system

Abstract

Provide a system. [Solution] a means for users to upload data files; a means for the server to store the received data in an internal database; a means for the server to automatically perform data integrity checking and preprocessing; A means for the server to apply multiple machine learning algorithms and select the most suitable model; a means for the server to perform prediction or classification using the selected model; A means for the server to visualize the results and display them on a dashboard or output them in a file format; A system including:

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] This is the section of tEase's patent specification that deals with "the problem that the invention aims to solve" and "the means for solving the problem."

[0005] Data-driven decision-making has become increasingly important in recent years. However, many companies and individuals lack the expertise in data science, making it difficult to perform complex data analysis. As a result, even with useful data, they are unable to make appropriate predictions or classifications, which can significantly delay decision-making and strategy formulation. Furthermore, current data analysis tools are often not user-friendly, and users continue to rely on experts with specific knowledge. The purpose of this invention is to solve these problems and provide a system that allows even users without data science expertise to easily perform advanced data analysis. [Means for solving the problem]

[0006] The present invention relates to a system that includes a means for users to upload data files, a means for a server to store the received data in an internal database, a means for the server to automatically perform data integrity checks and preprocessing, a means for the server to apply multiple machine learning algorithms to select the most suitable model, a means for the server to perform prediction or classification using the selected model, and a means for the server to visualize the results and display them on a dashboard or output them in file format. Specifically, the server performs hyperparameter tuning to build an optimal model, encodes categorical variables in the data, and normalizes or standardizes numerical variables, enabling highly accurate prediction and classification. This allows users to easily perform data analysis without specialized knowledge and make quick and accurate decisions based on the results.

[0007] "User" refers to any individual or organization that uploads data files to the PredictEase system and uses the results.

[0008] "Data File" refers to data that users upload to the PredictEase system, such as CSV or Excel files.

[0009] "Server" refers to the central processing unit of the PredictEase system, which receives, stores, pre-processes, selects models, predicts, classifies, and outputs results from data.

[0010] "Internal Database" refers to the storage device or system used by the Server to store data files.

[0011] "Integrity check" refers to the process by which the server detects missing or outliers in the data and checks the quality of the data.

[0012] "Preprocessing" refers to the processing that the server performs to convert data into an analyzable format, specifically tasks such as missing value processing, outlier processing, normalization, standardization, and encoding.

[0013] "Machine learning algorithms" refer to the statistical and computational methods that the server uses to build predictive models based on data.

[0014] "Hyperparameter tuning" refers to the process of adjusting parameters to optimize the performance of a machine learning model.

[0015] "Prediction" refers to the server predicting future data based on past data.

[0016] "Classification" refers to the server classifying data into different categories or classes.

[0017] "Visualization" refers to the server displaying prediction or classification results in a format that is easy for users to understand, such as a graph or table.

[0018] A "dashboard" refers to an interface that allows users to visually check prediction or classification results.

[0019] "Output in file format" means providing the results to the user in a format that can be downloaded (e.g., CSV or Excel file). [Brief explanation of the drawings]

[0020] [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

[0021] 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.

[0022] First, the terms used in the following description will be explained.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 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.

[0031] 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).

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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."

[0041] This invention relates to PredictEase, a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. Specifically, the system is composed of the following steps:

[0042] System Overview

[0043] Data Entry

[0044] 1. A user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[0045] 2. The server receives the uploaded data file and stores it in its internal database.

[0046] Data Preprocessing

[0047] 1. The server performs a consistency check on the data file to determine if there are any missing or outlier values.

[0048] 2. The server handles missing and outliers in an appropriate way and converts the data into a form that can be applied to the machine learning model.

[0049] 3. The server encodes categorical variables and normalizes or standardizes numeric variables.

[0050] Model Selection

[0051] 1. The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[0052] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[0053] Model Optimization

[0054] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[0055] Prediction and Classification

[0056] 1. The server uses the trained model to make a prediction or classification.

[0057] 2. The server aggregates the prediction and classification results and evaluates their performance.

[0058] Output of results

[0059] 1. The server visually displays prediction and classification results on a dashboard.

[0060] 2. Display the results in an interface so that users can view them in graphs and tables.

[0061] 3. The server makes the results available for download as a CSV or Excel file, if desired.

[0062] Specific examples

[0063] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[0064] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[0065] 2. The server stores the received sales data in an internal database and starts checking the data for consistency. For example, if there are missing values ​​in the sales data, it fills them with the average value and handles any outliers appropriately.

[0066] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[0067] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[0068] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[0069] 6. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[0070] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[0071] This makes it easy for users without data science knowledge to perform advanced data analysis and predictions, making PredictEase a powerful tool for supporting data-driven decision-making.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[0075] Step 2:

[0076] The server receives the uploaded data file, checks whether it is in the correct format, and then stores the data file in its internal database.

[0077] Step 3:

[0078] The server performs data integrity checks, specifically detecting missing and outlier values ​​and handling them in an appropriate way (e.g., imputing with the mean, removing outliers).

[0079] Step 4:

[0080] The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables to convert the data into a format that can be applied to machine learning models.

[0081] Step 5:

[0082] The server analyzes the features in the dataset and evaluates correlations and the importance of variables, thereby selecting important features and eliminating unnecessary features.

[0083] Step 6:

[0084] The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) to train each model.

[0085] Step 7:

[0086] The server evaluates the performance of each model using cross-validation, calculates evaluation metrics such as accuracy and F1 score for each model, and selects the optimal model.

[0087] Step 8:

[0088] The server tunes the hyperparameters of the selected model, using grid search or random search to find the optimal hyperparameters and build the final version of the model.

[0089] Step 9:

[0090] The server performs prediction or classification using the prepared test data, and then aggregates the prediction and classification results to evaluate their performance.

[0091] Step 10:

[0092] The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[0093] Step 11:

[0094] Users access the dashboard and view prediction and classification results displayed in graphs and tables.

[0095] Step 12:

[0096] If desired, users can download the results as a CSV or Excel file.

[0097] Through this series of steps, PredictEase enables users to easily perform highly accurate data analysis and predictions, even without data science knowledge.

