preprocessing

[0829] The server performs data preprocessing. In this step, missing values ​​are removed and features and target variables are separated. Specifically, features such as population density and weather are separated from target variables such as waste generation volume in order to input them into the machine learning model. The Python pandas library is used to format the data and process missing values.

[0830] 3. Model training

[0831] The server uses the preprocessed data to train a random forest regression model. Using the scikit-learn library, the model is built using the training data, enabling highly accurate predictions of waste generation.

[0832] 4. Prediction

[0833] The server uses the trained model to predict waste generation for a new dataset, such as population density or weather data for a specific day, and uses Python and a scikit-learn random forest regression model to make predictions based on this data.

[0834] 5. Collecting Emotional Data

[0835] The emotion engine collects emotional data from users such as factory managers. Specifically, it combines a camera using OpenCV with a Python script to recognize emotions from the user's facial expressions and analyzes positive or negative feedback.

[0836] 6. Optimizing collection routes

[0837] The server optimizes collection routes based on waste generation prediction results and user emotion data. This optimization process takes into account information obtained from emotion data and makes adjustments to increase user satisfaction. For example, if a user expresses negative emotions, the server takes measures such as changing the collection route or collection frequency.

[0838] Examples of concrete examples and prompts

[0839] For example, at a factory in a certain city, the server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. It then uses an emotion engine to collect feedback from the factory manager and takes this into consideration to optimize collection routes. An example of a prompt would be, "Predict the amount of waste generated within the factory and calculate the optimal collection route based on the user (manager) feedback."

[0840] In this way, the present invention simultaneously realizes efficient waste management and improved user satisfaction.

[0841] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0842] Step 1:

[0843] The server reads historical data about past waste generation and collection from a central database or CSV files. This input data includes waste generation volume, collection location, collection date, weather, population density, and other parameters. The obtained data frame is generated as output.

[0844] Step 2:

[0845] The server preprocesses the loaded data by removing missing values ​​and separating features (e.g., population density, weather) from the target variable (waste generation volume). The server uses the pandas library to format the data and process missing values, and outputs the preprocessed dataset.

[0846] Step 3:

[0847] The server trains a random forest regression model using the preprocessed dataset. Using the Scikit-learn library, it builds a model from the training data and gives it the ability to predict waste generation. The output is a trained machine learning model.

[0848] Step 4:

[0849] The server uses the trained model to predict waste generation for a new dataset, which may include population density and weather information for a specific day. By inputting this new data into the trained model, a waste generation prediction result is obtained. The output is a predicted value data.

[0850] Step 5:

[0851] Using the emotion engine, the device collects the user's emotion data. Specifically, it uses a camera and the OpenCV library to analyze the user's face and recognize their emotional status, such as positive or negative. The detection results are output as emotion data.

[0852] Step 6:

[0853] The server optimizes waste collection routes based on the prediction results and the user's emotional data. If the collected emotional data indicates negative feedback, it adjusts the collection route or collection frequency. For example, it recalculates the collection route to increase user satisfaction or changes the collection frequency. This optimized collection route information is output.

[0854] The above are the specific processing steps of the system that realizes the application example. The specific operations performed at each step realize efficient waste management and improved user satisfaction.

[0855] 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.

[0856] 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.

[0857] 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.

[0858] [Fourth embodiment]

[0859] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0860] 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.

[0861] 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).

[0862] 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.

[0863] 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.

[0864] 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).

[0865] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0866] 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.

[0867] 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.

[0868] 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.

[0869] 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.

[0870] 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.

[0871] 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."

[0872] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. The following describes in detail an embodiment of the present invention.

[0873] System configuration

[0874] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[0875] Next, the server preprocesses the loaded historical data, which includes the following steps:

[0876] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0877] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[0878] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation patterns.

[0879] The server then uses the trained model to make predictions on a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[0880] Finally, the server optimizes waste collection routes based on the prediction results. The optimization algorithm sorts collection points in descending order of waste volume and calculates efficient collection routes.

[0881] Program processing flow

[0882] 1. Data loading:

[0883] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[0884] 2. Data Preprocessing:

[0885] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[0886] 3. Model training:

[0887] The server trains a random forest regression model using the preprocessed data.

[0888] 4. Prediction:

[0889] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[0890] 5. Collection route optimization:

[0891] The server optimizes collection routes based on the prediction results, which are determined by sorting collection points in descending order of waste volume.

[0892] Specific examples

[0893] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0894] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0895] The processing flow will be explained below.

[0896] Step 1:

[0897] The server reads historical data from an external database or a CSV file, such as waste_data.csv, which contains information such as the amount of waste generated, collection locations, collection dates, weather, and population density.

