System

Abstract

A system is provided.SOLUTION: A system, comprising: means for obtaining a dataset; means for preprocessing the dataset; means for dividing the dataset into a training dataset and a test data; means for learning the dataset using a regression model; means for making a prediction using the test data; means for evaluating the prediction; and means for visualizing the evaluation.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

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

[0004] Modern urban planning requires rapid and effective responses to complex problems arising from the vast volume and diversity of data. In particular, as Japan's urbanization progresses rapidly, innovative methods are needed to simultaneously address a wide range of issues, including housing, transportation, and sustainability. However, traditional urban planning methods are limited in their ability to efficiently process, predict, and evaluate this diverse data. As a result, optimizing urban planning is difficult and effective decision-making is hindered. Therefore, there is a need for a system that can comprehensively analyze complex urban data and support sustainable, efficient, and human-centered urban planning. [Means for solving the problem]

[0005] The present invention solves these problems using the following means. A system is provided that includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, and a means for visualizing the evaluation results. This system efficiently processes vast amounts of diverse urban planning data and quickly performs predictions and evaluations. In particular, the adoption of a random forest regression model as the regression model improves prediction accuracy and provides high applicability even in complex urban environments. Furthermore, a means for separating explanatory variables and target variables from the data efficiently preprocesses the data, optimizing the model learning process. This makes it possible to provide effective and rapid solutions to the wide range of issues associated with Japan's urbanization.

[0006] "Means for acquiring data" refers to a function for collecting data required for the system from external data sources and incorporating it into the system.

[0007] "Preprocessing means" refers to functions for preparing data suitable for analysis and learning, such as cleaning acquired data, processing missing data, and normalizing data.

[0008] "Means for splitting data into training data and test data" is a function that splits a dataset into data for model learning (training data) and data for evaluation (test data).

[0009] "Means for learning data using regression models" refers to a function that uses statistical methods and machine learning algorithms to learn regularities and patterns from training data.

[0010] The "means for making predictions using test data" is a function for making predictions on test data using a trained model and generating predicted values.

[0011] The "means for evaluating prediction results" is a function for comparing the predicted values ​​generated by the model with the actual values ​​and evaluating the accuracy and performance of the model.

[0012] "Means for visualizing evaluation results" is a function that displays prediction results and evaluation results in a visual format such as a graph or chart, making them easy for users to understand. [Brief explanation of the drawings]

[0013] [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 illustrating 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

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] 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).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] The present invention is an urban planning support system that uses generative AI, and this system efficiently processes complex data related to urban planning and performs predictions and evaluations to support the optimization of urban planning. The following describes in detail an embodiment of the present invention.

[0035] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0036] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0037] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[0038] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[0039] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in a visual format such as graphs and charts, making them easy for users to understand.

[0040] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs training and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then view the results on their device and quickly and accurately understand how effective the proposed urban planning is.

[0041] This system will enable rapid and effective solutions to complex urban planning problems, and will enable effective planning support for the wide range of challenges that come with Japan's rapid urbanization.

[0042] The processing flow will be explained below.

[0043] Step 1:

[0044] Users upload urban planning data to their devices. This data is typically stored in CSV format and includes information on population density, transportation infrastructure, land use, etc. The devices then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0045] Step 2:

[0046] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, converting the received data into the appropriate format and preparing it for processing.

[0047] Step 3:

[0048] The server performs preprocessing, specifically removing missing values, cleaning the data, and normalizing it, to improve the quality of the data and make it suitable for model training.

[0049] Step 4:

[0050] The server divides the preprocessed data into explanatory variables (X) and response variables (y). The explanatory variables contain various features related to urban planning, while the response variables contain the values ​​to be predicted.

[0051] Step 5:

[0052] The server splits the data into training data and test data. Typically, 80% of the data is set as training data and 20% as test data, and they are used separately for learning and evaluation.

[0053] Step 6:

[0054] The server sets up a random forest regression model. Random forest regression models combine multiple decision trees to make predictions, resulting in high prediction accuracy. Hyperparameters are also configured here.

[0055] Step 7:

[0056] The server trains a random forest regression model using the training data, which learns patterns and features from the training data to improve its predictive capabilities.

[0057] Step 8:

[0058] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[0059] Step 9:

[0060] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[0061] Step 10:

[0062] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[0063] Step 11:

[0064] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and displays them to the user.

[0065] Step 12:

[0066] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[0067] Example 1

[0068] 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."

[0069] Conventional urban planning support systems require separate steps for data preprocessing, model training, and evaluation of prediction results, resulting in reduced work efficiency and making it difficult to optimize urban planning. Another issue is the time required for data transmission and preprocessing when processing large amounts of data. This means that users have to spend a great deal of time and money proposing urban plans and evaluating their effectiveness.

[0070] 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.

[0071] In this invention, the server includes a means for users to upload data from their terminals, a means for the server to receive the data and import it into a data frame, and a means for the server to pre-process the data, thereby enabling users to efficiently process the data and perform predictions and evaluations to support the optimization of urban planning.

[0072] "Means for users to upload data from their terminals" refers to the process by which users use their own terminals to send urban planning data to the system.

[0073] "Means by which the server receives data and imports it into a data frame" refers to the process by which the server receives data sent by the user and converts the data into a data frame format.

[0074] "Means by which the server pre-processes the data" refers to the process by which the server performs operations on the imported data, such as removing missing values, cleaning the data, and normalizing the data.

[0075] "Means by which the server separates data into explanatory variables and target variables" refers to the process by which the server separates preprocessed data into explanatory variables and target variables.

[0076] "Means by which the server divides data into training data and test data" refers to a process by which the server further divides the data divided into explanatory variables and target variables into training data and test data.

[0077] "Means by which the server uses a random forest regression model to learn data" refers to a process by which the server uses training data to train a random forest regression model.

[0078] "Means by which the server makes predictions using test data" refers to the process by which the server makes predictions on test data using a trained model.

[0079] "Means by which the server evaluates the prediction results" refers to the process by which the server compares the actual values ​​with the predicted values ​​to evaluate the prediction results, and uses an evaluation index such as mean squared error (MSE).

[0080] "Means for visualizing the evaluation results and transmitting the results to the terminal" refers to the process by which the server visualizes the evaluation results in graphs or charts and transmits them to the user's terminal.

[0081] A "random forest regression model" refers to a machine learning model that uses multiple decision trees to make highly accurate predictions.

[0082] "Means of transmitting data using a secure communication protocol" refers to the process of transmitting data to a server using a protocol for securely transmitting data (e.g., HTTPS).

[0083] MODE FOR CARRYING OUT THE INVENTION

[0084] This invention is an urban planning support system that efficiently processes data related to urban planning and performs forecasting and evaluation using a generative AI model. This system utilizes various hardware and software to allow users to upload urban planning data, preprocess it, split it into training data and test data, and perform model learning, prediction, evaluation, and visualization.

[0085] Uploading and Receiving Data

[0086] Users upload urban planning data in CSV format from their devices, including population density, transportation infrastructure, land use patterns, etc. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0087] Preprocessing and splitting data

[0088] The server imports the received data into a data frame using the Pandas library, then removes missing values, cleans the data, and normalizes the data by removing columns with more than 50% missing values ​​and imputing the remaining missing values ​​with the mean.

[0089] After preprocessing, the data is divided into explanatory variables (X) and target variables (y), and then split into training data and test data, with 80% as training data and 20% as test data.

[0090] Model training and prediction

[0091] The server sets up a random forest regression model using the RandomForestRegressor from the scikit-learn library, trains the model using the training data, and then makes predictions on the test data.

[0092] Evaluation and visualization of prediction results

[0093] After the prediction is made, the server evaluates the prediction results using the mean squared error (MSE). The evaluated results are visualized as graphs and charts using visualization libraries such as Matplotlib and Seaborn, making it easier for users to intuitively understand the evaluation results.

[0094] Specific examples

[0095] For example, if a user proposes a new urban planning scenario and wants to evaluate its effectiveness, they would use the system as follows: The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs learning and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then check the results on their device and quickly and accurately determine how effective the urban planning they submitted is.

[0096] Prompt Sentence Examples

[0097] Below are some example prompts to input to a generative AI model:

[0098] We provide new scenario data to evaluate future urban planning proposals. This data includes:

[0099] Population density (e.g. number of residents / km^2)

[0100] Transport infrastructure (e.g. number of bus stops, distance to train station)

[0101] Land use patterns (e.g., percentage commercial, percentage industrial)

[0102] Use this data to forecast and evaluate the proposed scenario. Specifically, follow these steps:

[0103] 1. Data preprocessing (removing missing values, cleaning data, normalizing)

[0104] 2. Split the data into training and testing (80% training data, 20% testing data)

[0105] 3. Training and predicting using a random forest regression model

[0106] 4. Evaluation using Mean Squared Error (MSE)

[0107] 5. Visualization and presentation of evaluation results

[0108] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0109] Step 1: Upload your data

[0110] The user prepares urban planning data (such as population density, transportation infrastructure, and land use patterns) in CSV format from their own device. The user opens the device's browser, selects the relevant file from the system's upload screen, and clicks the "Upload" button. The input is urban planning data in CSV format, and the output is data sent from the device to the server. This data is sent to the server using a secure communication protocol (e.g., HTTPS).

[0111] Step 2: Receiving the data

[0112] The server receives the data sent by the user. It verifies that the data was sent correctly over a secure protocol (e.g. HTTPS) and that the file format is CSV. The input is the CSV file sent to the server, and the output is the data in a valid DataFrame format. The server uses the Pandas library to import the data into a DataFrame.

[0113] Step 3: Preprocessing the data

[0114] The server performs preprocessing on the imported data. Specifically, it removes missing values, cleans invalid data, and normalizes the data. For example, it deletes columns with more than 50% missing values ​​and fills the remaining missing values ​​with the average value. It also converts inappropriate data (e.g., negative values) into appropriate values. The input is urban planning data in a data frame format, and the output is clean data after preprocessing.

[0115] Step 4: Splitting explanatory variables and response variables

[0116] After preprocessing is complete, the server divides the data into explanatory variables (e.g., transportation infrastructure, land use patterns) and target variables (e.g., land price increase rate). Specifically, it uses functions from the Pandas library to separate the target column from the entire data. The input is the preprocessed data, and the output is a dataset divided into explanatory variables and target variables.

[0117] Step 5: Split the training and test data

[0118] The server further splits the data split into explanatory variables and response variables into training data and test data. Using the train_test_split function from the scikit-learn library, 80% of the data is set as training data and 20% as test data. The input is the data split into explanatory variables and response variables, and the output is the training data and test data.

[0119] Step 6: Set up and train the model

[0120] The server sets up a random forest regression model using RandomForestRegressor from the scikit-learn library. Then, it trains the model using the training data (X_train, y_train). Specifically, it performs fitting on the training data and generates multiple decision trees. The input is 80% of the training data, and the output is the trained random forest regression model.

[0121] Step 7: Run a prediction

[0122] The server uses the trained random forest regression model to make predictions on the test data (X_test). The inputs are the trained model and 20% of the test data, and the output is the prediction result (y_pred). Specifically, the prediction is performed using the model's predict method.

[0123] Step 8: Evaluate the prediction results

[0124] To evaluate the prediction results, the server compares the actual value (y_test) with the predicted value (y_pred) and calculates the mean squared error (MSE). Specifically, it calculates MSE using the mean_squared_error function. The inputs are the actual value and the predicted value, and the output is MSE.

[0125] Step 9: Visualize and submit results

[0126] The server visualizes the evaluation results in graphs and charts and sends them to the user's device. It uses visualization libraries such as Matplotlib and Seaborn to convert the results into a format that is easy for users to understand. The input is the evaluation result data, and the output is the visualized evaluation results. The visualized results are sent to the user's device via a secure protocol.

[0127] Step 10: Check the results

[0128] Users can check the visualized evaluation results on their own devices, allowing them to intuitively and quickly understand the effectiveness of their proposed urban planning. The input is the visualization results sent from the server, and the output is the user's evaluation and feedback.

[0129] Through these steps, the system achieves the optimization and efficient evaluation of urban planning.The system uses generative AI models and prompt sentences to perform advanced data analysis.

[0130] (Application example 1)

[0131] 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."

[0132] Modern factories operate many robots, and maximizing their operational efficiency is essential. However, it is not easy to effectively analyze robot operation data and propose optimal operation schedules and placement. Another problem is that managers have few ways to grasp this information in real time and respond quickly.

[0133] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0134] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for acquiring robot operation data and proposing optimal operation schedules and placements, and a means for providing the proposal results via a smart device, thereby improving the efficiency of robot operation in factories and enabling managers to make quick decisions.

[0135] "Means for acquiring data" refers to the methods or processes for collecting data related to robot operations and transmitting it to the server.

[0136] "Preprocessing means" refers to a method for performing processes such as cleaning and normalization to make collected data easier to analyze.

[0137] "Means for splitting data into training data and test data" refers to a method for dividing a dataset into data for training a model and data for evaluation.

[0138] "Means for learning data using a regression model" refers to a method for constructing and learning a model using a regression algorithm in order to make numerical predictions based on data.

[0139] A "means for making predictions using test data" is a method for making predictions on test data using a trained model.

[0140] The "means for evaluating the prediction results" refers to a method for calculating evaluation indicators by comparing predicted values ​​with actual values ​​in order to evaluate the predictive performance of the model.

[0141] "Means for visualizing evaluation results" refers to a method for visually displaying evaluation results in graphs, charts, etc.

[0142] "Means for acquiring robot operation data and proposing optimal operation schedules and placements" refers to a method for collecting data on the operation status of robots and proposing efficient operation schedules and placements using a generative AI model.

[0143] "Means for providing proposal results via smart devices" refers to a method for displaying the generated operation schedule and placement proposal results to the administrator via a device such as a smartphone or tablet.

[0144] This invention is a factory robot operation management system that uses generative AI to efficiently process robot operation data within a factory and propose optimal operation schedules and placements, thereby maximizing the operational efficiency of the robots. The operation of the system is explained in detail below.

[0145] First, the user uploads the robot's operational data (e.g., operating time, error frequency, task completion time, location information, etc.) to the system from a device such as a smartphone or tablet. This data is saved in CSV format and sent to the server using a secure communication protocol (e.g., HTTPS).

[0146] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0147] The server then splits the data into training data and test data, which provides independent datasets for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server uses the test data to make predictions.

[0148] The prediction results are evaluated by the server. This evaluation uses the mean squared error (MSE), which compares the actual values ​​with the predicted values. The evaluation results are visualized in visual formats such as graphs and charts, making them easy for users to understand. The server also displays the generated operation schedules and placement proposals to managers via smart devices. Managers can check this information in real time and make decisions quickly.

[0149] For example, suppose a manager is planning a new robot operation schedule and wants to evaluate its optimality. The manager uploads current robot operation data from a terminal to the system. The terminal sends the data to a server, which receives, preprocesses, and splits the data, and then uses a random forest regression model to learn and predict. Finally, the server visualizes the evaluation results and sends them back to the terminal. The manager can view the results on the terminal and quickly and accurately understand how effective the new operation schedule is.

[0150] Generative AI models can be used to efficiently process complex data and help optimize robot operations.

[0151] Example prompt sentence:

[0152] "Please upload the operation data of the robots in your factory as a CSV file. For example, the following items (operating time, frequency of errors, task completion time, location information, etc.) will be included. The system will automatically generate an optimal operation schedule and propose the optimal placement of the robots."

[0153] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0154] Step 1:

[0155] The user prepares the robot's operation data (operating time, error frequency, task completion time, location information, etc.) and saves it in CSV format. The user selects this file using the terminal application and uploads it to the system. The operation data is in CSV format as input, and the data is sent to the server as output.

[0156] Step 2:

[0157] The terminal sends the CSV file to the server via a secure communication protocol (e.g., HTTPS). The terminal operates by uploading the file to a specified endpoint and performing encrypted communication to ensure security. The inputs are the CSV file data and communication protocol settings, and the output is the completion of data transfer to the server.

[0158] Step 3:

[0159] The server imports the CSV file received from the terminal into a data frame using the Pandas library. Specifically, the server executes the file reading code and loads the CSV data into memory. The received CSV file is used as input, and a Pandas data frame is generated as output.

[0160] Step 4:

[0161] The server preprocesses the data frame, specifically removing missing values, correcting outliers, and cleaning the data. It also normalizes the data to make it more analyzable. The input is a Pandas data frame, and the output is a clean, preprocessed data frame.

[0162] Step 5:

[0163] The server separates the preprocessed data into explanatory variables (X) and response variables (y). Specifically, the server selects feature columns from the data frame and splits them appropriately into X and y. The input is a clean data frame, and the output is a dataset of explanatory variables X and a dataset of response variables y.

[0164] Step 6:

[0165] The server splits the data into training data and test data. Specifically, the server uses the Scikit-learn library to split the dataset into 80% training data and 20% test data. The inputs are the explanatory variable X and the objective variable y, and the output is the training data and test data.

[0166] Step 7:

[0167] The server sets up a random forest regression model and trains it using the training data. Specifically, the server instantiates a RandomForestRegressor from the Scikit-learn library and inputs the training data to the model using the fit method. The inputs are the training datasets X_train and y_train, and the output is the trained model.

[0168] Step 8:

[0169] The server uses the trained model to predict the test data. Specifically, it uses the predict method to make predictions on the test data. The inputs are the trained model and the test dataset X_test, and the output is the prediction result.

[0170] Step 9:

[0171] The server evaluates the prediction results and calculates the mean squared error (MSE) to compare the actual value with the predicted value. Specifically, it calculates the error using the mean_squared_error function. The inputs are the prediction result y_pred and the actual value y_test, and the output is the evaluation result MSE.

[0172] Step 10:

[0173] The server visualizes the evaluation results in visual formats such as graphs and charts. Specifically, it uses the Matplotlib library to generate graphs and visually display the results. The inputs are the evaluation result MSE and the prediction result dataset, and the output is a visualized graph.

[0174] Step 11:

[0175] The server provides the generated operation schedule and placement proposal results via the smart device. Specifically, it sends data to the device using an API endpoint, allowing the administrator to view the results on the terminal. The proposal result data is input, and is displayed on the smart device as output.

[0176] Through the above processing steps, it is possible to efficiently analyze robot operation data within a factory and propose optimal operation schedules and placements.

[0177] 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.

[0178] The present invention is an urban planning support system that combines generative AI and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[0179] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0180] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0181] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[0182] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[0183] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in visual formats such as graphs and charts to make them easy for users to understand.