[0098] Example 1

[0099] 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."

[0100] Current data analysis and prediction systems are difficult to use effectively unless users have specialized data science knowledge. They also require complex procedures such as data preprocessing, optimal model selection, and hyperparameter tuning, so there is a need for a system that can perform these processes consistently and automatically. Furthermore, there is a need for a system that can easily visualize and output the results.

[0101] 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.

[0102] In this invention, the server includes means for users to upload data files, means for the computer to store the received data in an internal storage device, means for the computer to automatically check the consistency of the data and preprocess it, means for the computer to apply multiple machine learning algorithms and select the most suitable model, means for the computer to make predictions or classifications using the selected model, means for the computer to visualize the results and display them on a display device or output them in file format, and means for users to check the results and make them downloadable. This allows even users without specialized knowledge to easily and efficiently perform processes from uploading data to obtaining prediction results.

[0103] "User" refers to a person who uses this system, uploads data, and obtains analysis and prediction results.

[0104] A "data file" refers to a collection of digital data, such as CSV or Excel format, that is input into a system for analysis or prediction.

[0105] "Computer" refers to a collection of hardware and software for receiving, processing, analyzing data, and outputting results.

[0106] "Internal storage device" refers to a storage medium that a computer uses to temporarily or permanently store data files.

[0107] "Integrity check" refers to the process of verifying that a data file conforms to the expected format and structure and that there are no missing or outlier values.

[0108] "Preprocessing" refers to a series of steps that transform, filter, and encode data to improve the accuracy of analysis or prediction.

[0109] "Machine learning algorithms" refer to mathematical models and methods that automatically learn patterns and knowledge from data and make predictions and classifications.

[0110] A "model" is a computational system trained by machine learning algorithms that has the ability to make predictions or classifications based on input data.

[0111] "Hyperparameter tuning" refers to the process of adjusting the configuration parameters of an algorithm to optimize the performance of a machine learning model.

[0112] "Visualization" refers to the process of representing prediction and classification results as graphs or charts so that they can be understood intuitively.

[0113] "Display device" refers to hardware such as a monitor or display that presents visualized results to a user.

[0114] "Downloading" refers to the act of a user transferring and saving a data file of prediction or classification results from the system to their own device.

[0115] This invention relates to a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. In this system, users upload data files, and the server automatically preprocesses the data, selects and optimizes models, and predicts and visualizes the results.

[0116] System Overview

[0117] Data Entry

[0118] The process begins when a user accesses the web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The server receives the uploaded data file and stores it in its internal storage.

[0119] Data Preprocessing

[0120] The server performs a consistency check on the data files to check for missing or outliers. It then imputes missing values ​​with the mean and handles outliers appropriately. It also encodes categorical variables and normalizes or standardizes numerical variables. This process is important as a preprocessing step for prediction and classification.

[0121] Model Selection and Optimization

[0122] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It evaluates the performance of each model using cross-validation and selects the best model. It then optimizes the hyperparameters of the selected model using grid search or random search. This step improves the model's predictive accuracy.

[0123] Prediction and visualization

[0124] The server then uses the best trained model to make predictions and classifications for new data. The results are visualized and displayed on a dashboard, presented to users in the form of graphs and tables. Users can also download the results as CSV or Excel files for offline analysis.

[0125] Specific examples

[0126] For example, if a user in a company's planning department wants to forecast monthly sales data, they can use the system as follows:

[0127] 1. The user prepares past sales data in CSV file format and accesses the web application to upload the file.

[0128] 2. The server stores the received sales data in its internal storage device and starts checking the data for consistency. If missing values ​​are found, they are filled in with the average value, and any outliers are dealt with appropriately.

[0129] 3. The server preprocesses the data, encoding categorical variables and normalizing numerical variables.

[0130] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[0131] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[0132] 6. The server uses the optimal model to predict sales for the next month and displays the forecast results on the dashboard.

[0133] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[0134] Prompt Sentence Examples

[0135] "Use the following sales data to predict next month's sales. Use linear regression and random forest as your models. Impute missing values ​​with the mean and treat outliers appropriately."

[0136] In this way, the system efficiently processes everything from data input to output, providing users with highly accurate prediction results. This process eliminates the need for specialized knowledge, resulting in a system that is convenient and user-friendly.

[0137] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0138] Step 1:

[0139] A user accesses the PredictEase web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The input is the data file to be analyzed, and the output is the uploaded data file sent to the server.

[0140] Step 2:

[0141] The server receives the uploaded data file and saves it to its internal storage. The data file "sales_data.csv" is saved to the server. The input is the uploaded data file, and the output is saving the data file to its internal storage.

[0142] Step 3:

[0143] The server performs a consistency check on the data file. Specifically, it checks whether the number of columns and data types match, and checks for missing or outlier values. The input is the saved data file, and the output is the data consistency check result.

[0144] Step 4:

[0145] The server imputes missing values ​​with the mean value and filters or imputes outliers, if any. The input is data that has been checked for consistency, and the output is preprocessed data.

[0146] Step 5:

[0147] The server encodes categorical variables into numeric values ​​and normalizes or standardizes numeric variables. For example, categorical variables such as "store name" are one-hot encoded, and numeric variables such as "sales amount" are processed with min-max scaling. The input is the preprocessed data, and the output is the encoded and normalized data.

[0148] Step 6:

[0149] The server trains the model by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.). The input is the encoded and normalized data, and the output is a model trained with each algorithm.

[0150] Step 7:

[0151] The server evaluates the performance of each model using cross-validation and selects the best model. For example, cross-validation evaluates the mean squared error (MSE) or coefficient of determination (R²). The input is the trained model, and the output is the best model.

[0152] Step 8:

[0153] The server optimizes the hyperparameters of the selected optimal model using grid search or random search. The input is the optimal model, and the output is the model with optimized parameters.

[0154] Step 9:

[0155] The server uses the best-fit model it has trained to make predictions or classifications for new data. The input is the new data, and the output is the prediction or classification result.