[0898] Step 2:

[0899] The server preprocesses the data it loads. First, it removes or imputes missing values. Then it reshapes the data to separate it into features (e.g., population density, weather) and target variables (waste generation volume). This results in a clean dataset for use in machine learning models.

[0900] Step 3:

[0901] The server trains a random forest regression model using the preprocessed data. Special settings, such as n_estimators=100 and random_state=42, are used to ensure model stability. Once trained, the model is used for future predictions.

[0902] Step 4:

[0903] The server makes predictions using a new dataset, which includes population density and weather data for days that have not yet been collected, and uses the trained random forest regression model to predict waste generation for these new data points.

[0904] Step 5:

[0905] The server optimizes collection routes based on the predicted waste generation volume. Specifically, it sorts collection points in descending order of waste volume based on the prediction results and calculates the most efficient collection route. This improves waste collection efficiency and reduces operation costs.

[0906] Step 6:

[0907] The system notifies the user of the results of the collection route. Based on the notified data, the user can appropriately deploy collection vehicles and begin collection work. The user can then efficiently collect waste by following the presented optimized route.

[0908] Through these steps, the system can improve the accuracy of waste generation predictions and optimize collection routes, thereby reducing environmental impact and operational costs.

[0909] Example 1

[0910] 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."

[0911] Conventional waste collection systems have difficulty predicting waste generation volumes, leading to insufficient optimization of collection routes. It is particularly difficult to effectively incorporate external factors, such as population density and weather, that affect waste generation volumes. As a result, efficient operation of collection vehicles is hindered, resulting in increased operating costs and adverse environmental impacts.

[0912] 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.

[0913] In this invention, the server further includes means for reading historical data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for the historical data including information such as waste generation amount, collection location, collection date, weather, and population density, means for deleting or completing missing values, and means for separating features and target variables. This enables highly accurate prediction of waste generation amount and calculation of efficient collection routes based on the predictions.

[0914] "Historical data" is data that includes information about past waste generation and collection, such as waste generation volume, collection locations, collection dates, weather, population density, and the like.

[0915] "Preprocessing" refers to the process of removing or complementing missing values ​​in a dataset and separating the features and objective variables required for prediction.

[0916] A "machine learning model" is a model that learns patterns and rules from input data and makes predictions about new data.

[0917] A "random forest regression model" is a machine learning technique that uses a large number of decision trees to improve prediction accuracy.

[0918] "Prediction" is the act of using a trained machine learning model to estimate waste generation for a new dataset.

[0919] "Waste collection route optimization" refers to the calculation of the optimal route for efficiently visiting collection points based on the predicted amount of waste generated.

[0920] A "new dataset" is data that includes information such as population density and weather for future collection dates.

[0921] "Missing values" are values ​​that are missing in a dataset and must be handled appropriately.

[0922] "Features" are attributes or variables of input data that a machine learning model uses to make predictions.

[0923] A "target variable" is an attribute or variable of the output data that a machine learning model is intended to predict.

[0924] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection routes. Specifically, it consists of the following steps:

[0925] First, the server loads historical data on waste generation and collection from an external database or a CSV file. This historical data includes information such as the amount of waste generated, collection locations, collection dates, weather, and population density. For example, data can be loaded from a file called "waste_data.csv."

[0926] Next, the historical data loaded by the server is preprocessed. Preprocessing includes removing missing values ​​and separating features and target variables. Missing values ​​are values ​​that are missing from a dataset, and are handled appropriately using methods such as mean value imputation and data deletion. Features refer to the attributes and variables of data required for prediction, such as population density and weather, while target variables refer to the data to be predicted, such as waste generation volume.

[0927] Using the preprocessed data, the server trains a random forest regression model, a machine learning technique that uses a large number of decision trees to improve prediction accuracy. The server uses this random forest model to learn from past data and generate a model capable of predicting future waste generation patterns.

[0928] The trained model is then used to make predictions on a new dataset, which includes population density and weather data for a specific collection day. The user enters this information, and the server uses it to predict future waste generation. For example, to create a collection plan for a new week, the generative AI model can be given the following prompt:

[0929] "Predict waste generation based on population density and weather data for the new week, and optimize waste collection routes accordingly."

[0930] Finally, the server optimizes waste collection routes based on the prediction results. Using an optimization algorithm, collection points are sorted in descending order of waste volume, and an efficient collection route is calculated. This collection route information is output to a terminal and can be checked by the user. It is also reflected in collection vehicle schedules, improving operational efficiency.

[0931] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0932] Thus, the present invention is data-driven and provides a concrete approach to streamlining urban waste management, thereby reducing environmental impact and operational costs.

[0933] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0934] Step 1:

[0935] Data loading

[0936] The server reads historical data from an external database or a CSV file (e.g., "waste_data.csv"). The input data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. The server stores this data in memory and prepares it for subsequent preprocessing steps. Specifically, the server specifies the file path and opens it to read the data.