[0184] This is where the emotion engine, a key feature of the present invention, comes into play. The emotion engine uses technologies such as facial recognition, voice analysis, and text analysis to recognize the user's emotions. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice to identify emotions in real time.

[0185] The server then adjusts the prediction results based on the user's emotions recognized using this emotion engine. Specifically, if the user is stressed, the server provides more detailed explanations and supplementary information, while if the user is satisfied, the server displays concise results. The server also optimizes the visualization of the results based on the emotion. For example, if a positive emotion is recognized, the graph is displayed in brighter colors.

[0186] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and segments the data, and performs learning and prediction using a random forest regression model. The emotion engine then recognizes the user's emotions, and the server adjusts the prediction results and their visualization based on the emotion data. Finally, the server sends the results back to the device, and the user can view them on their device. This not only allows the user to quickly and accurately understand how effective the proposed urban planning is, but also provides emotionally sensitive feedback.

[0187] This system can provide fast and effective solutions to complex urban planning problems, and will provide advanced planning support for the diverse challenges that come with Japan's rapid urbanization. It also improves the user experience by recognizing user emotions and providing feedback based on those emotions, supporting more effective decision-making.

[0188] The processing flow will be explained below.

[0189] Step 1:

[0190] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format using their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0191] Step 2:

[0192] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, which converts the data into a format that can be used within the system.

[0193] Step 3:

[0194] The server preprocesses the received data, first by removing missing values, then by cleaning and normalizing the data, thereby ensuring data quality and improving the accuracy of the analysis.

[0195] Step 4:

[0196] The server divides the preprocessed data into explanatory variables (X) and objective variables (y). The explanatory variables contain various features related to urban planning, while the objective variables contain the values ​​to be predicted.

[0197] Step 5:

[0198] The server splits the data into training data and test data. Typically, 80% of the dataset is training data and 20% is test data. This split ensures a clear separation of the data used to train and evaluate the model.

[0199] Step 6:

[0200] The server sets up a random forest regression model. Random forests are models that use multiple decision trees to make predictions and have high predictive accuracy. At this stage, the necessary hyperparameters are also configured.

[0201] Step 7:

[0202] The server trains a random forest regression model using the training data, which allows the model to learn patterns and features to improve its predictive capabilities.

[0203] Step 8:

[0204] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[0205] Step 9:

[0206] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[0207] Step 10:

[0208] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[0209] Step 11:

[0210] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and prepares them for display to the user.

[0211] Step 12:

[0212] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[0213] Step 13:

[0214] The device activates an emotion engine to recognize the user's emotions. The emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to identify emotions, using technologies such as facial recognition, voice analysis, and text analysis.

[0215] Step 14:

[0216] The server receives the recognized user's emotion data, including, for example, whether the user is stressed or happy.

[0217] Step 15:

[0218] The server adjusts the prediction results based on the user's emotional data, for example, providing detailed explanations and additional information if the user is stressed, or displaying concise results if the user is happy.

[0219] Step 16:

[0220] The server optimizes the visualization of the results according to the user's emotions. For example, if a positive emotion is detected, the graph will be displayed in brighter colors.

[0221] Step 17:

[0222] The device displays optimized prediction results and visualized graphs to users, improving the user experience and helping them make more effective decisions.

[0223] Example 2

[0224] 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."

[0225] Conventional urban planning support systems are complex in the processing, prediction, and evaluation of large amounts of urban planning data, making them difficult to execute efficiently. Furthermore, they are unable to provide feedback or adjust results based on user emotions, making it difficult to optimize results for easy user understanding. This makes it difficult to support fast and effective decision-making, and there is a need for improved user experience, particularly one that takes emotions into consideration.

[0226] 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.

[0227] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for recognizing a user's emotion and adjusting the prediction results and display method based on the recognized emotion, and a means for transmitting data using a secure communication protocol, thereby enabling efficient processing of urban planning data and providing feedback according to the user's emotion.

[0228] The "means for acquiring data" is an input interface for users to upload urban planning data, which is then sent to the server.

[0229] "Preprocessing means" refers to automatically removing missing values ​​from the data, cleaning the data, normalizing the data, etc., and preparing the data in a format that can be analyzed.

[0230] A "means of splitting data into training and testing data" is a process of dividing a dataset into training and testing data in a certain proportion for the purpose of training and evaluating a model.

[0231] "Means of learning data using regression models" refers to machine learning models used to make predictions between data with relatively complex relationships, thereby enabling data analysis and predictions.

[0232] "Means for making predictions using test data" refers to the process of evaluating a trained model using a portion of the data to generate predictions.

[0233] The "means for evaluating the prediction results" refers to the use of evaluation indices (such as mean square error) to measure and judge the accuracy of the prediction results generated by the model.

[0234] "Means for visualizing evaluation results" refers to displaying prediction results and evaluation indicators in a visual format such as graphs or charts, making them easy for users to understand.

[0235] The "means for recognizing a user's emotions and adjusting the predicted results and display method based on the recognized emotions" refers to analyzing a user's emotions in real time using input devices such as a camera or microphone, and adapting the display format of information based on the results.

[0236] "Means of transmitting data using a secure communication protocol" refers to the communication means (e.g., HTTPS) used to securely transmit user data to the server, which encrypts and protects the data.

[0237] The present invention is an urban planning support system that combines a generative AI model and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[0238] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0239] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data undergoes preprocessing such as removing missing values, cleaning the data, and normalizing the data. After preprocessing, the data is divided into explanatory variables (X) and target variables (y).

[0240] The server then splits the data into training and test data: 80% of the data is training data and 20% is testing data, providing an independent dataset for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains the model using the training data.

[0241] After the model has completed training, the server uses the test data to make predictions. The prediction results are compared with the actual values ​​and evaluated using the mean squared error (MSE). The evaluation results are visualized in a visual format such as a graph or chart so that users can easily understand them.

[0242] This is where emotion recognition engines come in. Emotion recognition engines use technologies such as facial recognition, voice analysis, and text analysis to recognize a user's emotions. Specifically, they use cameras and microphones to analyze the user's facial expressions and tone of voice to identify emotions in real time. For example, if the user is satisfied, they can display a concise result, but if the user is stressed, they can provide a more detailed explanation or supplemental information.

[0243] The server adjusts the prediction results based on the user's recognized emotions and optimizes the visualization of the results. For example, if a positive emotion is recognized, the graph will be displayed in brighter colors. The concept is that dynamic feedback based on emotions improves the user experience.

[0244] As a concrete example, consider the case where a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data to the system from a terminal. Then, the user can enter the following prompt sentence:

[0245] "We will upload new urban planning data. The data includes population density, transportation infrastructure, and land use patterns."

[0246] The user's input data is sent to the server, where it undergoes a series of processes: preprocessing, modeling, prediction, and evaluation. Finally, an emotion recognition engine analyzes the user's reaction and adjusts the way the results are displayed. This allows users to quickly and accurately grasp the effects of urban planning and receive feedback that takes their emotions into consideration. This system will enable advanced planning support for the diverse challenges associated with Japan's rapid urbanization. It will also improve the user experience and support effective decision-making.

[0247] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0248] Step 1:

[0249] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices.

[0250] Input: City planning data in CSV format

[0251] Specifically, the user uses the system interface to select the target CSV file from the file selection dialog and clicks the upload button.

[0252] Output: HTTPS request from the terminal to the server

[0253] Step 2:

[0254] The terminal transmits data to the server using a secure communication protocol (eg, HTTPS).

[0255] Input: Urban planning data in CSV format uploaded by the user

[0256] Specifically, the device establishes secure communication using HTTPS and sends data to the server as a POST request.

[0257] Output: CSV data sent to the server

[0258] Step 3:

[0259] The server receives the data sent from the device and imports it into a data frame using the Pandas library.

[0260] Input: CSV data sent from the terminal

[0261] Specifically, the server reads the CSV file and converts it into a data frame using Pandas.

[0262] Output: Urban planning data in a data frame

[0263] Step 4:

[0264] The server performs preprocessing on the imported data, such as removing missing values, cleaning the data, and normalizing it.

[0265] Input: Dataframed urban planning data

[0266] Specifically, the server uses Pandas methods to remove missing values, clean outliers, and normalize each field.

[0267] Output: Preprocessed data frame

[0268] Step 5:

[0269] The server separates the preprocessed data into explanatory variables (X) and target variables (y).

[0270] Input: Preprocessed data frame

[0271] Specifically, the column containing the target variable is divided into y and the remaining columns into X.

[0272] Output: Dataset of explanatory and response variables

[0273] Step 6:

[0274] The server splits the data into training data (80%) and test data (20%).

[0275] Input: Dataset of explanatory and response variables

[0276] Specifically, the server splits the data using the train_test_split function of scikit-learn.

[0277] Output: A set of training and test data

[0278] Step 7:

[0279] The server sets up a random forest regression model and trains the model using the training data.

[0280] Input: training data

[0281] Specifically, the server instantiates a RandomForestRegressor from scikit-learn and trains the model using the fit method on the training data.

[0282] Output: A trained random forest model

[0283] Step 8:

[0284] The server uses the trained model to make predictions on the test data.

[0285] Input: Trained random forest model, test data

[0286] Specifically, the server uses the predict method on the test data to obtain the prediction results.

[0287] Output: Prediction results

[0288] Step 9:

[0289] The server compares the predicted results with the actual values ​​and calculates a performance metric (e.g., mean square error).

[0290] Input: Prediction results, actual values ​​of test data

[0291] Specifically, the server calculates the evaluation index using the mean_squared_error function of scikit-learn.

[0292] Output: Evaluation metric (e.g., mean squared error)

[0293] Step 10:

[0294] The server visualizes the evaluation results in visual formats such as graphs and charts.

[0295] Input: Evaluation Metrics

[0296] Specifically, the server visualizes the results using the Matplotlib library, drawing graphs and generating charts.

[0297] Output: Visualized evaluation results

[0298] Step 11:

[0299] The server activates an emotion recognition engine, analyzes the user's face and voice in real time via the camera and microphone, and identifies emotions.

[0300] Input: User's facial recognition and voice analysis data

[0301] Specifically, the server analyzes data acquired through the camera and microphone and recognizes emotions using facial recognition and voice analysis algorithms.

[0302] Output: Recognized user emotion data

[0303] Step 12:

[0304] The server adjusts how the prediction results are displayed based on the recognized user emotion.

[0305] Input: Recognized user emotion data, prediction results, evaluation results

[0306] Specifically, the server adds a detailed explanation if the user is feeling stressed, and displays the image in brighter colors if positive emotions are recognized.

[0307] Output: Display of sentiment-adjusted prediction and evaluation results

[0308] Step 13:

[0309] The server sends the final result to the terminal and displays it to the user.

[0310] Input: Emotion-adjusted predictions and evaluation results

[0311] Specifically, the server sends the results back to the terminal, which receives them and displays them to the user.

[0312] Output: The final result as it appears on your terminal.

[0313] (Application example 2)

[0314] 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."

[0315] In conventional brick-and-mortar store layout planning, store owners are faced with a huge amount of data, and processing that data and designing an optimal layout requires a great deal of time and effort. Furthermore, feedback and layout adjustments based on user needs and emotions are often not provided, resulting in insufficient optimization. This can have a negative impact on customer satisfaction and business performance. The present invention provides a system to solve these problems.

[0316] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0317] In this invention, the server includes means for acquiring data, means for preprocessing, means for dividing data into training data and test data, means for learning data using a regression model, means for making predictions using the test data, means for evaluating prediction results, means for visualizing the evaluation results, means for recognizing user emotions, means for adjusting prediction results based on emotions, and means for adjusting the display method of the results based on emotions, thereby enabling store owners to quickly and efficiently design optimal layouts and provide feedback that takes user emotions into consideration.

[0318] The "means for acquiring data" refers to a device or program for receiving data required by the system from outside.

[0319] The "preprocessing means" refers to a device or program that performs operations such as cleaning the acquired data, correcting missing values, and normalizing the data.

[0320] A "means for splitting data into training data and test data" is a device or program that splits a dataset appropriately for training and evaluating a model.

[0321] A "means for learning data using a regression model" is a device or program that executes a regression algorithm to build a predictive model from data.

[0322] A "means for making predictions using test data" is a device or program that uses a trained model to make predictions on new data sets.

[0323] A "means for evaluating prediction results" is a device or program that compares actual data with prediction results in order to evaluate the performance of a prediction model.

[0324] The "means for visualizing the evaluation results" is a device or program that displays the evaluation results in a visual format such as a graph or chart.

[0325] "Means for recognizing user emotions" refers to a device or program that uses a camera or microphone to analyze the user's facial expressions and tone of voice to identify emotions.

[0326] The "means for adjusting prediction results based on emotions" refers to a device or program that appropriately changes the prediction results or the content of the information provided based on the recognized emotions of the user.

[0327] The "means for adjusting the display method of results based on emotions" is a device or program that changes the displayed results or their format in response to the user's emotions.

[0328] This invention relates to a system for supporting layout optimization of brick-and-mortar stores. This system uses a generative AI model and an emotion recognition engine to propose optimal layouts based on data uploaded by users, and provides feedback according to the user's emotions.

[0329] System configuration

[0330] The present invention includes the following main means:

[0331] 1. How to obtain data

[0332] Data is received from the user terminal. This may be store data in CSV format, for example. The user terminal sends the data to the server using a secure communication protocol such as HTTPS.

[0333] 2. Pretreatment methods

[0334] The server uses the Pandas library to import the data, remove missing values, clean the data, and normalize it.

[0335] 3. A means of splitting the data into training and test data

[0336] Split the dataset into training and test data. A common practice is to use 80% of the data as training data and 20% as test data.

[0337] 4. How to train data using regression models

[0338] Set up a random forest regression model and train it using the training data.

[0339] 5. How to make predictions using test data

[0340] The model is used to make predictions on the test data and obtain the results.

[0341] 6. Means of evaluating prediction results

[0342] The accuracy of the prediction is evaluated using the mean squared error (MSE).

[0343] 7. Means of visualizing evaluation results

[0344] The evaluation results are displayed in visual formats such as graphs and charts. The Matplotlib library is used.

[0345] 8. How to Recognize User Emotions

[0346] It uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize emotions, and uses OpenCV and the SpeechRecognition library.

[0347] 9. How to adjust predictions based on emotions

[0348] Adjust predictions and information provided based on perceived emotions.

[0349] 10. A way to tailor how results are displayed based on sentiment

[0350] The color and level of detail of the results display are changed depending on the user's emotional state.

[0351] Program processing

[0352] The server uses the following infrastructure and software to process various data:

[0353] Pandas

[0354] Used for data preprocessing (cleaning, correction, normalization).

[0355] Scikit-Learn

[0356] Training and predicting with random forest regression models.

[0357] OpenCV

[0358] It is used to analyze camera images and recognize emotions from the user's facial expressions.

[0359] SpeechRecognition

[0360] Analyzes audio captured from a microphone to identify emotions.

[0361] Matplotlib

[0362] Prediction and evaluation results are displayed visually.

[0363] Specific examples

[0364] Suppose a store owner proposes a new product layout and wants to evaluate its effectiveness. The owner uploads store data saved in CSV file format to the app. The app receives this data, preprocesses it, and then uses a random forest regression model to learn and predict. The prediction results are evaluated and visualized in graphs and charts. When the user interacts with the system using the camera and microphone, the way the results are displayed is adjusted depending on their emotions. For example, if the user expresses positive emotions, the results are displayed in brighter colors.

[0365] Example prompt

[0366] "Please upload a CSV file of your new store layout. This data includes information such as customer movement patterns, sales performance, and product placement. Next, we will recognize your customers' real-time emotions, so please use your camera and microphone. We will then provide you with optimal feedback and layout evaluation results based on your emotions."

[0367] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0368] Step 1:

[0369] The user uploads store data (e.g., a CSV file containing customer movement patterns, sales performance, and product placement information) from the terminal.

[0370] Input: Store data in CSV format sent from the user's device.

[0371] Output: The raw CSV data sent to the server.

[0372] Step 2:

[0373] The server receives the data using a secure communication protocol such as HTTPS.

[0374] Input: The raw CSV data received.

[0375] Output: CSV data file stored on the server.

[0376] Step 3:

[0377] The server uses the Pandas library to import the CSV data into a data frame and preprocess the data (missing value correction, data cleaning, normalization).

[0378] Input: CSV format store data file.

[0379] Output: A preprocessed data frame.

[0380] Step 4:

[0381] Split the preprocessed data into training data and test data. Typically, 80% of the data is training data and 20% is test data.

[0382] Input: Preprocessed dataframe.

[0383] Output: Training dataset, Test dataset.

[0384] Step 5:

[0385] The server sets up a random forest regression model using the Scikit-Learn library and trains the model using the training data.

[0386] Input: Training dataset.

[0387] Output: A trained random forest regression model.

[0388] Step 6:

[0389] Using the trained model, the server makes predictions on the test data.

[0390] Input: Test dataset, trained random forest regression model.

[0391] Output: The predictions made by the model.

[0392] Step 7:

[0393] The server evaluates the accuracy of the prediction results using the mean square error (MSE).

[0394] Input: Actual test data, predicted outcome data.

[0395] Output: MSE value as evaluation metric.

[0396] Step 8:

[0397] To display the evaluation results in visual formats such as graphs and charts, we use the Matplotlib library.

[0398] Input: Evaluation metrics and prediction results.

[0399] Output: A visually displayed graph or chart.

[0400] Step 9:

[0401] The user interacts with the system using a camera and microphone, and the server recognizes the user's emotions using OpenCV and the SpeechRecognition library.

[0402] Input: Real-time captured video and audio data.

[0403] Output: The perceived emotional state of the user.

[0404] Step 10:

[0405] The server adjusts the prediction results and the content of the information it provides based on the recognized user's emotions.

[0406] Input: Prediction results, user sentiment data.

[0407] Output: Prediction results adjusted based on sentiment.

[0408] Step 11:

[0409] The server adjusts the color tone and detail of the result display based on the emotion and shows the final result to the user.

[0410] Input: Tuned prediction results, sentiment data.

[0411] Output: The final result with adjusted color and detail.

[0412] These steps can effectively optimize the layout of physical stores and provide feedback based on user sentiment.

[0413] 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.

[0414] 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.

[0415] 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.

[0416] [Second embodiment]

[0417] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0418] 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.

[0419] 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).

[0420] 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.

[0421] 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.

[0422] 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).

[0423] 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.

[0424] 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.