[0156] Step 10:

[0157] The server aggregates the prediction and classification results and calculates evaluation metrics such as accuracy and precision. The input is the prediction and classification results, and the output is a performance evaluation report.

[0158] Step 11:

[0159] The server visualizes the prediction and classification results and displays them on a dashboard. Users can check the results displayed in graphs and tables on the dashboard. The input is a performance evaluation report, and the output is a visualized dashboard display.

[0160] Step 12:

[0161] The server optionally makes the results available for download as CSV or Excel files, which users can download and analyze offline. The input is the visualized dashboard data, and the output is a downloadable file.

[0162] (Application example 1)

[0163] 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."

[0164] For modern self-driving vehicles, it is essential to analyze massive amounts of operational data in real time to detect and predict anomalies. However, this requires advanced knowledge of data science, which is a significant hurdle for non-expert users. Additionally, the pre-processing and modeling of operational data is complex and difficult to perform efficiently. There is a need for a system that can solve these problems and enable anyone to easily analyze data from self-driving vehicles.

[0165] 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.

[0166] In this invention, the server includes means for users to upload data files, means for the server to store the received data in an internal database, means for the server to automatically perform data consistency checks and preprocessing, means for the server to apply multiple machine learning algorithms and select the most appropriate model, means for the server to perform prediction or classification using the selected model, means for the server to visualize the results and display them on a dashboard or output them in file format, means for the server to analyze operational data of autonomous vehicles via an application installed on a smartphone and perform anomaly detection and prediction, and means for storing the trained model in the smartphone. This enables even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and prediction in real time.

[0167] "Means for users to upload data files" refers to an interface through which users submit files containing autonomous vehicle operation data to the web application.

[0168] "Means for storing the data received by the server in an internal database" refers to a processing device and storage technology for storing the received operation data in a database within the server.

[0169] "Means for the server to automatically perform data consistency checks and preprocessing" refers to technology in which the server checks operational data for missing values, abnormal values, etc., and automatically performs the necessary preprocessing (e.g., filling in missing values, cleaning data).

[0170] "A means by which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a technology in which the server tries out multiple models, evaluates their performance, and selects the model with the highest predictive accuracy.

[0171] "Means for the server to make predictions or classifications using the selected model" refers to a technology for making predictions or classifications for new operational data using the selected optimal model.

[0172] "Means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to the function by which the server displays the prediction or classification results in a visually easy-to-understand dashboard and outputs them in a file format (e.g., CSV, Excel) if necessary.

[0173] "Means of analyzing driving data of autonomous vehicles via an application installed on a smartphone to detect and predict abnormalities" refers to technology that analyzes driving data via a dedicated application on a smartphone, detects abnormal driving patterns, and predicts future driving conditions.

[0174] "Means for storing trained models within a smartphone" refers to technology that stores pre-trained machine learning models in a smartphone's storage device and makes them accessible to applications.

[0175] This invention relates to a system that enables users without specialized data science knowledge to easily analyze operational data of autonomous vehicles and perform anomaly detection and prediction. Specifically, the system is configured with the following steps.

[0176] System Overview

[0177] 1. Data Entry

[0178] Users upload autonomous vehicle operation data files (e.g., CSV or Excel files) via an application installed on their smartphones, and the uploaded data is sent to a server.

[0179] 2. Receipt and storage of data

[0180] The server stores the received operational data in an internal database, where it is quickly and securely stored and used for subsequent data analysis processes.

[0181] 3. Data integrity check and preprocessing

[0182] The server automatically performs integrity checks on the stored data, checking for missing or outlier values, and performs appropriate pre-processing, including imputing missing values ​​and cleaning the data.

[0183] 4. Model Selection

[0184] The server tries multiple machine learning algorithms and selects the best model, such as a random forest or a support vector machine. The selected model is then evaluated using cross-validation.

[0185] 5. Optimizing and training the model

[0186] The server then adjusts the hyperparameters of the selected model to achieve optimal performance, and the trained model is stored on the smartphone's memory.

[0187] 6. Prediction and Anomaly Detection

[0188] The selected model is used to predict and classify new driving data from the autonomous vehicle. The smartphone receives the analysis results and predicts abnormal driving patterns and future driving situations.

[0189] 7. Visualizing and outputting results

[0190] The server visually displays prediction and classification results on a dashboard. Users can view the results in graphs and tables through a smartphone interface. Users can also download the results in file format (e.g., CSV or Excel) if desired.

[0191] This allows even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and future predictions in real time.

[0192] Hardware and software used

[0193] Smartphone: Install the application and use it to upload data and display analysis results.

[0194] Server: Receives data, stores it, checks its integrity, preprocesses it, and selects, trains, and optimizes the model.

[0195] Python: A programming language used for data analysis and model training.

[0196] Pandas: A library used to load and preprocess data.

[0197] Scikit-learn: A library used to implement machine learning algorithms.

[0198] Joblib: A library used to save trained models.

[0199] Specific examples

[0200] For example, if a user wants to upload operational data (speed, GPS information, sensor information, etc.) from an autonomous vehicle and detect abnormal driving patterns, they can use the PredictDrive system via a smartphone application. The server receives the data, performs preprocessing, selects the optimal machine learning model, and trains it. The trained model is stored on the smartphone, and when new operational data is entered, it detects abnormalities in real time and displays the prediction results on a dashboard.

[0201] Prompt Sentence Examples

[0202] "Please tell us your next use case: Create a Python script that uses past driving data to predict whether the next driving pattern will be anomalous. The driving data includes speed, GPS information, and various sensor information. Use a random forest as the model, and create a visualization plot to present the results. Include detailed missing value handling and data standardization methods."

[0203] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0204] Step 1:

[0205] The user uploads the autonomous vehicle's operation data file (e.g., CSV format) using an application installed on their smartphone. This input data includes speed, GPS information, and sensor information. The data is then sent to the server via the smartphone.

[0206] Step 2:

[0207] The server stores the received operational data in an internal database. The stored data is used for subsequent processing within the server. The input here is the uploaded operational data, and the output is the data stored in the database.