[0937] Step 2:

[0938] Data preprocessing (missing value handling)

[0939] The server detects missing values ​​in the dataset loaded and performs deletion or imputation processing. Historical data is given as input. For example, if there are missing values ​​in weather data, they are imputed using historical average weather data. The output is a clean dataset with the missing values ​​processed. Specifically, it scans the dataset and processes missing values ​​in each column using the appropriate method.

[0940] Step 3:

[0941] Data preprocessing (separation of features and target variables)

[0942] The server uses a clean dataset to separate the features and target variables required for prediction. The input is a dataset with missing values ​​processed. The features include population density and weather data, and the target variables include waste generation volume. The output is a dataset separated into the features and target variables. Specifically, the data is separated based on each column.

[0943] Step 4:

[0944] Model training

[0945] The server trains a random forest regression model using the preprocessed data. The input is the separated features and the target variable. The server applies the random forest regression algorithm to generate and combine multiple decision trees. The output is a trained random forest model. Specifically, the data is shuffled to separate it into a training set and a validation set, and the model is optimized through cross-validation.

[0946] Step 5:

[0947] Entering New Data

[0948] The user inputs population density and weather data for a new collection date. The input is weather data and demographic information for a new week. The output is data formatted in a way that the server can use for predictions. Specifically, the user inputs new data through a web interface, and the server receives and preprocesses the data.

[0949] Step 6:

[0950] Model prediction

[0951] The server uses a new dataset to make predictions using the trained model. The input is the feature data for the new collection date. The output is a prediction of future waste generation volume. Specifically, the server inputs the preprocessed new data into the model and calculates the predicted value.

[0952] Step 7:

[0953] Collection route optimization

[0954] The server optimizes waste collection routes based on the prediction results. The input is the predicted waste generation data. The server uses an optimization algorithm to sort collection points in descending order of waste volume and calculates an efficient collection route. The output is optimized collection route information. Specifically, the algorithm is run based on the prediction results, and the collection route is calculated and visualized.

[0955] Step 8:

[0956] Output of results

[0957] The server outputs the optimized collection route to the terminal and displays it to the user. The optimized route information is given as input. The output is collection route information in a display format that can be confirmed by the user. Specifically, the calculation results are generated in table or map format and sent to the terminal.

[0958] (Application example 1)

[0959] 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."

[0960] Conventional waste collection systems lack the means to predict waste generation patterns and design efficient routes, resulting in poor collection efficiency. Similarly, in the supply of parts within factories, there is a lack of a means to design efficient supply routes based on past consumption data, resulting in wasted time and money in parts supply. To address these issues, the present invention aims to optimize waste collection and in-factory parts supply routes by applying preprocessing of historical data and machine learning models.

[0961] 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.

[0962] In this invention, the server includes a means for reading historical data, a means for preprocessing the read data, a means for training a machine learning model using the preprocessed data, a means for making predictions for a new data set using the trained model, a means for optimizing waste collection routes based on the prediction results, and a means for optimizing parts supply routes within the factory by applying a similar system, thereby enabling the efficiency of not only waste collection but also parts supply within the factory.

[0963] "Historical data" refers to recorded data about past events or activities, and in this context refers to waste generation and parts consumption data.

[0964] "Preprocessing" refers to the process of correcting or supplementing incomplete data and preparing it in an appropriate format prior to data analysis or training a machine learning model.

[0965] A "machine learning model" is a set of mathematical algorithms that learn patterns from past data and make predictions or classifications for new data.

[0966] "Training" is the process of using historical data to train a machine learning model and improve its accuracy.

[0967] "Prediction" is the act of using existing data and a trained machine learning model to estimate the value of future data points or events.

[0968] "Optimization" is the process of finding the most efficient use of available resources based on specific objectives and constraints.

[0969] "Waste collection route" refers to the specific route or sequence used to collect waste.

[0970] "Parts supply route" refers to the route and order for efficiently delivering parts within a factory to each workplace and process.

[0971] "Dataset" refers to a collection of data used for a particular calculation or analysis.

[0972] A "random forest regression model" is a type of machine learning algorithm that uses multiple decision trees to perform regression analysis.

[0973] The present invention is a system in which a server reads historical data, preprocesses the data, trains a machine learning model, makes predictions, and optimizes waste collection and in-factory parts supply routes. Hereinafter, an embodiment of the present invention will be described in detail.

[0974] System configuration

[0975] First, the server loads data on past waste generation and collection (historical data) from an external database or CSV file. Similarly, it loads data on in-factory parts consumption. This historical data includes information such as the amount of waste generated, collection location, collection date, weather, and population density. In-factory parts consumption data includes the amount of parts consumed, consumption location, consumption date, and work progress status in the factory.