[0425] 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.

[0426] 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.

[0427] 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.

[0428] 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."

[0429] The present invention is an urban planning support system that uses generative AI, and this system efficiently processes complex data related to urban planning and performs predictions and evaluations to support the optimization of urban planning. The following describes in detail an embodiment of the present invention.

[0430] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0431] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0432] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[0433] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[0434] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in a visual format such as graphs and charts, making them easy for users to understand.

[0435] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs training and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then view the results on their device and quickly and accurately understand how effective the proposed urban planning is.

[0436] This system will enable rapid and effective solutions to complex urban planning problems, and will enable effective planning support for the wide range of challenges that come with Japan's rapid urbanization.

[0437] The processing flow will be explained below.

[0438] Step 1:

[0439] Users upload urban planning data to their devices. This data is typically stored in CSV format and includes information on population density, transportation infrastructure, land use, etc. The devices then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0440] Step 2:

[0441] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, converting the received data into the appropriate format and preparing it for processing.

[0442] Step 3:

[0443] The server performs preprocessing, specifically removing missing values, cleaning the data, and normalizing it, to improve the quality of the data and make it suitable for model training.

[0444] Step 4:

[0445] The server divides the preprocessed data into explanatory variables (X) and response variables (y). The explanatory variables contain various features related to urban planning, while the response variables contain the values ​​to be predicted.

[0446] Step 5:

[0447] The server splits the data into training data and test data. Typically, 80% of the data is set as training data and 20% as test data, and they are used separately for learning and evaluation.

[0448] Step 6:

[0449] The server sets up a random forest regression model. Random forest regression models combine multiple decision trees to make predictions, resulting in high prediction accuracy. Hyperparameters are also configured here.

[0450] Step 7:

[0451] The server trains a random forest regression model using the training data, which learns patterns and features from the training data to improve its predictive capabilities.

[0452] Step 8:

[0453] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[0454] Step 9:

[0455] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[0456] Step 10:

[0457] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[0458] Step 11:

[0459] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and displays them to the user.

[0460] Step 12:

[0461] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[0462] Example 1

[0463] 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."

[0464] Conventional urban planning support systems require separate steps for data preprocessing, model training, and evaluation of prediction results, resulting in reduced work efficiency and making it difficult to optimize urban planning. Another issue is the time required for data transmission and preprocessing when processing large amounts of data. This means that users have to spend a great deal of time and money proposing urban plans and evaluating their effectiveness.

[0465] 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.

[0466] In this invention, the server includes a means for users to upload data from their terminals, a means for the server to receive the data and import it into a data frame, and a means for the server to pre-process the data, thereby enabling users to efficiently process the data and perform predictions and evaluations to support the optimization of urban planning.

[0467] "Means for users to upload data from their terminals" refers to the process by which users use their own terminals to send urban planning data to the system.

[0468] "Means by which the server receives data and imports it into a data frame" refers to the process by which the server receives data sent by the user and converts the data into a data frame format.

[0469] "Means by which the server pre-processes the data" refers to the process by which the server performs operations on the imported data, such as removing missing values, cleaning the data, and normalizing the data.

[0470] "Means by which the server separates data into explanatory variables and target variables" refers to the process by which the server separates preprocessed data into explanatory variables and target variables.

[0471] "Means by which the server divides data into training data and test data" refers to a process by which the server further divides the data divided into explanatory variables and target variables into training data and test data.

[0472] "Means by which the server uses a random forest regression model to learn data" refers to a process by which the server uses training data to train a random forest regression model.

[0473] "Means by which the server makes predictions using test data" refers to the process by which the server makes predictions on test data using a trained model.

[0474] "Means by which the server evaluates the prediction results" refers to the process by which the server compares the actual values ​​with the predicted values ​​to evaluate the prediction results, and uses an evaluation index such as mean squared error (MSE).

[0475] "Means for visualizing the evaluation results and transmitting the results to the terminal" refers to the process by which the server visualizes the evaluation results in graphs or charts and transmits them to the user's terminal.

[0476] A "random forest regression model" refers to a machine learning model that uses multiple decision trees to make highly accurate predictions.

[0477] "Means of transmitting data using a secure communication protocol" refers to the process of transmitting data to a server using a protocol for securely transmitting data (e.g., HTTPS).

[0478] MODE FOR CARRYING OUT THE INVENTION

[0479] This invention is an urban planning support system that efficiently processes data related to urban planning and performs forecasting and evaluation using a generative AI model. This system utilizes various hardware and software to allow users to upload urban planning data, preprocess it, split it into training data and test data, and perform model learning, prediction, evaluation, and visualization.

[0480] Uploading and Receiving Data

[0481] Users upload urban planning data in CSV format from their devices, including population density, transportation infrastructure, land use patterns, etc. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0482] Preprocessing and splitting data

[0483] The server imports the received data into a data frame using the Pandas library, then removes missing values, cleans the data, and normalizes the data by removing columns with more than 50% missing values ​​and imputing the remaining missing values ​​with the mean.

[0484] After preprocessing, the data is divided into explanatory variables (X) and target variables (y), and then split into training data and test data, with 80% as training data and 20% as test data.

[0485] Model training and prediction

[0486] The server sets up a random forest regression model using the RandomForestRegressor from the scikit-learn library, trains the model using the training data, and then makes predictions on the test data.

[0487] Evaluation and visualization of prediction results

[0488] After the prediction is made, the server evaluates the prediction results using the mean squared error (MSE). The evaluated results are visualized as graphs and charts using visualization libraries such as Matplotlib and Seaborn, making it easier for users to intuitively understand the evaluation results.

[0489] Specific examples

[0490] For example, if a user proposes a new urban planning scenario and wants to evaluate its effectiveness, they would use the system as follows: The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs learning and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then check the results on their device and quickly and accurately determine how effective the urban planning they submitted is.

[0491] Prompt Sentence Examples

[0492] Below are some example prompts to input to a generative AI model:

[0493] We provide new scenario data to evaluate future urban planning proposals. This data includes:

[0494] Population density (e.g. number of residents / km^2)

[0495] Transport infrastructure (e.g. number of bus stops, distance to train station)

[0496] Land use patterns (e.g., percentage commercial, percentage industrial)

[0497] Use this data to forecast and evaluate the proposed scenario. Specifically, follow these steps:

[0498] 1. Data preprocessing (removing missing values, cleaning data, normalizing)

[0499] 2. Split the data into training and testing (80% training data, 20% testing data)

[0500] 3. Training and predicting using a random forest regression model

[0501] 4. Evaluation using Mean Squared Error (MSE)

[0502] 5. Visualization and presentation of evaluation results

[0503] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0504] Step 1: Upload your data

[0505] The user prepares urban planning data (such as population density, transportation infrastructure, and land use patterns) in CSV format from their own device. The user opens the device's browser, selects the relevant file from the system's upload screen, and clicks the "Upload" button. The input is urban planning data in CSV format, and the output is data sent from the device to the server. This data is sent to the server using a secure communication protocol (e.g., HTTPS).

[0506] Step 2: Receiving the data

[0507] The server receives the data sent by the user. It verifies that the data was sent correctly over a secure protocol (e.g. HTTPS) and that the file format is CSV. The input is the CSV file sent to the server, and the output is the data in a valid DataFrame format. The server uses the Pandas library to import the data into a DataFrame.

[0508] Step 3: Preprocessing the data

[0509] The server performs preprocessing on the imported data. Specifically, it removes missing values, cleans invalid data, and normalizes the data. For example, it deletes columns with more than 50% missing values ​​and fills the remaining missing values ​​with the average value. It also converts inappropriate data (e.g., negative values) into appropriate values. The input is urban planning data in a data frame format, and the output is clean data after preprocessing.

[0510] Step 4: Splitting explanatory variables and response variables

[0511] After preprocessing is complete, the server divides the data into explanatory variables (e.g., transportation infrastructure, land use patterns) and target variables (e.g., land price increase rate). Specifically, it uses functions from the Pandas library to separate the target column from the entire data. The input is the preprocessed data, and the output is a dataset divided into explanatory variables and target variables.

[0512] Step 5: Split the training and test data

[0513] The server further splits the data split into explanatory variables and response variables into training data and test data. Using the train_test_split function from the scikit-learn library, 80% of the data is set as training data and 20% as test data. The input is the data split into explanatory variables and response variables, and the output is the training data and test data.

[0514] Step 6: Set up and train the model

[0515] The server sets up a random forest regression model using RandomForestRegressor from the scikit-learn library. Then, it trains the model using the training data (X_train, y_train). Specifically, it performs fitting on the training data and generates multiple decision trees. The input is 80% of the training data, and the output is the trained random forest regression model.

[0516] Step 7: Run a prediction

[0517] The server uses the trained random forest regression model to make predictions on the test data (X_test). The inputs are the trained model and 20% of the test data, and the output is the prediction result (y_pred). Specifically, the prediction is performed using the model's predict method.

[0518] Step 8: Evaluate the prediction results

[0519] To evaluate the prediction results, the server compares the actual value (y_test) with the predicted value (y_pred) and calculates the mean squared error (MSE). Specifically, it calculates MSE using the mean_squared_error function. The inputs are the actual value and the predicted value, and the output is MSE.

[0520] Step 9: Visualize and submit results

[0521] The server visualizes the evaluation results in graphs and charts and sends them to the user's device. It uses visualization libraries such as Matplotlib and Seaborn to convert the results into a format that is easy for users to understand. The input is the evaluation result data, and the output is the visualized evaluation results. The visualized results are sent to the user's device via a secure protocol.

[0522] Step 10: Check the results

[0523] Users can check the visualized evaluation results on their own devices, allowing them to intuitively and quickly understand the effectiveness of their proposed urban planning. The input is the visualization results sent from the server, and the output is the user's evaluation and feedback.

[0524] Through these steps, the system achieves the optimization and efficient evaluation of urban planning.The system uses generative AI models and prompt sentences to perform advanced data analysis.

[0525] (Application example 1)

[0526] 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."

[0527] Modern factories operate many robots, and maximizing their operational efficiency is essential. However, it is not easy to effectively analyze robot operation data and propose optimal operation schedules and placement. Another problem is that managers have few ways to grasp this information in real time and respond quickly.

[0528] 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.

[0529] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for acquiring robot operation data and proposing optimal operation schedules and placements, and a means for providing the proposal results via a smart device, thereby improving the efficiency of robot operation in factories and enabling managers to make quick decisions.

[0530] "Means for acquiring data" refers to the methods or processes for collecting data related to robot operations and transmitting it to the server.

[0531] "Preprocessing means" refers to a method for performing processes such as cleaning and normalization to make collected data easier to analyze.

[0532] "Means for splitting data into training data and test data" refers to a method for dividing a dataset into data for training a model and data for evaluation.

[0533] "Means for learning data using a regression model" refers to a method for constructing and learning a model using a regression algorithm in order to make numerical predictions based on data.

[0534] A "means for making predictions using test data" is a method for making predictions on test data using a trained model.

[0535] The "means for evaluating the prediction results" refers to a method for calculating evaluation indicators by comparing predicted values ​​with actual values ​​in order to evaluate the predictive performance of the model.

[0536] "Means for visualizing evaluation results" refers to a method for visually displaying evaluation results in graphs, charts, etc.

[0537] "Means for acquiring robot operation data and proposing optimal operation schedules and placements" refers to a method for collecting data on the operation status of robots and proposing efficient operation schedules and placements using a generative AI model.

[0538] "Means for providing proposal results via smart devices" refers to a method for displaying the generated operation schedule and placement proposal results to the administrator via a device such as a smartphone or tablet.

[0539] This invention is a factory robot operation management system that uses generative AI to efficiently process robot operation data within a factory and propose optimal operation schedules and placements, thereby maximizing the operational efficiency of the robots. The operation of the system is explained in detail below.

[0540] First, the user uploads the robot's operational data (e.g., operating time, error frequency, task completion time, location information, etc.) to the system from a device such as a smartphone or tablet. This data is saved in CSV format and sent to the server using a secure communication protocol (e.g., HTTPS).

[0541] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0542] The server then splits the data into training data and test data, which provides independent datasets for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server uses the test data to make predictions.

[0543] The prediction results are evaluated by the server. This evaluation uses the mean squared error (MSE), which compares the actual values ​​with the predicted values. The evaluation results are visualized in visual formats such as graphs and charts, making them easy for users to understand. The server also displays the generated operation schedules and placement proposals to managers via smart devices. Managers can check this information in real time and make decisions quickly.

[0544] For example, suppose a manager is planning a new robot operation schedule and wants to evaluate its optimality. The manager uploads current robot operation data from a terminal to the system. The terminal sends the data to a server, which receives, preprocesses, and splits the data, and then uses a random forest regression model to learn and predict. Finally, the server visualizes the evaluation results and sends them back to the terminal. The manager can view the results on the terminal and quickly and accurately understand how effective the new operation schedule is.

[0545] Generative AI models can be used to efficiently process complex data and help optimize robot operations.

[0546] Example prompt sentence:

[0547] "Please upload the operation data of the robots in your factory as a CSV file. For example, the following items (operating time, frequency of errors, task completion time, location information, etc.) will be included. The system will automatically generate an optimal operation schedule and propose the optimal placement of the robots."

[0548] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0549] Step 1:

[0550] The user prepares the robot's operation data (operating time, error frequency, task completion time, location information, etc.) and saves it in CSV format. The user selects this file using the terminal application and uploads it to the system. The operation data is in CSV format as input, and the data is sent to the server as output.

[0551] Step 2:

[0552] The terminal sends the CSV file to the server via a secure communication protocol (e.g., HTTPS). The terminal operates by uploading the file to a specified endpoint and performing encrypted communication to ensure security. The inputs are the CSV file data and communication protocol settings, and the output is the completion of data transfer to the server.

[0553] Step 3:

[0554] The server imports the CSV file received from the terminal into a data frame using the Pandas library. Specifically, the server executes the file reading code and loads the CSV data into memory. The received CSV file is used as input, and a Pandas data frame is generated as output.

[0555] Step 4:

[0556] The server preprocesses the data frame, specifically removing missing values, correcting outliers, and cleaning the data. It also normalizes the data to make it more analyzable. The input is a Pandas data frame, and the output is a clean, preprocessed data frame.

[0557] Step 5:

[0558] The server separates the preprocessed data into explanatory variables (X) and response variables (y). Specifically, the server selects feature columns from the data frame and splits them appropriately into X and y. The input is a clean data frame, and the output is a dataset of explanatory variables X and a dataset of response variables y.

[0559] Step 6:

[0560] The server splits the data into training data and test data. Specifically, the server uses the Scikit-learn library to split the dataset into 80% training data and 20% test data. The inputs are the explanatory variable X and the objective variable y, and the output is the training data and test data.

[0561] Step 7:

[0562] The server sets up a random forest regression model and trains it using the training data. Specifically, the server instantiates a RandomForestRegressor from the Scikit-learn library and inputs the training data to the model using the fit method. The inputs are the training datasets X_train and y_train, and the output is the trained model.

[0563] Step 8:

[0564] The server uses the trained model to predict the test data. Specifically, it uses the predict method to make predictions on the test data. The inputs are the trained model and the test dataset X_test, and the output is the prediction result.

[0565] Step 9:

[0566] The server evaluates the prediction results and calculates the mean squared error (MSE) to compare the actual value with the predicted value. Specifically, it calculates the error using the mean_squared_error function. The inputs are the prediction result y_pred and the actual value y_test, and the output is the evaluation result MSE.

[0567] Step 10:

[0568] The server visualizes the evaluation results in visual formats such as graphs and charts. Specifically, it uses the Matplotlib library to generate graphs and visually display the results. The inputs are the evaluation result MSE and the prediction result dataset, and the output is a visualized graph.

[0569] Step 11:

[0570] The server provides the generated operation schedule and placement proposal results via the smart device. Specifically, it sends data to the device using an API endpoint, allowing the administrator to view the results on the terminal. The proposal result data is input, and is displayed on the smart device as output.

[0571] Through the above processing steps, it is possible to efficiently analyze robot operation data within a factory and propose optimal operation schedules and placements.

[0572] 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.

[0573] The present invention is an urban planning support system that combines generative AI and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[0574] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0575] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0576] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[0577] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[0578] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in visual formats such as graphs and charts to make them easy for users to understand.

[0579] This is where the emotion engine, a key feature of the present invention, comes into play. The emotion engine uses technologies such as facial recognition, voice analysis, and text analysis to recognize the user's emotions. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice to identify emotions in real time.

[0580] The server then adjusts the prediction results based on the user's emotions recognized using this emotion engine. Specifically, if the user is stressed, the server provides more detailed explanations and supplementary information, while if the user is satisfied, the server displays concise results. The server also optimizes the visualization of the results based on the emotion. For example, if a positive emotion is recognized, the graph is displayed in brighter colors.

[0581] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and segments the data, and performs learning and prediction using a random forest regression model. The emotion engine then recognizes the user's emotions, and the server adjusts the prediction results and their visualization based on the emotion data. Finally, the server sends the results back to the device, and the user can view them on their device. This not only allows the user to quickly and accurately understand how effective the proposed urban planning is, but also provides emotionally sensitive feedback.

[0582] This system can provide fast and effective solutions to complex urban planning problems, and will provide advanced planning support for the diverse challenges that come with Japan's rapid urbanization. It also improves the user experience by recognizing user emotions and providing feedback based on those emotions, supporting more effective decision-making.

[0583] The processing flow will be explained below.

[0584] Step 1:

[0585] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format using their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0586] Step 2:

[0587] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, which converts the data into a format that can be used within the system.

[0588] Step 3:

[0589] The server preprocesses the received data, first by removing missing values, then by cleaning and normalizing the data, thereby ensuring data quality and improving the accuracy of the analysis.

[0590] Step 4:

[0591] The server divides the preprocessed data into explanatory variables (X) and objective variables (y). The explanatory variables contain various features related to urban planning, while the objective variables contain the values ​​to be predicted.

[0592] Step 5:

[0593] The server splits the data into training data and test data. Typically, 80% of the dataset is training data and 20% is test data. This split ensures a clear separation of the data used to train and evaluate the model.

[0594] Step 6:

[0595] The server sets up a random forest regression model. Random forests are models that use multiple decision trees to make predictions and have high predictive accuracy. At this stage, the necessary hyperparameters are also configured.

[0596] Step 7:

[0597] The server trains a random forest regression model using the training data, which allows the model to learn patterns and features to improve its predictive capabilities.