[0208] Step 3:

[0209] The server performs consistency checks on the stored data. Specifically, it checks for missing values ​​and outliers and performs data cleaning. For example, missing values ​​are imputed with the mean value. The input of this step is the stored data, and the output is the cleaned data.

[0210] Step 4:

[0211] The server preprocesses the data by encoding categorical variables and normalizing or standardizing numerical variables. Encoding converts categorical data into a numerical representation, and normalization is achieved by scaling the data range. The input is the cleaned data, and the output is the preprocessed data.

[0212] Step 5:

[0213] The server tries multiple machine learning algorithms and applies and trains models such as random forests and support vector machines. It evaluates the performance of each model through cross-validation and selects the most suitable model. The input is the preprocessed data, and the output is the selected optimal model.

[0214] Step 6:

[0215] The server adjusts the hyperparameters of the selected model and tunes it to achieve optimal performance. This process involves an iterative process to optimize the model parameters. The input is the selected model, and the output is the optimized model.

[0216] Step 7:

[0217] The trained optimization model is stored locally on the smartphone, which is then used for real-time prediction and classification. The input here is the optimization model, and the output is the model stored locally on the smartphone.

[0218] Step 8:

[0219] The smartphone uses the stored model to make predictions and detect anomalies in real time whenever new driving data is input. For example, if an abnormal driving pattern is detected, an alert is sent to the user. The input is new driving data, and the output is the prediction or anomaly detection result and a user notification.

[0220] Step 9:

[0221] The server visually displays the prediction and classification results on a dashboard. Users can check the results on their smartphones and, if necessary, download the results in file format (e.g., CSV or Excel). The input here is the prediction or classification results, and the output is the visualized dashboard and file output.

[0222] 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.

[0223] This invention combines the PredictEase system, which performs data prediction and classification, with an emotion engine that recognizes user emotions. This system is designed as a no-code tool that can be easily used by users without data science expertise, and provides appropriate feedback and preprocessing adjustments according to the user's emotional state.

[0224] System Overview

[0225] Data Entry

[0226] 1. The user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[0227] 2. The server receives the uploaded data file, checks its format, and then stores it in its internal database.

[0228] Data Preprocessing

[0229] 1. The server performs data integrity checks to detect missing or outlier values ​​and handles them in an appropriate way.

[0230] 2. The server encodes categorical variables and normalizes or standardizes numeric variables.

[0231] Using the Emotion Engine

[0232] 1. The server passes the user's input and actions on the interface to the emotion engine to recognize the user's emotions.

[0233] 2. The server uses the emotion engine to adjust data preprocessing and model selection strategies according to the user's emotional state. For example, if the user is impatient, it can shorten the preprocessing time or provide easy-to-understand feedback to the user.

[0234] Model Selection

[0235] 1. The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[0236] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[0237] Model Optimization

[0238] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[0239] Prediction and Classification

[0240] 1. The server uses the trained model to make a prediction or classification.

[0241] 2. The server aggregates the prediction and classification results and evaluates their performance.

[0242] Output of results

[0243] 1. The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[0244] 2. The user accesses the dashboard and checks the prediction and classification results displayed in graphs and tables.

[0245] 3. If desired, the user can download the results as a CSV or Excel file.

[0246] Specific examples

[0247] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[0248] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[0249] 2. The server stores the received sales data in an internal database and performs data consistency checks, such as filling in missing values ​​with the average value and handling outliers appropriately.

[0250] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[0251] 4. The server uses an emotion engine to recognize the user's emotional state based on their input and actions. For example, if the server determines that the user is impatient, it summarizes the processing results and provides quick feedback.

[0252] 5. The server applies multiple machine learning algorithms to train each model, for example, linear regression, random forest, decision tree, etc.

[0253] 6. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[0254] 7. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[0255] 8. The user can check the forecast results displayed in graphs and tables on the dashboard and download the results in CSV format if necessary.

[0256] This allows PredictEase to provide advanced data analysis and predictions while taking into account the user's emotional state. Even users without data science knowledge can use this system to obtain highly accurate data analysis results and make strategic decisions.

[0257] The processing flow will be explained below.

[0258] Step 1:

[0259] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[0260] Step 2:

[0261] The server receives the uploaded data file, checks whether it is in the correct format, and then stores it in the internal database. If the format is incorrect, an error message is displayed to the user.

[0262] Step 3:

[0263] The server checks the integrity of the data file, detecting missing and abnormal values ​​(e.g., extreme values ​​and outliers) and performing other processing such as filling in the missing values ​​with the mean.

[0264] Step 4:

[0265] The server performs data preprocessing, such as label encoding and one-hot encoding for categorical variables and standardization and normalization for numerical variables, to convert the data into a form suitable for use in machine learning models.

[0266] Step 5:

[0267] The server uses an emotion engine to recognize the user's emotional state, for example by analyzing the video captured by the user's webcam, mouse movements, typing speed, etc.

[0268] Step 6:

[0269] The server adjusts its data preprocessing and model selection strategies based on the results of the emotion engine: for example, if the user is impatient, it will shorten the preprocessing time and present the results in an easy-to-understand format.

[0270] Step 7:

[0271] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It splits the data into a training dataset and a test dataset and proceeds with learning for each model.

[0272] Step 8:

[0273] The server performs cross-validation and evaluates the performance of each model, calculating evaluation metrics such as accuracy, F1 score, and R^2 value, and selecting the best model.

[0274] Step 9:

[0275] The server performs hyperparameter tuning for the selected model, for example, using grid search or random search to find the optimal parameter set.

[0276] Step 10:

[0277] The server uses the optimized model to make predictions and classifications, and performs final evaluation of the model using test data.

[0278] Step 11:

[0279] The server visualizes the prediction and classification results, converting them into a format that can be displayed in graphs and tables on a dashboard.

[0280] Step 12:

[0281] Users can access the dashboard to see the results and can click on graphs and tables to access more detailed information.

[0282] Step 13:

[0283] If desired, users can download the results as CSV or Excel files, which can then be saved for further analysis and reporting.

[0284] Through these specific processing steps, PredictEase incorporates the user's emotional state while automating advanced data analysis, enabling highly accurate prediction and classification without specialized knowledge.