[0976] Next, the server preprocesses the loaded history data, which includes the following steps:

[0977] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[0978] 2. Separation of features and target variables: Separate the features required for prediction (e.g., population density, weather, factory work progress) from the target variables (waste generation volume, parts consumption volume).

[0979] The server uses the preprocessed data to train a random forest regression model, which has the ability to learn from past data and predict future waste generation and component consumption patterns.

[0980] The server then uses the trained model to make predictions on a new dataset, which includes population density, weather data, and factory work progress data for a specific collection or consumption date, and predicts waste generation and parts consumption based on these inputs.

[0981] Finally, the server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. The optimization algorithm sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes.

[0982] Hardware and Software Configuration

[0983] This system uses the following hardware and software:

[0984] Hardware: Servers, factory robots, smart glasses

[0985] Software: Python, Pandas, Scikit-learn

[0986] The server loads historical data and preprocesses it using Pandas. It then uses Scikit-learn to train a random forest regression model to make predictions based on new data sets. The optimized collection and delivery routes are then displayed or instructed to factory robots or smart glasses.

[0987] Specific examples

[0988] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. Based on this prediction, it optimizes collection routes to ensure efficient operation of collection vehicles.

[0989] Within the factory, parts consumption data from the past year is used to predict parts consumption based on the new week's data (population density and factory work progress).Based on this prediction, robots within the factory operate optimal supply routes, aiming to achieve efficient parts supply.

[0990] Example prompt sentence:

[0991] "Generate Python code that uses the past year's in-factory parts consumption data (e.g., CSV file) to predict parts consumption based on the new week's data (population density and weather). Include code that optimizes parts supply routes within the factory based on the predictions."

[0992] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0993] Step 1:

[0994] The server reads the historical data. Specifically, the server retrieves data on past waste generation and parts consumption from an external database or a CSV file (e.g., "factory_parts_data.csv"). The input is the historical data, and the output is the retrieved dataset.

[0995] Step 2:

[0996] The data loaded by the server is preprocessed. Specifically, missing values ​​are removed and the features required for prediction (e.g., population density, weather, factory work progress) are separated from the target variables (waste generation volume, parts consumption volume). The input is the acquired dataset, and the output is the preprocessed dataset.

[0997] Step 3:

[0998] The server uses the preprocessed data to train a machine learning model. Specifically, it uses a random forest regression model to learn patterns of waste generation and component consumption from historical data. The input is the preprocessed dataset, and the output is the trained model.

[0999] Step 4:

[1000] The server uses the trained model to make predictions for a new data set. Specifically, it predicts the amount of waste generated and parts consumed using inputs such as population density, weather data, and factory work progress data for a new collection or consumption date. The input is the new data set, and the output is the prediction result.

[1001] Step 5:

[1002] The server optimizes waste collection routes and in-plant parts supply routes based on the prediction results. Specifically, it sorts collection points and supply points in descending order of waste volume and parts consumption, and calculates efficient collection and supply routes. The input is the prediction results, and the output is optimized route information.

[1003] Step 6:

[1004] The server sends the optimized waste collection route and parts supply route to related terminals and devices (e.g., factory robots, smart glasses). Specifically, the route information is displayed or instructed on a display device or control system. The input is the optimized route information, and the output is a confirmation of transmission to the terminal.

[1005] 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.

[1006] The present invention is a system in which a server reads historical data, preprocesses the data, and trains a machine learning model to predict waste generation patterns and optimize waste collection routes. Furthermore, by combining it with an emotion engine that recognizes user emotions, the user experience can be improved. The following describes embodiments of the present invention in detail.

[1007] System configuration

[1008] First, the server loads historical data on waste generation and collection from an external database or CSV file, including information on waste generation volume, collection locations, collection dates, weather, population density, and more.

[1009] Next, the server preprocesses the loaded historical data, which includes the following steps:

[1010] 1. Removing Missing Values: Removing or imputing missing values ​​present in a dataset.

[1011] 2. Separation of features and target variables: The data is shaped to separate the features required for prediction (e.g., population density, weather) from the target variable (amount of waste generated).

[1012] The server uses the preprocessed data to train a random forest regression model, which is capable of learning waste generation patterns from past data and predicting future waste generation.

[1013] The server then uses the trained model to make waste generation predictions for a new dataset, which includes population density and weather data for a specific collection day, and predicts waste generation based on these inputs.

[1014] Furthermore, we use an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the collected results and system suggestions.

[1015] The server optimizes collection routes based on the waste generation prediction results and user emotion data. Based on the information obtained from the emotion engine, adjustments are made taking complaints and feedback into consideration, and a collection route that satisfies the user is calculated.