[0598] Step 8:

[0599] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[0600] Step 9:

[0601] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[0602] Step 10:

[0603] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[0604] Step 11:

[0605] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and prepares them for display to the user.

[0606] Step 12:

[0607] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[0608] Step 13:

[0609] The device activates an emotion engine to recognize the user's emotions. The emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to identify emotions, using technologies such as facial recognition, voice analysis, and text analysis.

[0610] Step 14:

[0611] The server receives the recognized user's emotion data, including, for example, whether the user is stressed or happy.

[0612] Step 15:

[0613] The server adjusts the prediction results based on the user's emotional data, for example, providing detailed explanations and additional information if the user is stressed, or displaying concise results if the user is happy.

[0614] Step 16:

[0615] The server optimizes the visualization of the results according to the user's emotions. For example, if a positive emotion is detected, the graph will be displayed in brighter colors.

[0616] Step 17:

[0617] The device displays optimized prediction results and visualized graphs to users, improving the user experience and helping them make more effective decisions.

[0618] Example 2

[0619] 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."

[0620] Conventional urban planning support systems are complex in the processing, prediction, and evaluation of large amounts of urban planning data, making them difficult to execute efficiently. Furthermore, they are unable to provide feedback or adjust results based on user emotions, making it difficult to optimize results for easy user understanding. This makes it difficult to support fast and effective decision-making, and there is a need for improved user experience, particularly one that takes emotions into consideration.

[0621] 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.

[0622] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for recognizing a user's emotion and adjusting the prediction results and display method based on the recognized emotion, and a means for transmitting data using a secure communication protocol, thereby enabling efficient processing of urban planning data and providing feedback according to the user's emotion.

[0623] The "means for acquiring data" is an input interface for users to upload urban planning data, which is then sent to the server.

[0624] "Preprocessing means" refers to automatically removing missing values ​​from the data, cleaning the data, normalizing the data, etc., and preparing the data in a format that can be analyzed.

[0625] A "means of splitting data into training and testing data" is a process of dividing a dataset into training and testing data in a certain proportion for the purpose of training and evaluating a model.

[0626] "Means of learning data using regression models" refers to machine learning models used to make predictions between data with relatively complex relationships, thereby enabling data analysis and predictions.

[0627] "Means for making predictions using test data" refers to the process of evaluating a trained model using a portion of the data to generate predictions.

[0628] The "means for evaluating the prediction results" refers to the use of evaluation indices (such as mean square error) to measure and judge the accuracy of the prediction results generated by the model.

[0629] "Means for visualizing evaluation results" refers to displaying prediction results and evaluation indicators in a visual format such as graphs or charts, making them easy for users to understand.

[0630] The "means for recognizing a user's emotions and adjusting the predicted results and display method based on the recognized emotions" refers to analyzing a user's emotions in real time using input devices such as a camera or microphone, and adapting the display format of information based on the results.

[0631] "Means of transmitting data using a secure communication protocol" refers to the communication means (e.g., HTTPS) used to securely transmit user data to the server, which encrypts and protects the data.

[0632] The present invention is an urban planning support system that combines a generative AI model and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[0633] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0634] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data undergoes preprocessing such as removing missing values, cleaning the data, and normalizing the data. After preprocessing, the data is divided into explanatory variables (X) and target variables (y).

[0635] The server then splits the data into training and test data: 80% of the data is training data and 20% is testing data, providing an independent dataset for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains the model using the training data.

[0636] After the model has completed training, the server uses the test data to make predictions. The prediction results are compared with the actual values ​​and evaluated using the mean squared error (MSE). The evaluation results are visualized in a visual format such as a graph or chart so that users can easily understand them.

[0637] This is where emotion recognition engines come in. Emotion recognition engines use technologies such as facial recognition, voice analysis, and text analysis to recognize a user's emotions. Specifically, they use cameras and microphones to analyze the user's facial expressions and tone of voice to identify emotions in real time. For example, if the user is satisfied, they can display a concise result, but if the user is stressed, they can provide a more detailed explanation or supplemental information.

[0638] The server adjusts the prediction results based on the user's recognized emotions and optimizes the visualization of the results. For example, if a positive emotion is recognized, the graph will be displayed in brighter colors. The concept is that dynamic feedback based on emotions improves the user experience.

[0639] As a concrete example, consider the case where a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data to the system from a terminal. Then, the user can enter the following prompt sentence:

[0640] "We will upload new urban planning data. The data includes population density, transportation infrastructure, and land use patterns."

[0641] The user's input data is sent to the server, where it undergoes a series of processes: preprocessing, modeling, prediction, and evaluation. Finally, an emotion recognition engine analyzes the user's reaction and adjusts the way the results are displayed. This allows users to quickly and accurately grasp the effects of urban planning and receive feedback that takes their emotions into consideration. This system will enable advanced planning support for the diverse challenges associated with Japan's rapid urbanization. It will also improve the user experience and support effective decision-making.

[0642] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0643] Step 1:

[0644] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices.

[0645] Input: City planning data in CSV format

[0646] Specifically, the user uses the system interface to select the target CSV file from the file selection dialog and clicks the upload button.

[0647] Output: HTTPS request from the terminal to the server

[0648] Step 2:

[0649] The terminal transmits data to the server using a secure communication protocol (eg, HTTPS).

[0650] Input: Urban planning data in CSV format uploaded by the user

[0651] Specifically, the device establishes secure communication using HTTPS and sends data to the server as a POST request.

[0652] Output: CSV data sent to the server

[0653] Step 3:

[0654] The server receives the data sent from the device and imports it into a data frame using the Pandas library.

[0655] Input: CSV data sent from the terminal

[0656] Specifically, the server reads the CSV file and converts it into a data frame using Pandas.

[0657] Output: Urban planning data in a data frame

[0658] Step 4:

[0659] The server performs preprocessing on the imported data, such as removing missing values, cleaning the data, and normalizing it.

[0660] Input: Dataframed urban planning data

[0661] Specifically, the server uses Pandas methods to remove missing values, clean outliers, and normalize each field.

[0662] Output: Preprocessed data frame

[0663] Step 5:

[0664] The server separates the preprocessed data into explanatory variables (X) and target variables (y).

[0665] Input: Preprocessed data frame

[0666] Specifically, the column containing the target variable is divided into y and the remaining columns into X.

[0667] Output: Dataset of explanatory and response variables

[0668] Step 6:

[0669] The server splits the data into training data (80%) and test data (20%).

[0670] Input: Dataset of explanatory and response variables

[0671] Specifically, the server splits the data using the train_test_split function of scikit-learn.

[0672] Output: A set of training and test data

[0673] Step 7:

[0674] The server sets up a random forest regression model and trains the model using the training data.

[0675] Input: training data

[0676] Specifically, the server instantiates a RandomForestRegressor from scikit-learn and trains the model using the fit method on the training data.

[0677] Output: A trained random forest model

[0678] Step 8:

[0679] The server uses the trained model to make predictions on the test data.

[0680] Input: Trained random forest model, test data

[0681] Specifically, the server uses the predict method on the test data to obtain the prediction results.

[0682] Output: Prediction results

[0683] Step 9:

[0684] The server compares the predicted results with the actual values ​​and calculates a performance metric (e.g., mean square error).

[0685] Input: Prediction results, actual values ​​of test data

[0686] Specifically, the server calculates the evaluation index using the mean_squared_error function of scikit-learn.

[0687] Output: Evaluation metric (e.g., mean squared error)

[0688] Step 10:

[0689] The server visualizes the evaluation results in visual formats such as graphs and charts.

[0690] Input: Evaluation Metrics

[0691] Specifically, the server visualizes the results using the Matplotlib library, drawing graphs and generating charts.

[0692] Output: Visualized evaluation results

[0693] Step 11:

[0694] The server activates an emotion recognition engine, analyzes the user's face and voice in real time via the camera and microphone, and identifies emotions.

[0695] Input: User's facial recognition and voice analysis data

[0696] Specifically, the server analyzes data acquired through the camera and microphone and recognizes emotions using facial recognition and voice analysis algorithms.

[0697] Output: Recognized user emotion data

[0698] Step 12:

[0699] The server adjusts how the prediction results are displayed based on the recognized user emotion.

[0700] Input: Recognized user emotion data, prediction results, evaluation results

[0701] Specifically, the server adds a detailed explanation if the user is feeling stressed, and displays the image in brighter colors if positive emotions are recognized.

[0702] Output: Display of sentiment-adjusted prediction and evaluation results

[0703] Step 13:

[0704] The server sends the final result to the terminal and displays it to the user.

[0705] Input: Emotion-adjusted predictions and evaluation results

[0706] Specifically, the server sends the results back to the terminal, which receives them and displays them to the user.

[0707] Output: The final result as it appears on your terminal.

[0708] (Application example 2)

[0709] 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."

[0710] In conventional brick-and-mortar store layout planning, store owners are faced with a huge amount of data, and processing that data and designing an optimal layout requires a great deal of time and effort. Furthermore, feedback and layout adjustments based on user needs and emotions are often not provided, resulting in insufficient optimization. This can have a negative impact on customer satisfaction and business performance. The present invention provides a system to solve these problems.

[0711] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0712] In this invention, the server includes means for acquiring data, means for preprocessing, means for dividing data into training data and test data, means for learning data using a regression model, means for making predictions using the test data, means for evaluating prediction results, means for visualizing the evaluation results, means for recognizing user emotions, means for adjusting prediction results based on emotions, and means for adjusting the display method of the results based on emotions, thereby enabling store owners to quickly and efficiently design optimal layouts and provide feedback that takes user emotions into consideration.

[0713] The "means for acquiring data" refers to a device or program for receiving data required by the system from outside.

[0714] The "preprocessing means" refers to a device or program that performs operations such as cleaning the acquired data, correcting missing values, and normalizing the data.

[0715] A "means for splitting data into training data and test data" is a device or program that splits a dataset appropriately for training and evaluating a model.

[0716] A "means for learning data using a regression model" is a device or program that executes a regression algorithm to build a predictive model from data.

[0717] A "means for making predictions using test data" is a device or program that uses a trained model to make predictions on new data sets.

[0718] A "means for evaluating prediction results" is a device or program that compares actual data with prediction results in order to evaluate the performance of a prediction model.

[0719] The "means for visualizing the evaluation results" is a device or program that displays the evaluation results in a visual format such as a graph or chart.

[0720] "Means for recognizing user emotions" refers to a device or program that uses a camera or microphone to analyze the user's facial expressions and tone of voice to identify emotions.

[0721] The "means for adjusting prediction results based on emotions" refers to a device or program that appropriately changes the prediction results or the content of the information provided based on the recognized emotions of the user.

[0722] The "means for adjusting the display method of results based on emotions" is a device or program that changes the displayed results or their format in response to the user's emotions.

[0723] This invention relates to a system for supporting layout optimization of brick-and-mortar stores. This system uses a generative AI model and an emotion recognition engine to propose optimal layouts based on data uploaded by users, and provides feedback according to the user's emotions.

[0724] System configuration

[0725] The present invention includes the following main means:

[0726] 1. How to obtain data

[0727] Data is received from the user terminal. This may be store data in CSV format, for example. The user terminal sends the data to the server using a secure communication protocol such as HTTPS.

[0728] 2. Pretreatment methods

[0729] The server uses the Pandas library to import the data, remove missing values, clean the data, and normalize it.

[0730] 3. A means of splitting the data into training and test data

[0731] Split the dataset into training and test data. A common practice is to use 80% of the data as training data and 20% as test data.

[0732] 4. How to train data using regression models

[0733] Set up a random forest regression model and train it using the training data.

[0734] 5. How to make predictions using test data

[0735] The model is used to make predictions on the test data and obtain the results.

[0736] 6. Means of evaluating prediction results

[0737] The accuracy of the prediction is evaluated using the mean squared error (MSE).

[0738] 7. Means of visualizing evaluation results

[0739] The evaluation results are displayed in visual formats such as graphs and charts. The Matplotlib library is used.

[0740] 8. How to Recognize User Emotions

[0741] It uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize emotions, and uses OpenCV and the SpeechRecognition library.

[0742] 9. How to adjust predictions based on emotions

[0743] Adjust predictions and information provided based on perceived emotions.

[0744] 10. A way to tailor how results are displayed based on sentiment

[0745] The color and level of detail of the results display are changed depending on the user's emotional state.

[0746] Program processing

[0747] The server uses the following infrastructure and software to process various data:

[0748] Pandas

[0749] Used for data preprocessing (cleaning, correction, normalization).

[0750] Scikit-Learn

[0751] Training and predicting with random forest regression models.

[0752] OpenCV

[0753] It is used to analyze camera images and recognize emotions from the user's facial expressions.

[0754] SpeechRecognition

[0755] Analyzes audio captured from a microphone to identify emotions.

[0756] Matplotlib

[0757] Prediction and evaluation results are displayed visually.

[0758] Specific examples

[0759] Suppose a store owner proposes a new product layout and wants to evaluate its effectiveness. The owner uploads store data saved in CSV file format to the app. The app receives this data, preprocesses it, and then uses a random forest regression model to learn and predict. The prediction results are evaluated and visualized in graphs and charts. When the user interacts with the system using the camera and microphone, the way the results are displayed is adjusted depending on their emotions. For example, if the user expresses positive emotions, the results are displayed in brighter colors.

[0760] Example prompt

[0761] "Please upload a CSV file of your new store layout. This data includes information such as customer movement patterns, sales performance, and product placement. Next, we will recognize your customers' real-time emotions, so please use your camera and microphone. We will then provide you with optimal feedback and layout evaluation results based on your emotions."

[0762] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0763] Step 1:

[0764] The user uploads store data (e.g., a CSV file containing customer movement patterns, sales performance, and product placement information) from the terminal.

[0765] Input: Store data in CSV format sent from the user's device.

[0766] Output: The raw CSV data sent to the server.

[0767] Step 2:

[0768] The server receives the data using a secure communication protocol such as HTTPS.

[0769] Input: The raw CSV data received.

[0770] Output: CSV data file stored on the server.

[0771] Step 3:

[0772] The server uses the Pandas library to import the CSV data into a data frame and preprocess the data (missing value correction, data cleaning, normalization).

[0773] Input: CSV format store data file.

[0774] Output: A preprocessed data frame.

[0775] Step 4:

[0776] Split the preprocessed data into training data and test data. Typically, 80% of the data is training data and 20% is test data.

[0777] Input: Preprocessed dataframe.

[0778] Output: Training dataset, Test dataset.

[0779] Step 5:

[0780] The server sets up a random forest regression model using the Scikit-Learn library and trains the model using the training data.

[0781] Input: Training dataset.

[0782] Output: A trained random forest regression model.

[0783] Step 6:

[0784] Using the trained model, the server makes predictions on the test data.

[0785] Input: Test dataset, trained random forest regression model.

[0786] Output: The predictions made by the model.

[0787] Step 7:

[0788] The server evaluates the accuracy of the prediction results using the mean square error (MSE).

[0789] Input: Actual test data, predicted outcome data.

[0790] Output: MSE value as evaluation metric.

[0791] Step 8:

[0792] To display the evaluation results in visual formats such as graphs and charts, we use the Matplotlib library.

[0793] Input: Evaluation metrics and prediction results.

[0794] Output: A visually displayed graph or chart.

[0795] Step 9:

[0796] The user interacts with the system using a camera and microphone, and the server recognizes the user's emotions using OpenCV and the SpeechRecognition library.

[0797] Input: Real-time captured video and audio data.

[0798] Output: The perceived emotional state of the user.

[0799] Step 10:

[0800] The server adjusts the prediction results and the content of the information it provides based on the recognized user's emotions.

[0801] Input: Prediction results, user sentiment data.

[0802] Output: Prediction results adjusted based on sentiment.

[0803] Step 11:

[0804] The server adjusts the color tone and detail of the result display based on the emotion and shows the final result to the user.

[0805] Input: Tuned prediction results, sentiment data.

[0806] Output: The final result with adjusted color and detail.

[0807] These steps can effectively optimize the layout of physical stores and provide feedback based on user sentiment.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] [Third embodiment]

[0812] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0813] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0814] 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).

[0815] 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.

[0816] 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.

[0817] 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).

[0818] 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.

[0819] 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.

[0820] 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.

[0821] 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.

[0822] 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.

[0823] 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."

[0824] The present invention is an urban planning support system that uses generative AI, and this system efficiently processes complex data related to urban planning and performs predictions and evaluations to support the optimization of urban planning. The following describes in detail an embodiment of the present invention.

[0825] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0826] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0827] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[0828] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[0829] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in a visual format such as graphs and charts, making them easy for users to understand.

[0830] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs training and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then view the results on their device and quickly and accurately understand how effective the proposed urban planning is.

[0831] This system will enable rapid and effective solutions to complex urban planning problems, and will enable effective planning support for the wide range of challenges that come with Japan's rapid urbanization.

[0832] The processing flow will be explained below.

[0833] Step 1:

[0834] Users upload urban planning data to their devices. This data is typically stored in CSV format and includes information on population density, transportation infrastructure, land use, etc. The devices then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0835] Step 2:

[0836] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, converting the received data into the appropriate format and preparing it for processing.

[0837] Step 3:

[0838] The server performs preprocessing, specifically removing missing values, cleaning the data, and normalizing it, to improve the quality of the data and make it suitable for model training.

[0839] Step 4:

[0840] The server divides the preprocessed data into explanatory variables (X) and response variables (y). The explanatory variables contain various features related to urban planning, while the response variables contain the values ​​to be predicted.

[0841] Step 5:

[0842] The server splits the data into training data and test data. Typically, 80% of the data is set as training data and 20% as test data, and they are used separately for learning and evaluation.

[0843] Step 6:

[0844] The server sets up a random forest regression model. Random forest regression models combine multiple decision trees to make predictions, resulting in high prediction accuracy. Hyperparameters are also configured here.

[0845] Step 7:

[0846] The server trains a random forest regression model using the training data, which learns patterns and features from the training data to improve its predictive capabilities.

[0847] Step 8:

[0848] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[0849] Step 9:

[0850] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[0851] Step 10:

[0852] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[0853] Step 11:

[0854] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and displays them to the user.

[0855] Step 12:

[0856] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[0857] Example 1

[0858] 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."

[0859] Conventional urban planning support systems require separate steps for data preprocessing, model training, and evaluation of prediction results, resulting in reduced work efficiency and making it difficult to optimize urban planning. Another issue is the time required for data transmission and preprocessing when processing large amounts of data. This means that users have to spend a great deal of time and money proposing urban plans and evaluating their effectiveness.