[0285] Example 2

[0286] 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."

[0287] Conventional data analysis systems are difficult for users without specialized knowledge to use, and require a lot of time and effort for data preprocessing and machine learning model selection. Furthermore, the systems operate without taking into account the user's emotional state, which can be stressful for users. Furthermore, it is difficult to perform highly accurate predictions and classifications while maintaining data quality.

[0288] 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.

[0289] In this invention, the server includes: [means for a user to upload a data file to be analyzed;] [means for the server to store the uploaded data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model; and [means for the server to visualize the results of prediction or classification and display them on a dashboard or output them in file format.] This makes it possible to easily perform highly accurate data analysis and prediction while taking into account the user's emotional state.

[0290] The "means for users to upload data files to be analyzed" refers to an interface that allows users to submit data files saved in formats such as CSV or Excel to the web application.

[0291] "Means by which the server stores uploaded data in its internal database" refers to the process by which the server receives data files uploaded by users, organizes and verifies the data, and stores it in its internal database.

[0292] "Means for the server to automatically perform data consistency checks and preprocessing" refers to a function that enables the server to detect missing or outlier values ​​in the data, complete or delete them, and automatically perform data preprocessing such as encoding categorical variables and normalizing numerical variables.

[0293] "Means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection" refers to the process in which the server analyzes the user's interactions such as keyboard input and mouse clicks based on an emotion engine, recognizes the user's emotional state, and dynamically adjusts the data preprocessing and machine learning model selection method based on this.

[0294] "Means in which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a method in which the server applies multiple machine learning algorithms, such as linear regression, random forest, decision tree, and support vector machine, trains and evaluates each model, and then selects the most appropriate model.

[0295] The "means for the server to make predictions or classifications using the selected model" is a function that makes predictions or classifications for new data using the optimal machine learning model that the server has previously selected.

[0296] "Means for the server to visualize the prediction or classification results and display them on a dashboard or output them in a file format" means the ability of the server to visualize the prediction or classification results obtained by the machine learning model in graph or table format and display them on a dashboard that can be accessed by the user, or output them in a file format such as CSV or Excel.

[0297] MODE FOR CARRYING OUT THE INVENTION

[0298] This invention relates to a system that streamlines data analysis, recognizes user emotions, and adjusts feedback. This system is based on the no-code data analysis tool "PredictEase," and can be used intuitively even by beginners.

[0299] In an example, a user first accesses the PredictEase web application through a web browser and uploads the data file to be analyzed, which can be in CSV or Excel format.

[0300] The server receives, validates, and stores data using a backend built in Python. Specifically, the received data is checked for consistency using the pandas library. Missing values ​​and outliers are detected and then appropriately imputed or deleted. This ensures the quality of the data.

[0301] Next, the data is preprocessed using scikit-learn: categorical variables are one-hot encoded and numerical variables are normalized or standardized. This preprocessing enables efficient and effective training with multiple machine learning algorithms.

[0302] Furthermore, the system incorporates an emotion engine that recognizes emotions by analyzing user behavior on the interface. For example, a user who frequently sees error messages can be determined to be impatient. Based on the analysis results of the emotion engine, the server dynamically adjusts data preprocessing and model selection strategies. This process reduces user stress and provides a comfortable user experience.

[0303] For model selection, the server uses scikit-learn or TENSORFLOW (registered trademark) to apply multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains models for each algorithm. It evaluates the performance of each model using cross-validation and selects the best model. This training and evaluation process maximizes the accuracy of predictions and classifications.

[0304] After the best model is selected, hyperparameter tuning is performed using GridSearchCV to build the best performing model. The server then uses this optimized model to make predictions or classifications on new data.

[0305] Finally, the server visualizes the prediction or classification results and outputs them in a dashboard format that is easy for users to understand, using visualization tools such as matplotlib and Plotly. Users can check the results through the dashboard and download them in CSV or Excel format if necessary.

[0306] Specific examples

[0307] For example, if a marketing manager at a company wants to use monthly sales data to predict next month's sales, he or she can use PredictEase as follows:

[0308] 1. The user prepares past sales data in CSV format and accesses the PredictEase web application to upload it.

[0309] 2. The server receives the data and uses the pandas library to check the data for consistency, impute missing values, and handle outliers appropriately.

[0310] 3. Use scikit-learn to perform data preprocessing, one-hot encoding categorical variables, and normalizing or standardizing numerical variables.

[0311] 4. The server uses an emotion engine to recognize the user's emotional state and adjusts pre-processing and feedback strategies according to the user's level of impatience.

[0312] 5. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow and selects the best model using cross-validation.

[0313] 6. Use GridSearchCV to tune the hyperparameters of the selected model and build the optimal model.

[0314] 7. The server uses this optimal model to predict next month's sales, visualizes the prediction results, and displays them on a dashboard.

[0315] 8. Users can check the prediction results on the dashboard and download them in CSV format if necessary.

[0316] This allows PredictEase to significantly reduce the effort required for data analysis and provide highly accurate predictions while taking into account the user's emotional state.

[0317] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0318] PredictEase System Program Processing Steps

[0319] Step 1: Upload your data files

[0320] Input: User-uploaded data file in CSV or Excel format.

[0321] Specific behavior:

[0322] 1. A user accesses the web application and selects a data file.

[0323] 2. The user clicks the "Upload" button.

[0324] Data processing: None at this time.

[0325] Output: The uploaded file is sent to the server.

[0326] Step 2: Receiving and storing data

[0327] Input: User uploaded data file.

[0328] Specific behavior:

[0329] 1. The server receives the uploaded file.

[0330] 2. The server checks the file format and stores it in its internal database.

[0331] Data processing: Check the format of the received file.

[0332] Output: The path of the data file stored in the database.

[0333] Step 3: Data integrity check and preprocessing

[0334] Input: Data files stored in the internal database.

[0335] Specific behavior:

[0336] 1. The server uses the pandas library to check the integrity of the data.

[0337] 2. The server detects missing values ​​and fills them with DataFrame.fillna().