[1016] Program processing flow

[1017] 1. Data loading:

[1018] The server retrieves historical data from an external database, for example, by reading data from a file called "waste_data.csv."

[1019] 2. Data Preprocessing:

[1020] The server preprocesses the data it has loaded. In this step, it removes missing values ​​and separates features from the target variable.

[1021] 3. Model training:

[1022] The server trains a random forest regression model using the preprocessed data.

[1023] 4. Prediction:

[1024] The server uses a new dataset (population density and weather data for days that have not yet been collected) to make predictions using the trained model.

[1025] 5. Collecting Emotional Data:

[1026] The server uses an emotion engine to collect user emotion data, measure how users react to the system, and collect positive or negative feedback.

[1027] 6. Collection route optimization:

[1028] The server optimizes collection routes based on waste generation prediction results and emotion data, thereby calculating efficient routes that will increase user satisfaction.

[1029] Specific examples

[1030] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation. It then uses an emotion engine to collect user feedback and takes this into account when optimizing collection routes.

[1031] In this way, the introduction of an emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[1032] The processing flow will be explained below.

[1033] Step 1:

[1034] The server reads historical data from an external database or a CSV file. For example, it retrieves information such as the amount of waste generated, collection location, collection date, weather, and population density from a file called waste_data.csv.

[1035] Step 2:

[1036] The server preprocesses the data it has loaded. Specifically, it removes or imputes missing values ​​and separates the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). This process results in a clean dataset.

[1037] Step 3:

[1038] The server trains a random forest regression model using the preprocessed data. It uses specific parameter settings (e.g., n_estimators=100, random_state=42) to train the model stably and efficiently.

[1039] Step 4:

[1040] The server uses the trained model to make predictions on a new dataset, which includes population density and weather data for days that haven't been collected. Based on these inputs, the server predicts new waste generation volumes.

[1041] Step 5:

[1042] The server uses an emotion engine to collect user emotion data. Specifically, it identifies situations where users provide feedback to the system, identifies positive or negative reactions, and records them in a database. This information is collected using methods such as text analysis and speech analysis.

[1043] Step 6:

[1044] The server analyzes the collected emotion data and optimizes the collection route based on the prediction results. Based on the emotion data, the collection route is adjusted to satisfy the user, balancing efficiency and user experience. For example, if a user expresses dissatisfaction with a previous collection route, improvements can be made based on that information.

[1045] Step 7:

[1046] The server notifies the user of the optimized collection route, including specific collection dates and times and route information. Based on this information, the user can operate the collection vehicle efficiently.

[1047] The above is a detailed explanation of each processing step for the server, terminal, and user, which allows the waste management system to achieve high prediction accuracy and user experience, while reducing environmental impact and operational costs.

[1048] Example 2

[1049] 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."

[1050] Conventional waste collection systems have low accuracy in predicting waste generation, which makes it difficult to optimize collection routes. Furthermore, adjustments were not made taking into account user feedback or emotions, preventing improvements in user satisfaction. Therefore, there was a need for efficient collection route calculations and adjustments that reflected user emotions.

[1051] 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.

[1052] In this invention, the server includes means for reading historical data, means for preprocessing the read data, means for training a predictive model using the preprocessed data, means for making predictions for a new data set using the trained predictive model, means for collecting emotion data, means for optimizing waste collection routes based on the prediction results, and means for adjusting the optimization results using the emotion data. This not only enables highly accurate prediction of waste generation patterns, but also realizes optimization of collection routes taking user emotions into consideration, enabling efficient collection operations and improved user satisfaction.

[1053] "Historical data" refers to data that includes information such as the amount of waste generated in the past, collection locations, collection dates, weather, and population density.

[1054] "Preprocessing" refers to the process of deleting missing values ​​from the loaded data and separating it into features and target variables.

[1055] A "predictive model" is a machine learning algorithm model that is trained using pre-processed data and used to predict future waste generation.

[1056] "Emotional data" is data that measures a user's emotions and collects negative or positive feedback.

[1057] "Waste collection route optimization" is the process of efficiently designing collection routes based on the results of forecasting waste generation amounts.

[1058] "Adjustment" refers to the process of rearranging collection routes optimized using emotion data to reflect user feedback.

[1059] A "regression algorithm" is a machine learning technique used to predict continuous values, and in the present system may include a random forest regression model.

[1060] A "new dataset" is data, including population density and weather data for future collection dates, that is used in the predictive model.

[1061] The present invention is a system in which a server reads historical data, pre-processes the data, and trains a predictive model to predict waste generation patterns and optimize waste collection routes. Furthermore, the server can collect sentiment data and make adjustments based on user feedback to improve the user experience. The following describes an embodiment of the present invention in detail.