[0860] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0861] In this invention, the server includes a means for users to upload data from their terminals, a means for the server to receive the data and import it into a data frame, and a means for the server to pre-process the data, thereby enabling users to efficiently process the data and perform predictions and evaluations to support the optimization of urban planning.

[0862] "Means for users to upload data from their terminals" refers to the process by which users use their own terminals to send urban planning data to the system.

[0863] "Means by which the server receives data and imports it into a data frame" refers to the process by which the server receives data sent by the user and converts the data into a data frame format.

[0864] "Means by which the server pre-processes the data" refers to the process by which the server performs operations on the imported data, such as removing missing values, cleaning the data, and normalizing the data.

[0865] "Means by which the server separates data into explanatory variables and target variables" refers to the process by which the server separates preprocessed data into explanatory variables and target variables.

[0866] "Means by which the server divides data into training data and test data" refers to a process by which the server further divides the data divided into explanatory variables and target variables into training data and test data.

[0867] "Means by which the server uses a random forest regression model to learn data" refers to a process by which the server uses training data to train a random forest regression model.

[0868] "Means by which the server makes predictions using test data" refers to the process by which the server makes predictions on test data using a trained model.

[0869] "Means by which the server evaluates the prediction results" refers to the process by which the server compares the actual values ​​with the predicted values ​​to evaluate the prediction results, and uses an evaluation index such as mean squared error (MSE).

[0870] "Means for visualizing the evaluation results and transmitting the results to the terminal" refers to the process by which the server visualizes the evaluation results in graphs or charts and transmits them to the user's terminal.

[0871] A "random forest regression model" refers to a machine learning model that uses multiple decision trees to make highly accurate predictions.

[0872] "Means of transmitting data using a secure communication protocol" refers to the process of transmitting data to a server using a protocol for securely transmitting data (e.g., HTTPS).

[0873] MODE FOR CARRYING OUT THE INVENTION

[0874] This invention is an urban planning support system that efficiently processes data related to urban planning and performs forecasting and evaluation using a generative AI model. This system utilizes various hardware and software to allow users to upload urban planning data, preprocess it, split it into training data and test data, and perform model learning, prediction, evaluation, and visualization.

[0875] Uploading and Receiving Data

[0876] Users upload urban planning data in CSV format from their devices, including population density, transportation infrastructure, land use patterns, etc. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0877] Preprocessing and splitting data

[0878] The server imports the received data into a data frame using the Pandas library, then removes missing values, cleans the data, and normalizes the data by removing columns with more than 50% missing values ​​and imputing the remaining missing values ​​with the mean.

[0879] After preprocessing, the data is divided into explanatory variables (X) and target variables (y), and then split into training data and test data, with 80% as training data and 20% as test data.

[0880] Model training and prediction

[0881] The server sets up a random forest regression model using the RandomForestRegressor from the scikit-learn library, trains the model using the training data, and then makes predictions on the test data.

[0882] Evaluation and visualization of prediction results

[0883] After the prediction is made, the server evaluates the prediction results using the mean squared error (MSE). The evaluated results are visualized as graphs and charts using visualization libraries such as Matplotlib and Seaborn, making it easier for users to intuitively understand the evaluation results.

[0884] Specific examples

[0885] For example, if a user proposes a new urban planning scenario and wants to evaluate its effectiveness, they would use the system as follows: The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs learning and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then check the results on their device and quickly and accurately determine how effective the urban planning they submitted is.

[0886] Prompt Sentence Examples

[0887] Below are some example prompts to input to a generative AI model:

[0888] We provide new scenario data to evaluate future urban planning proposals. This data includes:

[0889] Population density (e.g. number of residents / km^2)

[0890] Transport infrastructure (e.g. number of bus stops, distance to train station)

[0891] Land use patterns (e.g., percentage commercial, percentage industrial)

[0892] Use this data to forecast and evaluate the proposed scenario. Specifically, follow these steps:

[0893] 1. Data preprocessing (removing missing values, cleaning data, normalizing)

[0894] 2. Split the data into training and testing (80% training data, 20% testing data)

[0895] 3. Training and predicting using a random forest regression model

[0896] 4. Evaluation using Mean Squared Error (MSE)

[0897] 5. Visualization and presentation of evaluation results

[0898] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0899] Step 1: Upload your data

[0900] The user prepares urban planning data (such as population density, transportation infrastructure, and land use patterns) in CSV format from their own device. The user opens the device's browser, selects the relevant file from the system's upload screen, and clicks the "Upload" button. The input is urban planning data in CSV format, and the output is data sent from the device to the server. This data is sent to the server using a secure communication protocol (e.g., HTTPS).

[0901] Step 2: Receiving the data

[0902] The server receives the data sent by the user. It verifies that the data was sent correctly over a secure protocol (e.g. HTTPS) and that the file format is CSV. The input is the CSV file sent to the server, and the output is the data in a valid DataFrame format. The server uses the Pandas library to import the data into a DataFrame.

[0903] Step 3: Preprocessing the data

[0904] The server performs preprocessing on the imported data. Specifically, it removes missing values, cleans invalid data, and normalizes the data. For example, it deletes columns with more than 50% missing values ​​and fills the remaining missing values ​​with the average value. It also converts inappropriate data (e.g., negative values) into appropriate values. The input is urban planning data in a data frame format, and the output is clean data after preprocessing.

[0905] Step 4: Splitting explanatory variables and response variables

[0906] After preprocessing is complete, the server divides the data into explanatory variables (e.g., transportation infrastructure, land use patterns) and target variables (e.g., land price increase rate). Specifically, it uses functions from the Pandas library to separate the target column from the entire data. The input is the preprocessed data, and the output is a dataset divided into explanatory variables and target variables.

[0907] Step 5: Split the training and test data

[0908] The server further splits the data split into explanatory variables and response variables into training data and test data. Using the train_test_split function from the scikit-learn library, 80% of the data is set as training data and 20% as test data. The input is the data split into explanatory variables and response variables, and the output is the training data and test data.

[0909] Step 6: Set up and train the model

[0910] The server sets up a random forest regression model using RandomForestRegressor from the scikit-learn library. Then, it trains the model using the training data (X_train, y_train). Specifically, it performs fitting on the training data and generates multiple decision trees. The input is 80% of the training data, and the output is the trained random forest regression model.

[0911] Step 7: Run a prediction

[0912] The server uses the trained random forest regression model to make predictions on the test data (X_test). The inputs are the trained model and 20% of the test data, and the output is the prediction result (y_pred). Specifically, the prediction is performed using the model's predict method.

[0913] Step 8: Evaluate the prediction results

[0914] To evaluate the prediction results, the server compares the actual value (y_test) with the predicted value (y_pred) and calculates the mean squared error (MSE). Specifically, it calculates MSE using the mean_squared_error function. The inputs are the actual value and the predicted value, and the output is MSE.

[0915] Step 9: Visualize and submit results

[0916] The server visualizes the evaluation results in graphs and charts and sends them to the user's device. It uses visualization libraries such as Matplotlib and Seaborn to convert the results into a format that is easy for users to understand. The input is the evaluation result data, and the output is the visualized evaluation results. The visualized results are sent to the user's device via a secure protocol.

[0917] Step 10: Check the results

[0918] Users can check the visualized evaluation results on their own devices, allowing them to intuitively and quickly understand the effectiveness of their proposed urban planning. The input is the visualization results sent from the server, and the output is the user's evaluation and feedback.

[0919] Through these steps, the system achieves the optimization and efficient evaluation of urban planning.The system uses generative AI models and prompt sentences to perform advanced data analysis.

[0920] (Application example 1)

[0921] 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."

[0922] Modern factories operate many robots, and maximizing their operational efficiency is essential. However, it is not easy to effectively analyze robot operation data and propose optimal operation schedules and placement. Another problem is that managers have few ways to grasp this information in real time and respond quickly.

[0923] 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.

[0924] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for acquiring robot operation data and proposing optimal operation schedules and placements, and a means for providing the proposal results via a smart device, thereby improving the efficiency of robot operation in factories and enabling managers to make quick decisions.

[0925] "Means for acquiring data" refers to the methods or processes for collecting data related to robot operations and transmitting it to the server.

[0926] "Preprocessing means" refers to a method for performing processes such as cleaning and normalization to make collected data easier to analyze.

[0927] "Means for splitting data into training data and test data" refers to a method for dividing a dataset into data for training a model and data for evaluation.

[0928] "Means for learning data using a regression model" refers to a method for constructing and learning a model using a regression algorithm in order to make numerical predictions based on data.

[0929] A "means for making predictions using test data" is a method for making predictions on test data using a trained model.

[0930] The "means for evaluating the prediction results" refers to a method for calculating evaluation indicators by comparing predicted values ​​with actual values ​​in order to evaluate the predictive performance of the model.

[0931] "Means for visualizing evaluation results" refers to a method for visually displaying evaluation results in graphs, charts, etc.

[0932] "Means for acquiring robot operation data and proposing optimal operation schedules and placements" refers to a method for collecting data on the operation status of robots and proposing efficient operation schedules and placements using a generative AI model.

[0933] "Means for providing proposal results via smart devices" refers to a method for displaying the generated operation schedule and placement proposal results to the administrator via a device such as a smartphone or tablet.

[0934] This invention is a factory robot operation management system that uses generative AI to efficiently process robot operation data within a factory and propose optimal operation schedules and placements, thereby maximizing the operational efficiency of the robots. The operation of the system is explained in detail below.

[0935] First, the user uploads the robot's operational data (e.g., operating time, error frequency, task completion time, location information, etc.) to the system from a device such as a smartphone or tablet. This data is saved in CSV format and sent to the server using a secure communication protocol (e.g., HTTPS).

[0936] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0937] The server then splits the data into training data and test data, which provides independent datasets for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server uses the test data to make predictions.

[0938] The prediction results are evaluated by the server. This evaluation uses the mean squared error (MSE), which compares the actual values ​​with the predicted values. The evaluation results are visualized in visual formats such as graphs and charts, making them easy for users to understand. The server also displays the generated operation schedules and placement proposals to managers via smart devices. Managers can check this information in real time and make decisions quickly.

[0939] For example, suppose a manager is planning a new robot operation schedule and wants to evaluate its optimality. The manager uploads current robot operation data from a terminal to the system. The terminal sends the data to a server, which receives, preprocesses, and splits the data, and then uses a random forest regression model to learn and predict. Finally, the server visualizes the evaluation results and sends them back to the terminal. The manager can view the results on the terminal and quickly and accurately understand how effective the new operation schedule is.

[0940] Generative AI models can be used to efficiently process complex data and help optimize robot operations.

[0941] Example prompt sentence:

[0942] "Please upload the operation data of the robots in your factory as a CSV file. For example, the following items (operating time, frequency of errors, task completion time, location information, etc.) will be included. The system will automatically generate an optimal operation schedule and propose the optimal placement of the robots."

[0943] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0944] Step 1:

[0945] The user prepares the robot's operation data (operating time, error frequency, task completion time, location information, etc.) and saves it in CSV format. The user selects this file using the terminal application and uploads it to the system. The operation data is in CSV format as input, and the data is sent to the server as output.

[0946] Step 2:

[0947] The terminal sends the CSV file to the server via a secure communication protocol (e.g., HTTPS). The terminal operates by uploading the file to a specified endpoint and performing encrypted communication to ensure security. The inputs are the CSV file data and communication protocol settings, and the output is the completion of data transfer to the server.

[0948] Step 3:

[0949] The server imports the CSV file received from the terminal into a data frame using the Pandas library. Specifically, the server executes the file reading code and loads the CSV data into memory. The received CSV file is used as input, and a Pandas data frame is generated as output.

[0950] Step 4:

[0951] The server preprocesses the data frame, specifically removing missing values, correcting outliers, and cleaning the data. It also normalizes the data to make it more analyzable. The input is a Pandas data frame, and the output is a clean, preprocessed data frame.

[0952] Step 5:

[0953] The server separates the preprocessed data into explanatory variables (X) and response variables (y). Specifically, the server selects feature columns from the data frame and splits them appropriately into X and y. The input is a clean data frame, and the output is a dataset of explanatory variables X and a dataset of response variables y.

[0954] Step 6:

[0955] The server splits the data into training data and test data. Specifically, the server uses the Scikit-learn library to split the dataset into 80% training data and 20% test data. The inputs are the explanatory variable X and the objective variable y, and the output is the training data and test data.

[0956] Step 7:

[0957] The server sets up a random forest regression model and trains it using the training data. Specifically, the server instantiates a RandomForestRegressor from the Scikit-learn library and inputs the training data to the model using the fit method. The inputs are the training datasets X_train and y_train, and the output is the trained model.

[0958] Step 8:

[0959] The server uses the trained model to predict the test data. Specifically, it uses the predict method to make predictions on the test data. The inputs are the trained model and the test dataset X_test, and the output is the prediction result.

[0960] Step 9:

[0961] The server evaluates the prediction results and calculates the mean squared error (MSE) to compare the actual value with the predicted value. Specifically, it calculates the error using the mean_squared_error function. The inputs are the prediction result y_pred and the actual value y_test, and the output is the evaluation result MSE.

[0962] Step 10:

[0963] The server visualizes the evaluation results in visual formats such as graphs and charts. Specifically, it uses the Matplotlib library to generate graphs and visually display the results. The inputs are the evaluation result MSE and the prediction result dataset, and the output is a visualized graph.

[0964] Step 11:

[0965] The server provides the generated operation schedule and placement proposal results via the smart device. Specifically, it sends data to the device using an API endpoint, allowing the administrator to view the results on the terminal. The proposal result data is input, and is displayed on the smart device as output.

[0966] Through the above processing steps, it is possible to efficiently analyze robot operation data within a factory and propose optimal operation schedules and placements.

[0967] 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.

[0968] The present invention is an urban planning support system that combines generative AI and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[0969] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[0970] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[0971] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[0972] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[0973] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in visual formats such as graphs and charts to make them easy for users to understand.

[0974] This is where the emotion engine, a key feature of the present invention, comes into play. The emotion engine uses technologies such as facial recognition, voice analysis, and text analysis to recognize the user's emotions. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice to identify emotions in real time.

[0975] The server then adjusts the prediction results based on the user's emotions recognized using this emotion engine. Specifically, if the user is stressed, the server provides more detailed explanations and supplementary information, while if the user is satisfied, the server displays concise results. The server also optimizes the visualization of the results based on the emotion. For example, if a positive emotion is recognized, the graph is displayed in brighter colors.

[0976] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and segments the data, and performs learning and prediction using a random forest regression model. The emotion engine then recognizes the user's emotions, and the server adjusts the prediction results and their visualization based on the emotion data. Finally, the server sends the results back to the device, and the user can view them on their device. This not only allows the user to quickly and accurately understand how effective the proposed urban planning is, but also provides emotionally sensitive feedback.

[0977] This system can provide fast and effective solutions to complex urban planning problems, and will provide advanced planning support for the diverse challenges that come with Japan's rapid urbanization. It also improves the user experience by recognizing user emotions and providing feedback based on those emotions, supporting more effective decision-making.

[0978] The processing flow will be explained below.

[0979] Step 1:

[0980] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format using their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[0981] Step 2:

[0982] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, which converts the data into a format that can be used within the system.

[0983] Step 3:

[0984] The server preprocesses the received data, first by removing missing values, then by cleaning and normalizing the data, thereby ensuring data quality and improving the accuracy of the analysis.

[0985] Step 4:

[0986] The server divides the preprocessed data into explanatory variables (X) and objective variables (y). The explanatory variables contain various features related to urban planning, while the objective variables contain the values ​​to be predicted.

[0987] Step 5:

[0988] The server splits the data into training data and test data. Typically, 80% of the dataset is training data and 20% is test data. This split ensures a clear separation of the data used to train and evaluate the model.

[0989] Step 6:

[0990] The server sets up a random forest regression model. Random forests are models that use multiple decision trees to make predictions and have high predictive accuracy. At this stage, the necessary hyperparameters are also configured.

[0991] Step 7:

[0992] The server trains a random forest regression model using the training data, which allows the model to learn patterns and features to improve its predictive capabilities.

[0993] Step 8:

[0994] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[0995] Step 9:

[0996] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[0997] Step 10:

[0998] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[0999] Step 11:

[1000] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and prepares them for display to the user.

[1001] Step 12:

[1002] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[1003] Step 13:

[1004] The device activates an emotion engine to recognize the user's emotions. The emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to identify emotions, using technologies such as facial recognition, voice analysis, and text analysis.

[1005] Step 14:

[1006] The server receives the recognized user's emotion data, including, for example, whether the user is stressed or happy.

[1007] Step 15:

[1008] The server adjusts the prediction results based on the user's emotional data, for example, providing detailed explanations and additional information if the user is stressed, or displaying concise results if the user is happy.

[1009] Step 16:

[1010] The server optimizes the visualization of the results according to the user's emotions. For example, if a positive emotion is detected, the graph will be displayed in brighter colors.

[1011] Step 17:

[1012] The device displays optimized prediction results and visualized graphs to users, improving the user experience and helping them make more effective decisions.

[1013] Example 2

[1014] 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."

[1015] Conventional urban planning support systems are complex in the processing, prediction, and evaluation of large amounts of urban planning data, making them difficult to execute efficiently. Furthermore, they are unable to provide feedback or adjust results based on user emotions, making it difficult to optimize results for easy user understanding. This makes it difficult to support fast and effective decision-making, and there is a need for improved user experience, particularly one that takes emotions into consideration.

[1016] 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.

[1017] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for recognizing a user's emotion and adjusting the prediction results and display method based on the recognized emotion, and a means for transmitting data using a secure communication protocol, thereby enabling efficient processing of urban planning data and providing feedback according to the user's emotion.

[1018] The "means for acquiring data" is an input interface for users to upload urban planning data, which is then sent to the server.

[1019] "Preprocessing means" refers to automatically removing missing values ​​from the data, cleaning the data, normalizing the data, etc., and preparing the data in a format that can be analyzed.

[1020] A "means of splitting data into training and testing data" is a process of dividing a dataset into training and testing data in a certain proportion for the purpose of training and evaluating a model.

[1021] "Means of learning data using regression models" refers to machine learning models used to make predictions between data with relatively complex relationships, thereby enabling data analysis and predictions.

[1022] "Means for making predictions using test data" refers to the process of evaluating a trained model using a portion of the data to generate predictions.

[1023] The "means for evaluating the prediction results" refers to the use of evaluation indices (such as mean square error) to measure and judge the accuracy of the prediction results generated by the model.