[0338] 3. The server detects outliers and handles them appropriately.

[0339] Data processing: missing value imputation, outlier treatment.

[0340] Output: Consistency checked and preprocessed data.

[0341] Step 4: Data Encoding and Normalization

[0342] Input: Data that has been integrity checked and preprocessed.

[0343] Specific behavior:

[0344] 1. The server encodes categorical variables using scikit-learn's OneHotEncoder.

[0345] 2. The server normalizes the numerical variables using scikit-learn's StandardScaler.

[0346] Data processing: categorical data encoding, numerical data normalization.

[0347] Output: The dataset after encoding and normalization.

[0348] Step 5: Recognizing the user's emotional state

[0349] Input: Data about user actions on the interface (clicks, keystrokes, etc.).

[0350] Specific behavior:

[0351] 1. The server passes the behavioral data to the emotion engine.

[0352] 2. The server receives the analysis results of the emotion engine.

[0353] Data calculation: User behavior data analysis.

[0354] Output: User's emotional state (stressed, anxious, etc.).

[0355] Step 6: Preprocessing and model selection adjustments

[0356] Input: User emotional state and the encoded and normalized dataset.

[0357] Specific behavior:

[0358] 1. The server adjusts data preprocessing and model selection methods based on the emotional state.

[0359] 2. The server simplifies the data preprocessing step.

[0360] Data processing: Adjusting data preprocessing strategies.

[0361] Output: Preprocessed data reflecting emotional states.

[0362] Step 7: Train the machine learning model

[0363] Input: Preprocessed data.

[0364] Specific behavior:

[0365] 1. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow.

[0366] 2. The server trains each model and performs cross-validation.

[0367] Data computation: model training, cross-validation.

[0368] Output: Performance evaluation results for each machine learning model.

[0369] Step 8: Selecting the best model

[0370] Input: Performance evaluation results for each model.

[0371] Specific behavior:

[0372] 1. The server selects the optimal model based on performance evaluation.

[0373] Data calculation: Comparison of performance evaluation results.

[0374] Output: The best model selected.

[0375] Step 9: Hyperparameter tuning

[0376] Input: The selected best model.

[0377] Specific behavior:

[0378] 1. The server optimizes the hyperparameters using GridSearchCV.

[0379] Data computation: Hyperparameter optimization.

[0380] Output: An optimized machine learning model.

[0381] Step 10: Perform prediction and classification

[0382] Input: Optimized machine learning model and new data.

[0383] Specific behavior:

[0384] 1. The server uses the optimized model to make predictions or classifications on new data.

[0385] Data calculations: Predicting or classifying new data.

[0386] Output: Prediction or classification results.

[0387] Step 11: Visualizing and outputting results

[0388] Input: Prediction or classification results.

[0389] Specific behavior:

[0390] 1. The server visualizes the results using matplotlib, Plotly, etc.

[0391] 2. The server visualizes the results and displays them on a dashboard.

[0392] 3. Export the results in CSV or Excel format as needed.

[0393] Data arithmetic: visualization and format transformation of results.

[0394] Output: Dashboard view and CSV / Excel format files.

[0395] (Application example 2)

[0396] 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."

[0397] Conventional data analysis systems do not take into account the user's emotional state, resulting in situations where users feel stressed or where data preprocessing and model selection are suboptimal. Furthermore, product recommendation systems on online shopping sites do not reflect user emotions and are limited to uniform recommendation accuracy. This prevents improvements to the user experience and increases in purchasing motivation.

[0398] 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.

[0399] In this invention, the server includes: [means for a user to upload a data file;] [means for the server to store the received data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to use an emotion engine that recognizes the user's emotional state;] [means for the server to adjust preprocessing and model selection based on the user's emotional state;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model;] [means for the server to visualize the results and display them on a dashboard or output them in file format;] [means for collecting user emotion data via a smartphone or other device; and [means for the server to recommend products.] This enables real-time recognition of the user's emotional state, corresponding data preprocessing and model selection, and even emotion-based product recommendations.

[0400] The "means for users to upload data files" refers to the means by which users add data files to be analyzed to the system.

[0401] The "means for storing data received by the server in an internal database" refers to the means for the server to store data files received from users in a database.

[0402] "Means for the server to automatically check data consistency and preprocess" refers to automated means for the server to check data consistency and handle missing values ​​and outliers.

[0403] The "means for the server to use an emotion engine to recognize the user's emotional state" refers to a means for utilizing an engine to detect and analyze the user's emotions and identify a specific emotional state.

[0404] The "means for the server to adjust preprocessing and model selection based on the emotional state of the user" refers to a means for changing the process of data preprocessing and appropriate model selection in accordance with the emotional state of the user detected using the emotion engine.

[0405] "Means in which the server applies multiple machine learning algorithms and selects the most appropriate model" refers to a means in which data is analyzed using multiple machine learning technologies and the optimal algorithm is selected from among them.

[0406] The "means for the server to make predictions or classifications using the selected model" refers to the means for making predictions or classifications of data using the optimal trained model.

[0407] The "means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to a means for displaying the results of prediction or classification in a visually easy-to-understand manner on a dashboard or outputting them as a file.

[0408] "Means for collecting user emotional data via smartphones or other devices" refers to means for collecting user emotional information in real time using smartphones or other sensor devices.

[0409] The "means for the server to recommend products" refers to a means for recommending products suitable for a user based on collected data and the user's emotional state.

[0410] This invention is an emotion-adaptive product recommendation system that applies the PredictEase system, which incorporates an emotion engine that recognizes the user's emotional state, to an online shopping site. This system can be accessed by users via smartphones or other devices.

[0411] Data Entry and Preprocessing

[0412] Users access the online shopping site's application and upload the necessary data files. The server receives the data and stores it in an internal database. At that time, the server checks the data for consistency and automatically processes missing or outlier values.

[0413] Recognition of emotional states

[0414] The server collects user emotional data via smartphones and other devices (cameras, microphones, emotion-sensing devices, etc.). The emotion engine analyzes this data and identifies the user's emotional state. For example, it uses facial recognition and voice analysis technology to determine whether the user is feeling stressed.