[1062] First, the server reads historical data on past waste generation and collection from an external database or CSV file. This historical data includes information such as waste generation volume, collection location, collection date, weather, population density, etc. The server obtains the data from, for example, a database management system or file system.

[1063] Next, the server preprocesses the historical data it has loaded. This includes removing or completing missing values, and separating the features required for prediction (e.g., population density, weather) from the target variable (waste generation volume). A data analysis library (e.g., Pandas) is used for data preprocessing.

[1064] The server uses the preprocessed data to train a predictive model. This predictive model has the ability to learn waste generation patterns from past data and predict future waste generation volumes. For training, a regression algorithm from a machine learning library (e.g., Scikit-learn) is used.

[1065] The server then uses the trained model to predict waste generation for a new dataset, which includes population density and weather data for the next collection day or week, and predicts waste generation based on these inputs.

[1066] Furthermore, the server uses an emotion engine to collect user emotion data, which plays an important role in analyzing how users feel about the waste collection system. The server performs emotion analysis of user feedback using, for example, a text analysis API.

[1067] The server optimizes waste collection routes based on the prediction results and emotion data. Optimization algorithms such as linear programming and genetic algorithms are used to optimize collection routes. The emotion data is also used to adjust collection routes to reflect user feedback and complaints.

[1068] Specific examples

[1069] For example, let's assume that this invention is applied to a city's waste management system. The server reads data from a CSV file for the past year, preprocesses the data, and trains a predictive model. After training, it inputs population density and weather data for the next week to predict future waste generation. It then uses an emotion engine to collect feedback from users and optimizes collection routes accordingly.

[1070] Examples of prompts include:

[1071] "Predict next week's waste generation volume based on the past year's waste generation volume data, collection locations, collection dates, weather, and population density information. Additionally, suggest optimal collection routes based on user feedback."

[1072] In this way, the introduction of an emotion engine can improve not only the efficiency of waste collection but also user satisfaction, thereby reducing the impact on the environment and reducing operating costs at the same time.

[1073] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1074] Step 1:

[1075] The server reads historical data. Specifically, the server retrieves data such as past waste generation volume, collection locations, collection dates, weather, and population density from an external database or the "waste_data.csv" file. The input is the "waste_data.csv" file, and the output is a dataset in Pandas DataFrame format.

[1076] Step 2:

[1077] The server preprocesses the historical data. Specifically, the server removes or imputes missing values ​​and separates the data into the features and target variables required for prediction. The input is a DataFrame of historical data, and the output is the preprocessed feature data and target variable data in DataFrame format.

[1078] Step 3:

[1079] The server trains a predictive model. Specifically, the server trains a machine learning model using the preprocessed data. It uses Scikit-learn's random forest regression model. The input is the preprocessed feature data and the objective variable data, and the output is a trained predictive model.

[1080] Step 4:

[1081] The server makes a prediction based on the new dataset. Specifically, the server uses a predictive model to predict waste generation volume using population density and weather data for the next collection day as input. The input is the new dataset, and the output is the predicted waste generation volume.

[1082] Step 5:

[1083] The server collects emotion data. Specifically, it collects feedback from users via their devices and performs emotion analysis using an emotion engine. The input is the user's feedback (in text format), and the output is the emotion analysis result (positive, negative, etc.).

[1084] Step 6:

[1085] The server optimizes collection routes using emotion data. Specifically, the server calculates the optimal waste collection route based on the prediction results and emotion data. Linear programming and genetic algorithms are used for optimization. The inputs are the predicted waste generation volume and emotion data, and the output is the optimized collection route.

[1086] The above are the specific processing steps of this system's program. The input and output at each step, as well as the specific data processing and data calculation, have been explained, and the overall flow of the system has been clarified.

[1087] (Application example 2)

[1088] 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."

[1089] In modern factories, efficient waste management and the development of optimal collection routes are important challenges. In particular, irregular waste generation patterns and low user (factory manager) satisfaction with collection services pose concerns about increased operational costs and environmental impact. Furthermore, adjusting collection routes based on user feedback is likely to contribute to further efficiency and improved satisfaction. Given this background, existing waste management systems are required to be equipped with highly accurate waste generation predictions and the ability to collect and reflect user sentiment data.

[1090] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for reading history data, means for preprocessing the read data, means for training a machine learning model using the preprocessed data, means for making predictions for a new data set using the trained model, means for optimizing waste collection routes based on the prediction results, means for recognizing and collecting user emotion data, and means for adjusting collection routes based on the user emotion data. This not only improves the accuracy of waste generation predictions and provides optimized collection routes, but also enables efficient waste management with high user satisfaction by adjusting the routes based on user emotions.

[1091] "Historical data" means information relating to past waste generation and collection, including data such as waste generation volume, collection locations, collection dates, weather, and population density.