[1024] "Means for visualizing evaluation results" refers to displaying prediction results and evaluation indicators in a visual format such as graphs or charts, making them easy for users to understand.

[1025] The "means for recognizing a user's emotions and adjusting the predicted results and display method based on the recognized emotions" refers to analyzing a user's emotions in real time using input devices such as a camera or microphone, and adapting the display format of information based on the results.

[1026] "Means of transmitting data using a secure communication protocol" refers to the communication means (e.g., HTTPS) used to securely transmit user data to the server, which encrypts and protects the data.

[1027] The present invention is an urban planning support system that combines a generative AI model and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[1028] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[1029] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data undergoes preprocessing such as removing missing values, cleaning the data, and normalizing the data. After preprocessing, the data is divided into explanatory variables (X) and target variables (y).

[1030] The server then splits the data into training and test data: 80% of the data is training data and 20% is testing data, providing an independent dataset for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains the model using the training data.

[1031] After the model has completed training, the server uses the test data to make predictions. The prediction results are compared with the actual values ​​and evaluated using the mean squared error (MSE). The evaluation results are visualized in a visual format such as a graph or chart so that users can easily understand them.

[1032] This is where emotion recognition engines come in. Emotion recognition engines use technologies such as facial recognition, voice analysis, and text analysis to recognize a user's emotions. Specifically, they use cameras and microphones to analyze the user's facial expressions and tone of voice to identify emotions in real time. For example, if the user is satisfied, they can display a concise result, but if the user is stressed, they can provide a more detailed explanation or supplemental information.

[1033] The server adjusts the prediction results based on the user's recognized emotions and optimizes the visualization of the results. For example, if a positive emotion is recognized, the graph will be displayed in brighter colors. The concept is that dynamic feedback based on emotions improves the user experience.

[1034] As a concrete example, consider the case where a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data to the system from a terminal. Then, the user can enter the following prompt sentence:

[1035] "We will upload new urban planning data. The data includes population density, transportation infrastructure, and land use patterns."

[1036] The user's input data is sent to the server, where it undergoes a series of processes: preprocessing, modeling, prediction, and evaluation. Finally, an emotion recognition engine analyzes the user's reaction and adjusts the way the results are displayed. This allows users to quickly and accurately grasp the effects of urban planning and receive feedback that takes their emotions into consideration. This system will enable advanced planning support for the diverse challenges associated with Japan's rapid urbanization. It will also improve the user experience and support effective decision-making.

[1037] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1038] Step 1:

[1039] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices.

[1040] Input: City planning data in CSV format

[1041] Specifically, the user uses the system interface to select the target CSV file from the file selection dialog and clicks the upload button.

[1042] Output: HTTPS request from the terminal to the server

[1043] Step 2:

[1044] The terminal transmits data to the server using a secure communication protocol (eg, HTTPS).

[1045] Input: Urban planning data in CSV format uploaded by the user

[1046] Specifically, the device establishes secure communication using HTTPS and sends data to the server as a POST request.

[1047] Output: CSV data sent to the server

[1048] Step 3:

[1049] The server receives the data sent from the device and imports it into a data frame using the Pandas library.

[1050] Input: CSV data sent from the terminal

[1051] Specifically, the server reads the CSV file and converts it into a data frame using Pandas.

[1052] Output: Urban planning data in a data frame

[1053] Step 4:

[1054] The server performs preprocessing on the imported data, such as removing missing values, cleaning the data, and normalizing it.

[1055] Input: Dataframed urban planning data

[1056] Specifically, the server uses Pandas methods to remove missing values, clean outliers, and normalize each field.

[1057] Output: Preprocessed data frame

[1058] Step 5:

[1059] The server separates the preprocessed data into explanatory variables (X) and target variables (y).

[1060] Input: Preprocessed data frame

[1061] Specifically, the column containing the target variable is divided into y and the remaining columns into X.

[1062] Output: Dataset of explanatory and response variables

[1063] Step 6:

[1064] The server splits the data into training data (80%) and test data (20%).

[1065] Input: Dataset of explanatory and response variables

[1066] Specifically, the server splits the data using the train_test_split function of scikit-learn.

[1067] Output: A set of training and test data

[1068] Step 7:

[1069] The server sets up a random forest regression model and trains the model using the training data.

[1070] Input: training data

[1071] Specifically, the server instantiates a RandomForestRegressor from scikit-learn and trains the model using the fit method on the training data.

[1072] Output: A trained random forest model

[1073] Step 8:

[1074] The server uses the trained model to make predictions on the test data.

[1075] Input: Trained random forest model, test data

[1076] Specifically, the server uses the predict method on the test data to obtain the prediction results.

[1077] Output: Prediction results

[1078] Step 9:

[1079] The server compares the predicted results with the actual values ​​and calculates a performance metric (e.g., mean square error).

[1080] Input: Prediction results, actual values ​​of test data

[1081] Specifically, the server calculates the evaluation index using the mean_squared_error function of scikit-learn.

[1082] Output: Evaluation metric (e.g., mean squared error)

[1083] Step 10:

[1084] The server visualizes the evaluation results in visual formats such as graphs and charts.

[1085] Input: Evaluation Metrics

[1086] Specifically, the server visualizes the results using the Matplotlib library, drawing graphs and generating charts.

[1087] Output: Visualized evaluation results

[1088] Step 11:

[1089] The server activates an emotion recognition engine, analyzes the user's face and voice in real time via the camera and microphone, and identifies emotions.

[1090] Input: User's facial recognition and voice analysis data

[1091] Specifically, the server analyzes data acquired through the camera and microphone and recognizes emotions using facial recognition and voice analysis algorithms.

[1092] Output: Recognized user emotion data

[1093] Step 12:

[1094] The server adjusts how the prediction results are displayed based on the recognized user emotion.

[1095] Input: Recognized user emotion data, prediction results, evaluation results

[1096] Specifically, the server adds a detailed explanation if the user is feeling stressed, and displays the image in brighter colors if positive emotions are recognized.

[1097] Output: Display of sentiment-adjusted prediction and evaluation results

[1098] Step 13:

[1099] The server sends the final result to the terminal and displays it to the user.

[1100] Input: Emotion-adjusted predictions and evaluation results

[1101] Specifically, the server sends the results back to the terminal, which receives them and displays them to the user.

[1102] Output: The final result as it appears on your terminal.

[1103] (Application example 2)

[1104] 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."

[1105] In conventional brick-and-mortar store layout planning, store owners are faced with a huge amount of data, and processing that data and designing an optimal layout requires a great deal of time and effort. Furthermore, feedback and layout adjustments based on user needs and emotions are often not provided, resulting in insufficient optimization. This can have a negative impact on customer satisfaction and business performance. The present invention provides a system to solve these problems.

[1106] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1107] In this invention, the server includes means for acquiring data, means for preprocessing, means for dividing data into training data and test data, means for learning data using a regression model, means for making predictions using the test data, means for evaluating prediction results, means for visualizing the evaluation results, means for recognizing user emotions, means for adjusting prediction results based on emotions, and means for adjusting the display method of the results based on emotions, thereby enabling store owners to quickly and efficiently design optimal layouts and provide feedback that takes user emotions into consideration.

[1108] The "means for acquiring data" refers to a device or program for receiving data required by the system from outside.

[1109] The "preprocessing means" refers to a device or program that performs operations such as cleaning the acquired data, correcting missing values, and normalizing the data.

[1110] A "means for splitting data into training data and test data" is a device or program that splits a dataset appropriately for training and evaluating a model.

[1111] A "means for learning data using a regression model" is a device or program that executes a regression algorithm to build a predictive model from data.

[1112] A "means for making predictions using test data" is a device or program that uses a trained model to make predictions on new data sets.

[1113] A "means for evaluating prediction results" is a device or program that compares actual data with prediction results in order to evaluate the performance of a prediction model.

[1114] The "means for visualizing the evaluation results" is a device or program that displays the evaluation results in a visual format such as a graph or chart.

[1115] "Means for recognizing user emotions" refers to a device or program that uses a camera or microphone to analyze the user's facial expressions and tone of voice to identify emotions.

[1116] The "means for adjusting prediction results based on emotions" refers to a device or program that appropriately changes the prediction results or the content of the information provided based on the recognized emotions of the user.

[1117] The "means for adjusting the display method of results based on emotions" is a device or program that changes the displayed results or their format in response to the user's emotions.

[1118] This invention relates to a system for supporting layout optimization of brick-and-mortar stores. This system uses a generative AI model and an emotion recognition engine to propose optimal layouts based on data uploaded by users, and provides feedback according to the user's emotions.

[1119] System configuration

[1120] The present invention includes the following main means:

[1121] 1. How to obtain data

[1122] Data is received from the user terminal. This may be store data in CSV format, for example. The user terminal sends the data to the server using a secure communication protocol such as HTTPS.

[1123] 2. Pretreatment methods

[1124] The server uses the Pandas library to import the data, remove missing values, clean the data, and normalize it.

[1125] 3. A means of splitting the data into training and test data

[1126] Split the dataset into training and test data. A common practice is to use 80% of the data as training data and 20% as test data.

[1127] 4. How to train data using regression models

[1128] Set up a random forest regression model and train it using the training data.

[1129] 5. How to make predictions using test data

[1130] The model is used to make predictions on the test data and obtain the results.

[1131] 6. Means of evaluating prediction results

[1132] The accuracy of the prediction is evaluated using the mean squared error (MSE).

[1133] 7. Means of visualizing evaluation results

[1134] The evaluation results are displayed in visual formats such as graphs and charts. The Matplotlib library is used.

[1135] 8. How to Recognize User Emotions

[1136] It uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize emotions, and uses OpenCV and the SpeechRecognition library.

[1137] 9. How to adjust predictions based on emotions

[1138] Adjust predictions and information provided based on perceived emotions.

[1139] 10. A way to tailor how results are displayed based on sentiment

[1140] The color and level of detail of the results display are changed depending on the user's emotional state.

[1141] Program processing

[1142] The server uses the following infrastructure and software to process various data:

[1143] Pandas

[1144] Used for data preprocessing (cleaning, correction, normalization).

[1145] Scikit-Learn

[1146] Training and predicting with random forest regression models.

[1147] OpenCV

[1148] It is used to analyze camera images and recognize emotions from the user's facial expressions.

[1149] SpeechRecognition

[1150] Analyzes audio captured from a microphone to identify emotions.

[1151] Matplotlib

[1152] Prediction and evaluation results are displayed visually.

[1153] Specific examples

[1154] Suppose a store owner proposes a new product layout and wants to evaluate its effectiveness. The owner uploads store data saved in CSV file format to the app. The app receives this data, preprocesses it, and then uses a random forest regression model to learn and predict. The prediction results are evaluated and visualized in graphs and charts. When the user interacts with the system using the camera and microphone, the way the results are displayed is adjusted depending on their emotions. For example, if the user expresses positive emotions, the results are displayed in brighter colors.

[1155] Example prompt

[1156] "Please upload a CSV file of your new store layout. This data includes information such as customer movement patterns, sales performance, and product placement. Next, we will recognize your customers' real-time emotions, so please use your camera and microphone. We will then provide you with optimal feedback and layout evaluation results based on your emotions."

[1157] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1158] Step 1:

[1159] The user uploads store data (e.g., a CSV file containing customer movement patterns, sales performance, and product placement information) from the terminal.

[1160] Input: Store data in CSV format sent from the user's device.

[1161] Output: The raw CSV data sent to the server.

[1162] Step 2:

[1163] The server receives the data using a secure communication protocol such as HTTPS.

[1164] Input: The raw CSV data received.

[1165] Output: CSV data file stored on the server.

[1166] Step 3:

[1167] The server uses the Pandas library to import the CSV data into a data frame and preprocess the data (missing value correction, data cleaning, normalization).

[1168] Input: CSV format store data file.

[1169] Output: A preprocessed data frame.

[1170] Step 4:

[1171] Split the preprocessed data into training data and test data. Typically, 80% of the data is training data and 20% is test data.

[1172] Input: Preprocessed dataframe.

[1173] Output: Training dataset, Test dataset.

[1174] Step 5:

[1175] The server sets up a random forest regression model using the Scikit-Learn library and trains the model using the training data.

[1176] Input: Training dataset.

[1177] Output: A trained random forest regression model.

[1178] Step 6:

[1179] Using the trained model, the server makes predictions on the test data.

[1180] Input: Test dataset, trained random forest regression model.

[1181] Output: The predictions made by the model.

[1182] Step 7:

[1183] The server evaluates the accuracy of the prediction results using the mean square error (MSE).

[1184] Input: Actual test data, predicted outcome data.

[1185] Output: MSE value as evaluation metric.

[1186] Step 8:

[1187] To display the evaluation results in visual formats such as graphs and charts, we use the Matplotlib library.

[1188] Input: Evaluation metrics and prediction results.

[1189] Output: A visually displayed graph or chart.

[1190] Step 9:

[1191] The user interacts with the system using a camera and microphone, and the server recognizes the user's emotions using OpenCV and the SpeechRecognition library.

[1192] Input: Real-time captured video and audio data.

[1193] Output: The perceived emotional state of the user.

[1194] Step 10:

[1195] The server adjusts the prediction results and the content of the information it provides based on the recognized user's emotions.

[1196] Input: Prediction results, user sentiment data.

[1197] Output: Prediction results adjusted based on sentiment.

[1198] Step 11:

[1199] The server adjusts the color tone and detail of the result display based on the emotion and shows the final result to the user.

[1200] Input: Tuned prediction results, sentiment data.

[1201] Output: The final result with adjusted color and detail.

[1202] These steps can effectively optimize the layout of physical stores and provide feedback based on user sentiment.

[1203] 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.

[1204] 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.

[1205] 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.

[1206] [Fourth embodiment]

[1207] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1208] 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.

[1209] 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).

[1210] 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.

[1211] 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.

[1212] 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).

[1213] 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.

[1214] 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.

[1215] 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.

[1216] 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.

[1217] 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.

[1218] 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.

[1219] 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."

[1220] The present invention is an urban planning support system that uses generative AI, and this system efficiently processes complex data related to urban planning and performs predictions and evaluations to support the optimization of urban planning. The following describes in detail an embodiment of the present invention.

[1221] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[1222] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[1223] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[1224] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[1225] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in a visual format such as graphs and charts, making them easy for users to understand.

[1226] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs training and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then view the results on their device and quickly and accurately understand how effective the proposed urban planning is.

[1227] This system will enable rapid and effective solutions to complex urban planning problems, and will enable effective planning support for the wide range of challenges that come with Japan's rapid urbanization.

[1228] The processing flow will be explained below.

[1229] Step 1:

[1230] Users upload urban planning data to their devices. This data is typically stored in CSV format and includes information on population density, transportation infrastructure, land use, etc. The devices then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[1231] Step 2:

[1232] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, converting the received data into the appropriate format and preparing it for processing.

[1233] Step 3:

[1234] The server performs preprocessing, specifically removing missing values, cleaning the data, and normalizing it, to improve the quality of the data and make it suitable for model training.

[1235] Step 4:

[1236] The server divides the preprocessed data into explanatory variables (X) and response variables (y). The explanatory variables contain various features related to urban planning, while the response variables contain the values ​​to be predicted.

[1237] Step 5:

[1238] The server splits the data into training data and test data. Typically, 80% of the data is set as training data and 20% as test data, and they are used separately for learning and evaluation.

[1239] Step 6:

[1240] The server sets up a random forest regression model. Random forest regression models combine multiple decision trees to make predictions, resulting in high prediction accuracy. Hyperparameters are also configured here.

[1241] Step 7:

[1242] The server trains a random forest regression model using the training data, which learns patterns and features from the training data to improve its predictive capabilities.

[1243] Step 8:

[1244] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[1245] Step 9:

[1246] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[1247] Step 10:

[1248] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[1249] Step 11:

[1250] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and displays them to the user.

[1251] Step 12:

[1252] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[1253] Example 1

[1254] 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."

[1255] Conventional urban planning support systems require separate steps for data preprocessing, model training, and evaluation of prediction results, resulting in reduced work efficiency and making it difficult to optimize urban planning. Another issue is the time required for data transmission and preprocessing when processing large amounts of data. This means that users have to spend a great deal of time and money proposing urban plans and evaluating their effectiveness.

[1256] 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.

[1257] In this invention, the server includes a means for users to upload data from their terminals, a means for the server to receive the data and import it into a data frame, and a means for the server to pre-process the data, thereby enabling users to efficiently process the data and perform predictions and evaluations to support the optimization of urban planning.

[1258] "Means for users to upload data from their terminals" refers to the process by which users use their own terminals to send urban planning data to the system.

[1259] "Means by which the server receives data and imports it into a data frame" refers to the process by which the server receives data sent by the user and converts the data into a data frame format.

[1260] "Means by which the server pre-processes the data" refers to the process by which the server performs operations on the imported data, such as removing missing values, cleaning the data, and normalizing the data.

[1261] "Means by which the server separates data into explanatory variables and target variables" refers to the process by which the server separates preprocessed data into explanatory variables and target variables.

[1262] "Means by which the server divides data into training data and test data" refers to a process by which the server further divides the data divided into explanatory variables and target variables into training data and test data.

[1263] "Means by which the server uses a random forest regression model to learn data" refers to a process by which the server uses training data to train a random forest regression model.

[1264] "Means by which the server makes predictions using test data" refers to the process by which the server makes predictions on test data using a trained model.

[1265] "Means by which the server evaluates the prediction results" refers to the process by which the server compares the actual values ​​with the predicted values ​​to evaluate the prediction results, and uses an evaluation index such as mean squared error (MSE).

[1266] "Means for visualizing the evaluation results and transmitting the results to the terminal" refers to the process by which the server visualizes the evaluation results in graphs or charts and transmits them to the user's terminal.

[1267] A "random forest regression model" refers to a machine learning model that uses multiple decision trees to make highly accurate predictions.

[1268] "Means of transmitting data using a secure communication protocol" refers to the process of transmitting data to a server using a protocol for securely transmitting data (e.g., HTTPS).

[1269] MODE FOR CARRYING OUT THE INVENTION

[1270] This invention is an urban planning support system that efficiently processes data related to urban planning and performs forecasting and evaluation using a generative AI model. This system utilizes various hardware and software to allow users to upload urban planning data, preprocess it, split it into training data and test data, and perform model learning, prediction, evaluation, and visualization.