[0415] Model selection and tuning

[0416] Based on the user's emotional state identified by the emotion engine, the server adaptively adjusts the data preprocessing and model selection processes, thereby shortening processing time and providing easy-to-understand feedback to users experiencing stress.

[0417] The server analyzes the data by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, etc.) to select the optimal model, and then performs hyperparameter tuning of the selected model to achieve optimal performance.

[0418] Predictions and product recommendations

[0419] The server uses the best-fit model trained on the system to make predictions or classifications based on the user's data. It also considers the user's emotional state and purchasing history to recommend suitable products. The results are visualized and displayed intuitively on the user's application screen. Users can also download the results as a file if desired.

[0420] Hardware and software used

[0421] Hardware: Smartphone (camera, microphone), emotion sensing device (e.g., EEG headset)

[0422] Software: Sentiment analysis software (e.g., Sentiment API), machine learning libraries (e.g., TensorFlow, Scikit-learn)

[0423] Examples and prompts

[0424] For example, if a user is feeling stressed, the "Emotion-Adaptive Smart Recommender" will recommend products using prompts like the following:

[0425] User Emotion: Stress

[0426] Past purchase history: Aroma candles, relaxation CDs, herbal tea

[0427] Recommended Products:

[0428] 1. Lavender aroma oil

[0429] 2. Yoga mat

[0430] 3. Relaxation Cafe Gift Card

[0431] This system can recommend optimal products by taking into account the user's emotional state in real time, which is expected to improve the user experience and satisfaction.

[0432] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0433] Step 1:

[0434] The user accesses the application of the online shopping site and uploads the data file to be analyzed (e.g., a CSV file).

[0435] Input: Data file uploaded by the user.

[0436] Output: The data file that is transferred to the server.

[0437] Specific operation: The user uses a smartphone or personal computer to select a data file from the file selection screen and click the upload button.

[0438] Step 2:

[0439] The server stores the received data files in an internal database, performs data consistency checks, and automatically handles missing and outlier values.

[0440] Input: Data file received from user.

[0441] Output: Preprocessed data that has passed integrity checks.

[0442] What it does: The server stores the data in a database and runs scripts to impute missing values ​​(e.g., impute with the mean) and remove / correct outliers.

[0443] Step 3:

[0444] The server collects user emotional data via the smartphone's camera, microphone, and emotion sensing device.

[0445] Input: User facial recognition data, voice data, and input from other sensors.

[0446] Output: Data on the user's emotional state.

[0447] Specific operation: The server performs facial expression analysis from camera images using an emotion recognition API, and performs emotion analysis from microphone audio.

[0448] Step 4:

[0449] The server uses an emotion engine to analyze the collected emotion data and identify the user's emotional state.

[0450] Input: Collected emotion data.

[0451] Output: The user's specific emotional state (e.g., stress, joy, excitement).

[0452] Specific operation: The server runs the emotion engine, analyzes the data, and outputs the emotional state as a number or category.

[0453] Step 5:

[0454] The server adaptively adjusts the data preprocessing and model selection processes based on the user's emotional state.

[0455] Input: The user's emotional state.

[0456] Output: Adjusted preprocessing parameters and model selection strategy.

[0457] What it does: The server considers the emotion data and performs simplified pre-processing to reduce processing time or generates easy-to-understand feedback.

[0458] Step 6:

[0459] The server applies multiple machine learning algorithms and selects the most suitable model.

[0460] Input: Preprocessed data.

[0461] Output: Best model via cross-validation.

[0462] What it does: The server applies different algorithms (e.g., linear regression, random forest) and selects the best model based on the results of cross-validation.

[0463] Step 7:

[0464] The server tunes the hyperparameters of the selected model to achieve optimal performance.

[0465] Input: A cross-validated model.

[0466] Output: The optimal tuned model.

[0467] Specific operation: Set the range of hyperparameters and search for the optimal parameters using grid search or random search.

[0468] Step 8:

[0469] The server uses the best trained model to make predictions and product recommendations.

[0470] Input: User data and emotional state.

[0471] Output: A list of product recommendations.

[0472] What it does: Apply the model to rank and recommend products based on user sentiment and purchase history.

[0473] Step 9:

[0474] The server visualizes the recommendation results and displays them on the user's application screen.

[0475] Input: A list of recommended products.

[0476] Output: Visualized dashboard or file format.

[0477] What it does: Generates results in the form of graphs and lists, displays them in a dashboard that is intuitive and easy for users to understand, and provides the results as a downloadable file if required.

[0478] 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.

[0479] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0480] 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.

[0481] [Second embodiment]

[0482] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0483] 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.

[0484] 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).

[0485] 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.

[0486] 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.

[0487] 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).

[0488] 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.

[0489] 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.

[0490] 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.

[0491] 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.

[0492] 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.

[0493] 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."

[0494] This invention relates to PredictEase, a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. Specifically, the system is composed of the following steps:

[0495] System Overview

[0496] Data Entry

[0497] 1. A user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[0498] 2. The server receives the uploaded data file and stores it in its internal database.

[0499] Data Preprocessing

[0500] 1. The server performs a consistency check on the data file to determine if there are any missing or outlier values.

[0501] 2. The server handles missing and outliers in an appropriate way and converts the data into a form that can be applied to the machine learning model.

[0502] 3. The server encodes categorical variables and normalizes or standardizes numeric variables.

[0503] Model Selection

[0504] 1. The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[0505] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[0506] Model Optimization

[0507] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[0508] Prediction and Classification

[0509] 1. The server uses the trained model to make a prediction or classification.

[0510] 2. The server aggregates the prediction and classification results and evaluates their performance.

[0511] Output of results

[0512] 1. The server visually displays prediction and classification results on a dashboard.

[0513] 2. Display the results in an interface so that users can view them in graphs and tables.

[0514] 3. The server makes the results available for download as a CSV or Excel file, if desired.

[0515] Specific examples

[0516] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[0517] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[0518] 2. The server stores the received sales data in an internal database and starts checking the data for consistency. For example, if there are missing values ​​in the sales data, it fills them with the average value and handles any outliers appropriately.