[1092] "Preprocessing" is a data processing method that removes missing values ​​and separates features and target variables in order to prepare the data in a format suitable for machine learning model training.

[1093] A "machine learning model" is an algorithm that uses given data to learn patterns and make predictions or classifications for future data.

[1094] A "random forest regression model" is a machine learning model that makes predictions by combining multiple decision trees, and is particularly effective when applied to regression problems, enabling highly accurate predictions.

[1095] An "emotion engine" is a system that recognizes users' emotions, collects and analyzes that data, and is used to understand users' feedback and emotional state.

[1096] "Collection route optimization" is a means of minimizing costs and time by adjusting collection routes based on prediction results and user emotion data in order to achieve efficient waste collection.

[1097] "User emotion data" refers to data collected by the emotion engine regarding the user's emotional state, including positive or negative feedback.

[1098] A "new dataset" is new input data, such as population density or weather on an uncollected day, that the trained model will use to make predictions.

[1099] "Missing values" are incomplete or missing data in a dataset that should be appropriately handled to improve the accuracy of the model.

[1100] This invention is a system for streamlining waste management within factories. The system utilizes historical data and machine learning models to predict waste generation patterns and optimize collection routes. Furthermore, by collecting user sentiment data, the system can be applied to adjusting collection routes.

[1101] Hardware and software used

[1102] Hardware: Factory robots, servers

[1103] Software: Python, TensorFlow, OpenCV (for emotion recognition), pandas (data preprocessing), scikit-learn (machine learning)

[1104] System program processing explanation

[1105] 1. Data loading

[1106] The server reads data on historical waste generation and collection from an external database or a CSV file, including information on waste generation volume, collection location, collection date, weather, population density, etc. For example, it retrieves data from a file called "waste_data.csv."

[1107] 2. Data Preprocessing

[1108] The server performs data preprocessing. In this step, missing values ​​are removed and features and target variables are separated. Specifically, features such as population density and weather are separated from target variables such as waste generation volume in order to input them into the machine learning model. The Python pandas library is used to format the data and process missing values.

[1109] 3. Model training

[1110] The server uses the preprocessed data to train a random forest regression model. Using the scikit-learn library, the model is built using the training data, enabling highly accurate predictions of waste generation.

[1111] 4. Prediction

[1112] The server uses the trained model to predict waste generation for a new dataset, such as population density or weather data for a specific day, and uses Python and a scikit-learn random forest regression model to make predictions based on this data.

[1113] 5. Collecting Emotional Data

[1114] The emotion engine collects emotional data from users such as factory managers. Specifically, it combines a camera using OpenCV with a Python script to recognize emotions from the user's facial expressions and analyzes positive or negative feedback.

[1115] 6. Optimizing collection routes

[1116] The server optimizes collection routes based on waste generation prediction results and user emotion data. This optimization process takes into account information obtained from emotion data and makes adjustments to increase user satisfaction. For example, if a user expresses negative emotions, the server takes measures such as changing the collection route or collection frequency.

[1117] Examples of concrete examples and prompts

[1118] For example, at a factory in a certain city, the server reads data from the past year from a CSV file, preprocesses the data, and trains the model. After training, it inputs population density and weather data for the new week to accurately predict waste generation volume. It then uses an emotion engine to collect feedback from the factory manager and takes this into consideration to optimize collection routes. An example of a prompt would be, "Predict the amount of waste generated within the factory and calculate the optimal collection route based on the user (manager) feedback."

[1119] In this way, the present invention simultaneously realizes efficient waste management and improved user satisfaction.

[1120] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1121] Step 1:

[1122] The server reads historical data about past waste generation and collection from a central database or CSV files. This input data includes waste generation volume, collection location, collection date, weather, population density, and other parameters. The obtained data frame is generated as output.

[1123] Step 2:

[1124] The server preprocesses the loaded data. Specifically, it removes missing values ​​and separates features (e.g., population density, weather) from the target variable (waste generation volume). It uses the pandas library to format the data and process missing values, and outputs the preprocessed dataset.

[1125] Step 3:

[1126] The server trains a random forest regression model using the preprocessed dataset. Using the Scikit-learn library, it builds a model from the training data and gives it the ability to predict waste generation. The output is a trained machine learning model.

[1127] Step 4:

[1128] The server uses the trained model to predict waste generation for a new dataset, which may include population density and weather information for a specific day. By inputting this new data into the trained model, a waste generation prediction result is obtained. The output is a predicted value data.

[1129] Step 5:

[1130] The device uses the emotion engine to collect user emotion data. Specifically, it uses a camera and the OpenCV library to analyze the user's face and recognize their emotional status, such as positive or negative. The detection results are output as emotion data.