[1271] Uploading and Receiving Data

[1272] Users upload urban planning data in CSV format from their devices, including population density, transportation infrastructure, land use patterns, etc. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[1273] Preprocessing and splitting data

[1274] The server imports the received data into a data frame using the Pandas library, then removes missing values, cleans the data, and normalizes the data by removing columns with more than 50% missing values ​​and imputing the remaining missing values ​​with the mean.

[1275] After preprocessing, the data is divided into explanatory variables (X) and target variables (y), and then split into training data and test data, with 80% as training data and 20% as test data.

[1276] Model training and prediction

[1277] The server sets up a random forest regression model using the RandomForestRegressor from the scikit-learn library, trains the model using the training data, and then makes predictions on the test data.

[1278] Evaluation and visualization of prediction results

[1279] After the prediction is made, the server evaluates the prediction results using the mean squared error (MSE). The evaluated results are visualized as graphs and charts using visualization libraries such as Matplotlib and Seaborn, making it easier for users to intuitively understand the evaluation results.

[1280] Specific examples

[1281] For example, if a user proposes a new urban planning scenario and wants to evaluate its effectiveness, they would use the system as follows: The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and splits the data, and performs learning and prediction using a random forest regression model. Finally, the server visualizes the evaluation results and sends them back to the device. The user can then check the results on their device and quickly and accurately determine how effective the urban planning they submitted is.

[1282] Prompt Sentence Examples

[1283] Below are some example prompts to input to a generative AI model:

[1284] We provide new scenario data to evaluate future urban planning proposals. This data includes:

[1285] Population density (e.g. number of residents / km^2)

[1286] Transport infrastructure (e.g. number of bus stops, distance to train station)

[1287] Land use patterns (e.g., percentage commercial, percentage industrial)

[1288] Use this data to forecast and evaluate the proposed scenario. Specifically, follow these steps:

[1289] 1. Data preprocessing (removing missing values, cleaning data, normalizing)

[1290] 2. Split the data into training and testing (80% training data, 20% testing data)

[1291] 3. Training and predicting using a random forest regression model

[1292] 4. Evaluation using Mean Squared Error (MSE)

[1293] 5. Visualization and presentation of evaluation results

[1294] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1295] Step 1: Upload your data

[1296] The user prepares urban planning data (such as population density, transportation infrastructure, and land use patterns) in CSV format from their own device. The user opens the device's browser, selects the relevant file from the system's upload screen, and clicks the "Upload" button. The input is urban planning data in CSV format, and the output is data sent from the device to the server. This data is sent to the server using a secure communication protocol (e.g., HTTPS).

[1297] Step 2: Receiving the data

[1298] The server receives the data sent by the user. It verifies that the data was sent correctly over a secure protocol (e.g. HTTPS) and that the file format is CSV. The input is the CSV file sent to the server, and the output is the data in a valid DataFrame format. The server uses the Pandas library to import the data into a DataFrame.

[1299] Step 3: Preprocessing the data

[1300] The server performs preprocessing on the imported data. Specifically, it removes missing values, cleans invalid data, and normalizes the data. For example, it deletes columns with more than 50% missing values ​​and fills the remaining missing values ​​with the average value. It also converts inappropriate data (e.g., negative values) into appropriate values. The input is urban planning data in a data frame format, and the output is clean data after preprocessing.

[1301] Step 4: Splitting explanatory variables and response variables

[1302] After preprocessing is complete, the server divides the data into explanatory variables (e.g., transportation infrastructure, land use patterns) and target variables (e.g., land price increase rate). Specifically, it uses functions from the Pandas library to separate the target column from the entire data. The input is the preprocessed data, and the output is a dataset divided into explanatory variables and target variables.

[1303] Step 5: Split the training and test data

[1304] The server further splits the data split into explanatory variables and response variables into training data and test data. Using the train_test_split function from the scikit-learn library, 80% of the data is set as training data and 20% as test data. The input is the data split into explanatory variables and response variables, and the output is the training data and test data.

[1305] Step 6: Set up and train the model

[1306] The server sets up a random forest regression model using RandomForestRegressor from the scikit-learn library. Then, it trains the model using the training data (X_train, y_train). Specifically, it performs fitting on the training data and generates multiple decision trees. The input is 80% of the training data, and the output is the trained random forest regression model.

[1307] Step 7: Run a prediction

[1308] The server uses the trained random forest regression model to make predictions on the test data (X_test). The inputs are the trained model and 20% of the test data, and the output is the prediction result (y_pred). Specifically, the prediction is performed using the model's predict method.

[1309] Step 8: Evaluate the prediction results

[1310] To evaluate the prediction results, the server compares the actual value (y_test) with the predicted value (y_pred) and calculates the mean squared error (MSE). Specifically, it calculates MSE using the mean_squared_error function. The inputs are the actual value and the predicted value, and the output is MSE.

[1311] Step 9: Visualize and submit results

[1312] The server visualizes the evaluation results in graphs and charts and sends them to the user's device. It uses visualization libraries such as Matplotlib and Seaborn to convert the results into a format that is easy for users to understand. The input is the evaluation result data, and the output is the visualized evaluation results. The visualized results are sent to the user's device via a secure protocol.

[1313] Step 10: Check the results

[1314] Users can check the visualized evaluation results on their own devices, allowing them to intuitively and quickly understand the effectiveness of their proposed urban planning. The input is the visualization results sent from the server, and the output is the user's evaluation and feedback.

[1315] Through these steps, the system achieves the optimization and efficient evaluation of urban planning.The system uses generative AI models and prompt sentences to perform advanced data analysis.

[1316] (Application example 1)

[1317] 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."

[1318] Modern factories operate many robots, and maximizing their operational efficiency is essential. However, it is not easy to effectively analyze robot operation data and propose optimal operation schedules and placement. Another problem is that managers have few ways to grasp this information in real time and respond quickly.

[1319] 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.

[1320] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for acquiring robot operation data and proposing optimal operation schedules and placements, and a means for providing the proposal results via a smart device, thereby improving the efficiency of robot operation in factories and enabling managers to make quick decisions.

[1321] "Means for acquiring data" refers to the methods or processes for collecting data related to robot operations and transmitting it to the server.

[1322] "Preprocessing means" refers to a method for performing processes such as cleaning and normalization to make collected data easier to analyze.

[1323] "Means for splitting data into training data and test data" refers to a method for dividing a dataset into data for training a model and data for evaluation.

[1324] "Means for learning data using a regression model" refers to a method for constructing and learning a model using a regression algorithm in order to make numerical predictions based on data.

[1325] A "means for making predictions using test data" is a method for making predictions on test data using a trained model.

[1326] The "means for evaluating the prediction results" refers to a method for calculating evaluation indicators by comparing predicted values ​​with actual values ​​in order to evaluate the predictive performance of the model.

[1327] "Means for visualizing evaluation results" refers to a method for visually displaying evaluation results in graphs, charts, etc.

[1328] "Means for acquiring robot operation data and proposing optimal operation schedules and placements" refers to a method for collecting data on the operation status of robots and proposing efficient operation schedules and placements using a generative AI model.

[1329] "Means for providing proposal results via smart devices" refers to a method for displaying the generated operation schedule and placement proposal results to the administrator via a device such as a smartphone or tablet.

[1330] This invention is a factory robot operation management system that uses generative AI to efficiently process robot operation data within a factory and propose optimal operation schedules and placements, thereby maximizing the operational efficiency of the robots. The operation of the system is explained in detail below.

[1331] First, the user uploads the robot's operational data (e.g., operating time, error frequency, task completion time, location information, etc.) to the system from a device such as a smartphone or tablet. This data is saved in CSV format and sent to the server using a secure communication protocol (e.g., HTTPS).

[1332] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[1333] The server then splits the data into training data and test data, which provides independent datasets for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server uses the test data to make predictions.

[1334] The prediction results are evaluated by the server. This evaluation uses the mean squared error (MSE), which compares the actual values ​​with the predicted values. The evaluation results are visualized in visual formats such as graphs and charts, making them easy for users to understand. The server also displays the generated operation schedules and placement proposals to managers via smart devices. Managers can check this information in real time and make decisions quickly.

[1335] For example, suppose a manager is planning a new robot operation schedule and wants to evaluate its optimality. The manager uploads current robot operation data from a terminal to the system. The terminal sends the data to a server, which receives, preprocesses, and splits the data, and then uses a random forest regression model to learn and predict. Finally, the server visualizes the evaluation results and sends them back to the terminal. The manager can view the results on the terminal and quickly and accurately understand how effective the new operation schedule is.

[1336] Generative AI models can be used to efficiently process complex data and help optimize robot operations.

[1337] Example prompt sentence:

[1338] "Please upload the operation data of the robots in your factory as a CSV file. For example, the following items (operating time, frequency of errors, task completion time, location information, etc.) will be included. The system will automatically generate an optimal operation schedule and propose the optimal placement of the robots."

[1339] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1340] Step 1:

[1341] The user prepares the robot's operation data (operating time, error frequency, task completion time, location information, etc.) and saves it in CSV format. The user selects this file using the terminal application and uploads it to the system. The operation data is in CSV format as input, and the data is sent to the server as output.

[1342] Step 2:

[1343] The terminal sends the CSV file to the server via a secure communication protocol (e.g., HTTPS). The terminal operates by uploading the file to a specified endpoint and performing encrypted communication to ensure security. The inputs are the CSV file data and communication protocol settings, and the output is the completion of data transfer to the server.

[1344] Step 3:

[1345] The server imports the CSV file received from the terminal into a data frame using the Pandas library. Specifically, the server executes the file reading code and loads the CSV data into memory. The received CSV file is used as input, and a Pandas data frame is generated as output.

[1346] Step 4:

[1347] The server preprocesses the data frame, specifically removing missing values, correcting outliers, and cleaning the data. It also normalizes the data to make it more analyzable. The input is a Pandas data frame, and the output is a clean, preprocessed data frame.

[1348] Step 5:

[1349] The server separates the preprocessed data into explanatory variables (X) and response variables (y). Specifically, the server selects feature columns from the data frame and splits them appropriately into X and y. The input is a clean data frame, and the output is a dataset of explanatory variables X and a dataset of response variables y.

[1350] Step 6:

[1351] The server splits the data into training data and test data. Specifically, the server uses the Scikit-learn library to split the dataset into 80% training data and 20% test data. The inputs are the explanatory variable X and the objective variable y, and the output is the training data and test data.

[1352] Step 7:

[1353] The server sets up a random forest regression model and trains it using the training data. Specifically, the server instantiates a RandomForestRegressor from the Scikit-learn library and inputs the training data to the model using the fit method. The inputs are the training datasets X_train and y_train, and the output is the trained model.

[1354] Step 8:

[1355] The server uses the trained model to predict the test data. Specifically, it uses the predict method to make predictions on the test data. The inputs are the trained model and the test dataset X_test, and the output is the prediction result.

[1356] Step 9:

[1357] The server evaluates the prediction results and calculates the mean squared error (MSE) to compare the actual value with the predicted value. Specifically, it calculates the error using the mean_squared_error function. The inputs are the prediction result y_pred and the actual value y_test, and the output is the evaluation result MSE.

[1358] Step 10:

[1359] The server visualizes the evaluation results in visual formats such as graphs and charts. Specifically, it uses the Matplotlib library to generate graphs and visually display the results. The inputs are the evaluation result MSE and the prediction result dataset, and the output is a visualized graph.

[1360] Step 11:

[1361] The server provides the generated operation schedule and placement proposal results via the smart device. Specifically, it sends data to the device using an API endpoint, allowing the administrator to view the results on the terminal. The proposal result data is input, and is displayed on the smart device as output.

[1362] Through the above processing steps, it is possible to efficiently analyze robot operation data within a factory and propose optimal operation schedules and placements.

[1363] 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.

[1364] The present invention is an urban planning support system that combines generative AI and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[1365] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) from their own devices. This data is typically stored in CSV format. The devices then send the data to the server using a secure communication protocol (e.g., HTTPS).

[1366] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data is first preprocessed. Specifically, missing values ​​are removed, the data is cleaned, and the data is normalized. After preprocessing is complete, the data is separated into explanatory variables (X) and target variables (y).

[1367] The server then splits the data into training and test data, with 80% of the data set as training data and 20% as test data, providing independent datasets for evaluating the model's training and predictive accuracy.

[1368] The server then sets up a random forest regression model and trains it using the training data. Random forest regression models have high predictive accuracy because they use multiple decision trees to make predictions. After the model has completed training, the server makes predictions using the test data.

[1369] The prediction results are evaluated by the server using the mean squared error (MSE) that compares the actual values ​​with the predicted values, and the evaluation results are visualized in visual formats such as graphs and charts to make them easy for users to understand.

[1370] This is where the emotion engine, a key feature of the present invention, comes into play. The emotion engine uses technologies such as facial recognition, voice analysis, and text analysis to recognize the user's emotions. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice to identify emotions in real time.

[1371] The server then adjusts the prediction results based on the user's emotions recognized using this emotion engine. Specifically, if the user is stressed, the server provides more detailed explanations and supplementary information, while if the user is satisfied, the server displays concise results. The server also optimizes the visualization of the results based on the emotion. For example, if a positive emotion is recognized, the graph is displayed in brighter colors.

[1372] For example, suppose a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data from their device to the system. The device then sends the data to the server, which receives, preprocesses, and segments the data, and performs learning and prediction using a random forest regression model. The emotion engine then recognizes the user's emotions, and the server adjusts the prediction results and their visualization based on the emotion data. Finally, the server sends the results back to the device, and the user can view them on their device. This not only allows the user to quickly and accurately understand how effective the proposed urban planning is, but also provides emotionally sensitive feedback.

[1373] This system can provide fast and effective solutions to complex urban planning problems, and will provide advanced planning support for the diverse challenges that come with Japan's rapid urbanization. It also improves the user experience by recognizing user emotions and providing feedback based on those emotions, supporting more effective decision-making.

[1374] The processing flow will be explained below.

[1375] Step 1:

[1376] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format using their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[1377] Step 2:

[1378] The server receives the CSV data sent from the device and imports it into a data frame using the Pandas library, which converts the data into a format that can be used within the system.

[1379] Step 3:

[1380] The server preprocesses the received data, first by removing missing values, then by cleaning and normalizing the data, thereby ensuring data quality and improving the accuracy of the analysis.

[1381] Step 4:

[1382] The server divides the preprocessed data into explanatory variables (X) and objective variables (y). The explanatory variables contain various features related to urban planning, while the objective variables contain the values ​​to be predicted.

[1383] Step 5:

[1384] The server splits the data into training data and test data. Typically, 80% of the dataset is training data and 20% is test data. This split ensures a clear separation of the data used to train and evaluate the model.

[1385] Step 6:

[1386] The server sets up a random forest regression model. Random forests are models that use multiple decision trees to make predictions and have high predictive accuracy. At this stage, the necessary hyperparameters are also configured.

[1387] Step 7:

[1388] The server trains a random forest regression model using the training data, which allows the model to learn patterns and features to improve its predictive capabilities.

[1389] Step 8:

[1390] The server uses the trained model to make predictions on test data, thereby obtaining prediction results for evaluating the model's performance.

[1391] Step 9:

[1392] The server evaluates the prediction results, specifically calculating the mean squared error (MSE) and comparing the predicted values ​​with the actual values ​​to check the accuracy of the model.

[1393] Step 10:

[1394] The server visualizes the evaluation results, displaying them visually in the form of graphs and charts so that users can easily understand them.

[1395] Step 11:

[1396] The server sends the generated evaluation results and visualized graphs to the terminal, which receives them and prepares them for display to the user.

[1397] Step 12:

[1398] Users can evaluate urban plans by checking the evaluation results and visualized graphs displayed on their devices, allowing them to quickly and accurately grasp how effective the proposed urban plan is.

[1399] Step 13:

[1400] The device activates an emotion engine to recognize the user's emotions. The emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to identify emotions, using technologies such as facial recognition, voice analysis, and text analysis.

[1401] Step 14:

[1402] The server receives the recognized user's emotion data, including, for example, whether the user is stressed or happy.

[1403] Step 15:

[1404] The server adjusts the prediction results based on the user's emotional data, for example, providing detailed explanations and additional information if the user is stressed, or displaying concise results if the user is happy.

[1405] Step 16:

[1406] The server optimizes the visualization of the results according to the user's emotions. For example, if a positive emotion is detected, the graph will be displayed in brighter colors.

[1407] Step 17:

[1408] The device displays optimized prediction results and visualized graphs to users, improving the user experience and helping them make more effective decisions.

[1409] Example 2

[1410] 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."

[1411] Conventional urban planning support systems are complex in the processing, prediction, and evaluation of large amounts of urban planning data, making them difficult to execute efficiently. Furthermore, they are unable to provide feedback or adjust results based on user emotions, making it difficult to optimize results for easy user understanding. This makes it difficult to support fast and effective decision-making, and there is a need for improved user experience, particularly one that takes emotions into consideration.

[1412] 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.

[1413] In this invention, the server includes a means for acquiring data, a means for preprocessing, a means for dividing data into training data and test data, a means for learning data using a regression model, a means for making predictions using the test data, a means for evaluating the prediction results, a means for visualizing the evaluation results, a means for recognizing a user's emotion and adjusting the prediction results and display method based on the recognized emotion, and a means for transmitting data using a secure communication protocol, thereby enabling efficient processing of urban planning data and providing feedback according to the user's emotion.

[1414] The "means for acquiring data" is an input interface for users to upload urban planning data, which is then sent to the server.

[1415] "Preprocessing means" refers to automatically removing missing values ​​from the data, cleaning the data, normalizing the data, etc., and preparing the data in a format that can be analyzed.

[1416] A "means of splitting data into training and testing data" is a process of dividing a dataset into training and testing data in a certain proportion for the purpose of training and evaluating a model.

[1417] "Means of learning data using regression models" refers to machine learning models used to make predictions between data with relatively complex relationships, thereby enabling data analysis and predictions.

[1418] "Means for making predictions using test data" refers to the process of evaluating a trained model using a portion of the data to generate predictions.

[1419] The "means for evaluating the prediction results" refers to the use of evaluation indices (such as mean square error) to measure and judge the accuracy of the prediction results generated by the model.

[1420] "Means for visualizing evaluation results" refers to displaying prediction results and evaluation indicators in a visual format such as graphs or charts, making them easy for users to understand.

[1421] The "means for recognizing a user's emotions and adjusting the predicted results and display method based on the recognized emotions" refers to analyzing a user's emotions in real time using input devices such as a camera or microphone, and adapting the display format of information based on the results.

[1422] "Means of transmitting data using a secure communication protocol" refers to the communication means (e.g., HTTPS) used to securely transmit user data to the server, which encrypts and protects the data.

[1423] The present invention is an urban planning support system that combines a generative AI model and an emotion recognition engine, which efficiently processes complex data related to urban planning, performs predictions and evaluations, and optimizes the adjustment and visualization of results based on the user's emotions. The following describes in detail an embodiment of the present invention.

[1424] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices, which then transmit the data to the server using a secure communication protocol (e.g., HTTPS).

[1425] The server receives the data sent from the terminal and imports it into a data frame using the Pandas library. The imported data undergoes preprocessing such as removing missing values, cleaning the data, and normalizing the data. After preprocessing, the data is divided into explanatory variables (X) and target variables (y).

[1426] The server then splits the data into training and test data: 80% of the data is training data and 20% is testing data, providing an independent dataset for evaluating the model's training and predictive accuracy. The server then sets up a random forest regression model and trains the model using the training data.

[1427] After the model has completed training, the server uses the test data to make predictions. The prediction results are compared with the actual values ​​and evaluated using the mean squared error (MSE). The evaluation results are visualized in a visual format such as a graph or chart so that users can easily understand them.

[1428] This is where emotion recognition engines come in. Emotion recognition engines use technologies such as facial recognition, voice analysis, and text analysis to recognize a user's emotions. Specifically, they use cameras and microphones to analyze the user's facial expressions and tone of voice to identify emotions in real time. For example, if the user is satisfied, they can display a concise result, but if the user is stressed, they can provide a more detailed explanation or supplemental information.

[1429] The server adjusts the prediction results based on the user's recognized emotions and optimizes the visualization of the results. For example, if a positive emotion is recognized, the graph will be displayed in brighter colors. The concept is that dynamic feedback based on emotions improves the user experience.

[1430] As a concrete example, consider the case where a user proposes a new urban planning scenario and wants to evaluate its effectiveness. The user uploads the proposed urban planning data to the system from a terminal. Then, the user can enter the following prompt sentence:

[1431] "We will upload new urban planning data. The data includes population density, transportation infrastructure, and land use patterns."

[1432] The user's input data is sent to the server, where it undergoes a series of processes: preprocessing, modeling, prediction, and evaluation. Finally, an emotion recognition engine analyzes the user's reaction and adjusts the way the results are displayed. This allows users to quickly and accurately grasp the effects of urban planning and receive feedback that takes their emotions into consideration. This system will enable advanced planning support for the diverse challenges associated with Japan's rapid urbanization. It will also improve the user experience and support effective decision-making.

[1433] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1434] Step 1:

[1435] Users upload urban planning data (e.g., population density, transportation infrastructure, land use patterns, etc.) in CSV format from their own devices.

[1436] Input: City planning data in CSV format

[1437] Specifically, the user uses the system interface to select the target CSV file from the file selection dialog and clicks the upload button.

[1438] Output: HTTPS request from the terminal to the server

[1439] Step 2:

[1440] The terminal transmits data to the server using a secure communication protocol (eg, HTTPS).

[1441] Input: Urban planning data in CSV format uploaded by the user

[1442] Specifically, the device establishes secure communication using HTTPS and sends data to the server as a POST request.

[1443] Output: CSV data sent to the server

[1444] Step 3:

[1445] The server receives the data sent from the device and imports it into a data frame using the Pandas library.

[1446] Input: CSV data sent from the terminal

[1447] Specifically, the server reads the CSV file and converts it into a data frame using Pandas.

[1448] Output: Urban planning data in a data frame

[1449] Step 4:

[1450] The server performs preprocessing on the imported data, such as removing missing values, cleaning the data, and normalizing it.

[1451] Input: Dataframed urban planning data

[1452] Specifically, the server uses Pandas methods to remove missing values, clean outliers, and normalize each field.

[1453] Output: Preprocessed data frame

[1454] Step 5:

[1455] The server separates the preprocessed data into explanatory variables (X) and target variables (y).

[1456] Input: Preprocessed data frame

[1457] Specifically, the column containing the target variable is divided into y and the remaining columns into X.

[1458] Output: Dataset of explanatory and response variables

[1459] Step 6:

[1460] The server splits the data into training data (80%) and test data (20%).

[1461] Input: Dataset of explanatory and response variables

[1462] Specifically, the server splits the data using the train_test_split function of scikit-learn.

[1463] Output: A set of training and test data

[1464] Step 7:

[1465] The server sets up a random forest regression model and trains the model using the training data.

[1466] Input: training data

[1467] Specifically, the server instantiates a RandomForestRegressor from scikit-learn and trains the model using the fit method on the training data.

[1468] Output: A trained random forest model

[1469] Step 8:

[1470] The server uses the trained model to make predictions on the test data.

[1471] Input: Trained random forest model, test data

[1472] Specifically, the server uses the predict method on the test data to obtain the prediction results.

[1473] Output: Prediction results

[1474] Step 9:

[1475] The server compares the predicted results with the actual values ​​and calculates a performance metric (e.g., mean square error).

[1476] Input: Prediction results, actual values ​​of test data

[1477] Specifically, the server calculates the evaluation index using the mean_squared_error function of scikit-learn.

[1478] Output: Evaluation metric (e.g., mean squared error)

[1479] Step 10:

[1480] The server visualizes the evaluation results in visual formats such as graphs and charts.

[1481] Input: Evaluation Metrics

[1482] Specifically, the server visualizes the results using the Matplotlib library, drawing graphs and generating charts.

[1483] Output: Visualized evaluation results

[1484] Step 11:

[1485] The server activates an emotion recognition engine, analyzes the user's face and voice in real time via the camera and microphone, and identifies emotions.

[1486] Input: User's facial recognition and voice analysis data

[1487] Specifically, the server analyzes data acquired through the camera and microphone and recognizes emotions using facial recognition and voice analysis algorithms.

[1488] Output: Recognized user emotion data

[1489] Step 12:

[1490] The server adjusts how the prediction results are displayed based on the recognized user emotion.

[1491] Input: Recognized user emotion data, prediction results, evaluation results

[1492] Specifically, the server adds a detailed explanation if the user is feeling stressed, and displays the image in brighter colors if positive emotions are recognized.

[1493] Output: Display of sentiment-adjusted prediction and evaluation results

[1494] Step 13:

[1495] The server sends the final result to the terminal and displays it to the user.

[1496] Input: Emotion-adjusted predictions and evaluation results

[1497] Specifically, the server sends the results back to the terminal, which receives them and displays them to the user.

[1498] Output: The final result as it appears on your terminal.

[1499] (Application example 2)

[1500] 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."

[1501] In conventional brick-and-mortar store layout planning, store owners are faced with a huge amount of data, and processing that data and designing an optimal layout requires a great deal of time and effort. Furthermore, feedback and layout adjustments based on user needs and emotions are often not provided, resulting in insufficient optimization. This can have a negative impact on customer satisfaction and business performance. The present invention provides a system to solve these problems.

[1502] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1503] In this invention, the server includes means for acquiring data, means for preprocessing, means for dividing data into training data and test data, means for learning data using a regression model, means for making predictions using the test data, means for evaluating prediction results, means for visualizing the evaluation results, means for recognizing user emotions, means for adjusting prediction results based on emotions, and means for adjusting the display method of the results based on emotions, thereby enabling store owners to quickly and efficiently design optimal layouts and provide feedback that takes user emotions into consideration.

[1504] The "means for acquiring data" refers to a device or program for receiving data required by the system from outside.

[1505] The "preprocessing means" refers to a device or program that performs operations such as cleaning the acquired data, correcting missing values, and normalizing the data.

[1506] A "means for splitting data into training data and test data" is a device or program that splits a dataset appropriately for training and evaluating a model.

[1507] A "means for learning data using a regression model" is a device or program that executes a regression algorithm to build a predictive model from data.

[1508] A "means for making predictions using test data" is a device or program that uses a trained model to make predictions on new data sets.

[1509] A "means for evaluating prediction results" is a device or program that compares actual data with prediction results in order to evaluate the performance of a prediction model.

[1510] The "means for visualizing the evaluation results" is a device or program that displays the evaluation results in a visual format such as a graph or chart.

[1511] "Means for recognizing user emotions" refers to a device or program that uses a camera or microphone to analyze the user's facial expressions and tone of voice to identify emotions.

[1512] The "means for adjusting prediction results based on emotions" refers to a device or program that appropriately changes the prediction results or the content of the information provided based on the recognized emotions of the user.

[1513] The "means for adjusting the display method of results based on emotions" is a device or program that changes the displayed results or their format in response to the user's emotions.

[1514] This invention relates to a system for supporting layout optimization of brick-and-mortar stores. This system uses a generative AI model and an emotion recognition engine to propose optimal layouts based on data uploaded by users, and provides feedback according to the user's emotions.

[1515] System configuration

[1516] The present invention includes the following main means:

[1517] 1. How to obtain data

[1518] Data is received from the user terminal. This may be store data in CSV format, for example. The user terminal sends the data to the server using a secure communication protocol such as HTTPS.

[1519] 2. Pretreatment methods

[1520] The server uses the Pandas library to import the data, remove missing values, clean the data, and normalize it.

[1521] 3. A means of splitting the data into training and test data

[1522] Split the dataset into training and test data. A common practice is to use 80% of the data as training data and 20% as test data.

[1523] 4. How to train data using regression models

[1524] Set up a random forest regression model and train it using the training data.

[1525] 5. How to make predictions using test data

[1526] The model is used to make predictions on the test data and obtain the results.

[1527] 6. Means of evaluating prediction results

[1528] The accuracy of the prediction is evaluated using the mean squared error (MSE).

[1529] 7. Means of visualizing evaluation results

[1530] The evaluation results are displayed in visual formats such as graphs and charts. The Matplotlib library is used.

[1531] 8. How to Recognize User Emotions

[1532] It uses a camera and microphone to analyze the user's facial expressions and tone of voice to recognize emotions, and uses OpenCV and the SpeechRecognition library.

[1533] 9. How to adjust predictions based on emotions

[1534] Adjust predictions and information provided based on perceived emotions.

[1535] 10. A way to tailor how results are displayed based on sentiment

[1536] The color and level of detail of the results display are changed depending on the user's emotional state.

[1537] Program processing

[1538] The server uses the following infrastructure and software to process various data:

[1539] Pandas

[1540] Used for data preprocessing (cleaning, correction, normalization).

[1541] Scikit-Learn

[1542] Training and predicting with random forest regression models.

[1543] OpenCV

[1544] It is used to analyze camera images and recognize emotions from the user's facial expressions.

[1545] SpeechRecognition

[1546] Analyzes audio captured from a microphone to identify emotions.

[1547] Matplotlib

[1548] Prediction and evaluation results are displayed visually.

[1549] Specific examples

[1550] Suppose a store owner proposes a new product layout and wants to evaluate its effectiveness. The owner uploads store data saved in CSV file format to the app. The app receives this data, preprocesses it, and then uses a random forest regression model to learn and predict. The prediction results are evaluated and visualized in graphs and charts. When the user interacts with the system using the camera and microphone, the way the results are displayed is adjusted depending on their emotions. For example, if the user expresses positive emotions, the results are displayed in brighter colors.

[1551] Example prompt

[1552] "Please upload a CSV file of your new store layout. This data includes information such as customer movement patterns, sales performance, and product placement. Next, we will recognize your customers' real-time emotions, so please use your camera and microphone. We will then provide you with optimal feedback and layout evaluation results based on your emotions."

[1553] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1554] Step 1:

[1555] The user uploads store data (e.g., a CSV file containing customer movement patterns, sales performance, and product placement information) from the terminal.

[1556] Input: Store data in CSV format sent from the user's device.

[1557] Output: The raw CSV data sent to the server.

[1558] Step 2:

[1559] The server receives the data using a secure communication protocol such as HTTPS.

[1560] Input: The raw CSV data received.

[1561] Output: CSV data file stored on the server.

[1562] Step 3:

[1563] The server uses the Pandas library to import the CSV data into a data frame and preprocess the data (missing value correction, data cleaning, normalization).

[1564] Input: CSV format store data file.

[1565] Output: A preprocessed data frame.

[1566] Step 4:

[1567] Split the preprocessed data into training data and test data. Typically, 80% of the data is training data and 20% is test data.

[1568] Input: Preprocessed dataframe.

[1569] Output: Training dataset, Test dataset.

[1570] Step 5:

[1571] The server sets up a random forest regression model using the Scikit-Learn library and trains the model using the training data.

[1572] Input: Training dataset.

[1573] Output: A trained random forest regression model.

[1574] Step 6:

[1575] Using the trained model, the server makes predictions on the test data.

[1576] Input: Test dataset, trained random forest regression model.

[1577] Output: The predictions made by the model.

[1578] Step 7:

[1579] The server evaluates the accuracy of the prediction results using the mean square error (MSE).

[1580] Input: Actual test data, predicted outcome data.

[1581] Output: MSE value as evaluation metric.

[1582] Step 8:

[1583] To display the evaluation results in visual formats such as graphs and charts, we use the Matplotlib library.

[1584] Input: Evaluation metrics and prediction results.

[1585] Output: A visually displayed graph or chart.

[1586] Step 9:

[1587] The user interacts with the system using a camera and microphone, and the server recognizes the user's emotions using OpenCV and the SpeechRecognition library.

[1588] Input: Real-time captured video and audio data.

[1589] Output: The perceived emotional state of the user.

[1590] Step 10:

[1591] The server adjusts the prediction results and the content of the information it provides based on the recognized user's emotions.

[1592] Input: Prediction results, user sentiment data.

[1593] Output: Prediction results adjusted based on sentiment.

[1594] Step 11:

[1595] The server adjusts the color tone and detail of the result display based on the emotion and shows the final result to the user.

[1596] Input: Tuned prediction results, sentiment data.

[1597] Output: The final result with adjusted color and detail.

[1598] These steps can effectively optimize the layout of physical stores and provide feedback based on user sentiment.

[1599] 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.

[1600] 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.

[1601] 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.

[1602] 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.

[1603] 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.

[1604] 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.

[1605] 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).

[1606] 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.

[1607] 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."

[1608] 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.

[1609] 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).

[1610] 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.

[1611] 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.

[1612] 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.

[1613] 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.

[1614] 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.

[1615] 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.

[1616] 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.

[1617] 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.

[1618] 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.

[1619] 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.

[1620] The following is further disclosed regarding the above embodiment.

[1621] (Claim 1)

[1622] A means of obtaining the data;

[1623] A means for performing pre-processing;

[1624] a means for splitting the data into training data and test data;

[1625] a means for learning data using a regression model;

[1626] a means of making predictions using test data;

[1627] a means for evaluating the prediction results;

[1628] A system including a means for visualizing evaluation results.

[1629] (Claim 2)

[1630] 10. The system of claim 1, wherein the regression model is a random forest regression model.

[1631] (Claim 3)

[1632] 10. The system of claim 1, further comprising means for separating explanatory and response variables of the data.

[1633] "Example 1"

[1634] (Claim 1)

[1635] a means for a user to upload data from a terminal;

[1636] A means for the server to receive and import the data into a data frame;

[1637] a means for the server to pre-process the data;

[1638] A means for the server to separate the data into explanatory variables and target variables,

[1639] a means for the server to split the data into training data and test data;

[1640] a means for the server to learn data using a random forest regression model;

[1641] a means for the server to make predictions using the test data;

[1642] a means for the server to evaluate the prediction results;

[1643] The system includes a means for visualizing the evaluation results and transmitting the results to a terminal.

[1644] (Claim 2)

[1645] 10. The system of claim 1, wherein the random forest regression model makes predictions using multiple decision trees.

[1646] (Claim 3)

[1647] 10. The system of claim 1, further comprising means for transmitting the data in a secure communication protocol.

[1648] "Application Example 1"

[1649] (Claim 1)

[1650] A means of obtaining the data;

[1651] A means for performing pre-processing;

[1652] a means for splitting the data into training data and test data;

[1653] a means for learning data using a regression model;

[1654] a means of making predictions using test data;

[1655] a means for evaluating the prediction results;

[1656] A means for visualizing the evaluation results;

[1657] A means to acquire robot operation data and propose optimal operation schedules and placements,

[1658] A system including a means for providing the proposal results via a smart device.

[1659] (Claim 2)

[1660] 10. The system of claim 1, wherein the regression model is a random forest regression model.

[1661] (Claim 3)

[1662] 10. The system of claim 1, further comprising means for separating explanatory and response variables of the data.

[1663] "Example 2: Combining Emotion Engines"

[1664] (Claim 1)

[1665] A means of obtaining the data;

[1666] A means for performing pre-processing;

[1667] a means for splitting the data into training data and test data;

[1668] a means for learning data using a regression model;

[1669] a means of making predictions using test data;

[1670] a means for evaluating the prediction results;

[1671] A means for visualizing the evaluation results;

[1672] means for recognizing a user's emotion and adjusting the prediction results and display method based on the recognized emotion;

[1673] A system including means for transmitting data using a secure communication protocol.

[1674] (Claim 2)

[1675] 10. The system of claim 1, wherein the regression model is a random forest regression model.

[1676] (Claim 3)

[1677] 10. The system of claim 1, further comprising means for separating explanatory and response variables of the data.

[1678] "Application example 2 when combining emotion engines"

[1679] (Claim 1)

[1680] A means of obtaining the data;

[1681] A means for performing pre-processing;

[1682] a means for splitting the data into training data and test data;

[1683] a means for learning data using a regression model;

[1684] a means of making predictions using test data;

[1685] a means for evaluating the prediction results;

[1686] A means for visualizing the evaluation results;

[1687] a means of recognizing a user's emotions;

[1688] a means for adjusting the predicted outcome based on the emotion;

[1689] The system includes a means for adjusting how results are displayed based on emotion.

[1690] (Claim 2)

[1691] 10. The system of claim 1, wherein the regression model is a random forest regression model.

[1692] (Claim 3)

[1693] 10. The system of claim 1, further comprising means for separating explanatory and response variables of the data. [Explanation of symbols]

[1694] 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 of obtaining the data; A means for performing pre-processing; a means for splitting the data into training data and test data; a means for learning data using a regression model; a means of making predictions using test data; a means for evaluating the prediction results; A system including a means for visualizing evaluation results.

2. The system of claim 1 , wherein the regression model is a random forest regression model.

3. The system of claim 1 further comprising means for separating explanatory and response variables of the data.

Images