[1131] Step 6:

[1132] The server optimizes waste collection routes based on the prediction results and the user's emotional data. If the collected emotional data indicates negative feedback, it adjusts the collection route or collection frequency. For example, it recalculates the collection route to increase user satisfaction or changes the collection frequency. This optimized collection route information is output.

[1133] The above are the specific processing steps of the system that realizes the application example. The specific operations performed at each step realize efficient waste management and improved user satisfaction.

[1134] 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.

[1135] 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.

[1136] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1137] 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.

[1138] 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.

[1139] 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.

[1140] 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).

[1141] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1142] 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."

[1143] 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.

[1144] 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).

[1145] 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.

[1146] 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.

[1147] 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.

[1148] 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.

[1149] 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.

[1150] 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.

[1151] 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.

[1152] 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.

[1153] 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.

[1154] 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.

[1155] The following is further disclosed regarding the above embodiment.

[1156] (Claim 1)

[1157] a means for reading historical data;

[1158] A means for preprocessing the loaded data;

[1159] means for training a machine learning model using the preprocessed data;

[1160] a means for using the trained model to make predictions on new datasets;

[1161] a means for optimizing waste collection routes based on the prediction results;

[1162] A system including:

[1163] (Claim 2)

[1164] The system of claim 1, further comprising means for removing missing values ​​and separating the feature and the target variable.

[1165] (Claim 3)

[1166] 2. The system of claim 1, wherein the machine learning model is a random forest regression model.

[1167] "Example 1"

[1168] (Claim 1)

[1169] a means for reading historical data;

[1170] A means for preprocessing the loaded data;

[1171] means for training a machine learning model using the preprocessed data;

[1172] a means for using the trained model to make predictions on new datasets;

[1173] a means for optimizing waste collection routes based on the prediction results;

[1174] The historical data includes information such as waste generation volume, collection location, collection date, weather, population density, etc.;

[1175] a means of removing or imputing missing values;

[1176] The system further includes means for separating the features and the target variable.

[1177] (Claim 2)

[1178] 10. The system of claim 1, further comprising means for training the preprocessed data with a random forest regression model.

[1179] (Claim 3)

[1180] The system of claim 1, further comprising means for performing a prediction using the trained model based on weather information and population density information included in the new dataset, and optimizing waste collection routes based on the prediction results.

[1181] "Application Example 1"

[1182] (Claim 1)

[1183] a means for reading historical data;

[1184] A means for preprocessing the loaded data;

[1185] means for training a machine learning model using the preprocessed data;

[1186] a means for using the trained model to make predictions on new datasets;

[1187] a means for optimizing waste collection routes based on the prediction results;

[1188] Applying a similar system to optimize parts supply routes within factories,

[1189] A system including:

[1190] (Claim 2)

[1191] The system of claim 1, further comprising means for removing missing values ​​and separating the feature and the target variable.

[1192] (Claim 3)

[1193] 2. The system of claim 1, wherein the machine learning model is a random forest regression model.

[1194] "Example 2: Combining Emotion Engines"

[1195] (Claim 1)

[1196] a means for reading historical data;

[1197] A means for preprocessing the loaded data;

[1198] means for training a predictive model using the preprocessed data;

[1199] means for making predictions on new datasets using the trained predictive model;

[1200] a means for collecting emotion data;

[1201] a means for optimizing waste collection routes based on the prediction results;

[1202] a means for utilizing sentiment data to adjust optimization results;

[1203] A system including:

[1204] (Claim 2)

[1205] The system of claim 1, further comprising means for removing missing values ​​and separating the feature and the target variable.

[1206] (Claim 3)

[1207] 10. The system of claim 1, wherein the predictive model uses a regression algorithm.

[1208] "Application example 2 when combining emotion engines"

[1209] (Claim 1)

[1210] a means for reading historical data;

[1211] A means for preprocessing the loaded data;

[1212] means for training a machine learning model using the preprocessed data;

[1213] a means for using the trained model to make predictions on new datasets;

[1214] a means for optimizing waste collection routes based on the prediction results;

[1215] means for recognizing and collecting user emotion data;

[1216] means for adjusting a collection route based on user emotion data;

[1217] A system including:

[1218] (Claim 2)

[1219] The system of claim 1, further comprising means for removing missing values ​​and separating the feature and the target variable.

[1220] (Claim 3)

[1221] 2. The system of claim 1, wherein the machine learning model is a random forest regression model. [Explanation of symbols]

[1222] 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 reading historical data; A means for preprocessing the loaded data; means for training a machine learning model using the preprocessed data; a means for using the trained model to make predictions on new datasets; a means for optimizing waste collection routes based on the prediction results; A system including:

2. The system according to claim 1 , further comprising means for deleting missing values ​​and separating the feature and the target variable.

3. The system of claim 1 , wherein the machine learning model is a random forest regression model.

Images