[0519] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[0520] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[0521] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[0522] 6. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[0523] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[0524] This makes it easy for users without data science knowledge to perform advanced data analysis and predictions, making PredictEase a powerful tool for supporting data-driven decision-making.

[0525] The processing flow will be explained below.

[0526] Step 1:

[0527] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[0528] Step 2:

[0529] The server receives the uploaded data file, checks whether it is in the correct format, and then stores the data file in its internal database.

[0530] Step 3:

[0531] The server performs data integrity checks, specifically detecting missing and outlier values ​​and handling them in an appropriate way (e.g., imputing with the mean, removing outliers).

[0532] Step 4:

[0533] The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables to convert the data into a format that can be applied to machine learning models.

[0534] Step 5:

[0535] The server analyzes the features in the dataset and evaluates correlations and the importance of variables, thereby selecting important features and eliminating unnecessary features.

[0536] Step 6:

[0537] The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) to train each model.

[0538] Step 7:

[0539] The server evaluates the performance of each model using cross-validation, calculates evaluation metrics such as accuracy and F1 score for each model, and selects the optimal model.

[0540] Step 8:

[0541] The server tunes the hyperparameters of the selected model, using grid search or random search to find the optimal hyperparameters and build the final version of the model.

[0542] Step 9:

[0543] The server performs prediction or classification using the prepared test data, and then aggregates the prediction and classification results to evaluate their performance.

[0544] Step 10:

[0545] The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[0546] Step 11:

[0547] Users access the dashboard and view prediction and classification results displayed in graphs and tables.

[0548] Step 12:

[0549] If desired, users can download the results as a CSV or Excel file.

[0550] Through this series of steps, PredictEase enables users to easily perform highly accurate data analysis and predictions, even without data science knowledge.

[0551] Example 1

[0552] 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."

[0553] Current data analysis and prediction systems are difficult to use effectively unless users have specialized data science knowledge. They also require complex procedures such as data preprocessing, optimal model selection, and hyperparameter tuning, so there is a need for a system that can perform these processes consistently and automatically. Furthermore, there is a need for a system that can easily visualize and output the results.

[0554] 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.

[0555] In this invention, the server includes means for users to upload data files, means for the computer to store the received data in an internal storage device, means for the computer to automatically check the consistency of the data and preprocess it, means for the computer to apply multiple machine learning algorithms and select the most suitable model, means for the computer to make predictions or classifications using the selected model, means for the computer to visualize the results and display them on a display device or output them in file format, and means for users to check the results and make them downloadable. This allows even users without specialized knowledge to easily and efficiently perform processes from uploading data to obtaining prediction results.

[0556] "User" refers to a person who uses this system, uploads data, and obtains analysis and prediction results.

[0557] A "data file" refers to a collection of digital data, such as CSV or Excel format, that is input into a system for analysis or prediction.

[0558] "Computer" refers to a collection of hardware and software for receiving, processing, analyzing data, and outputting results.

[0559] "Internal storage device" refers to a storage medium that a computer uses to temporarily or permanently store data files.

[0560] "Integrity check" refers to the process of verifying that a data file conforms to the expected format and structure and that there are no missing or outlier values.

[0561] "Preprocessing" refers to a series of steps that transform, filter, and encode data to improve the accuracy of analysis or prediction.

[0562] "Machine learning algorithms" refer to mathematical models and methods that automatically learn patterns and knowledge from data and make predictions and classifications.

[0563] A "model" is a computational system trained by machine learning algorithms that has the ability to make predictions or classifications based on input data.

[0564] "Hyperparameter tuning" refers to the process of adjusting the configuration parameters of an algorithm to optimize the performance of a machine learning model.

[0565] "Visualization" refers to the process of representing prediction and classification results as graphs or charts so that they can be understood intuitively.

[0566] "Display device" refers to hardware such as a monitor or display that presents visualized results to a user.

[0567] "Downloading" refers to the act of a user transferring and saving a data file of prediction or classification results from the system to their own device.

[0568] This invention relates to a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. In this system, users upload data files, and the server automatically preprocesses the data, selects and optimizes models, and predicts and visualizes the results.

[0569] System Overview

[0570] Data Entry

[0571] The process begins when a user accesses the web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The server receives the uploaded data file and stores it in its internal storage.

[0572] Data Preprocessing

[0573] The server performs a consistency check on the data files to check for missing or outliers. It then imputes missing values ​​with the mean and handles outliers appropriately. It also encodes categorical variables and normalizes or standardizes numerical variables. This process is important as a preprocessing step for prediction and classification.

[0574] Model Selection and Optimization

[0575] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It evaluates the performance of each model using cross-validation and selects the best model. It then optimizes the hyperparameters of the selected model using grid search or random search. This step improves the model's predictive accuracy.

[0576] Prediction and visualization

[0577] The server then uses the best trained model to make predictions and classifications for new data. The results are visualized and displayed on a dashboard, presented to users in the form of graphs and tables. Users can also download the results as CSV or Excel files for offline analysis.

[0578] Specific examples

[0579] For example, if a user in a company's planning department wants to forecast monthly sales data, they can use the system as follows:

[0580] 1. The user prepares past sales data in CSV file format and accesses the web application to upload the file.

[0581] 2. The server stores the received sales data in its internal storage device and starts checking the data for consistency. If missing values ​​are found, they are filled in with the average value, and any outliers are dealt with appropriately.

[0582] 3. The server preprocesses the data, encoding categorical variables and normalizing numerical variables.

[0583] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[0584] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[0585] 6. The server uses the optimal model to predict sales for the next month and displays the forecast results on the dashboard.

[0586] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[0587] Prompt Sentence Examples

[0588] "Use the following sales data to predict next month's sales. Use linear regression and random forest as your models. Impute missing values ​​with the mean and treat outliers appropriately."

[0589] In this way, the system efficiently processes everything from data input to output, providing users with highly accurate prediction results. This process eliminates the need for specialized knowledge, resulting in a system that is convenient and user-friendly.

[0590] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0591] Step 1:

[0592] A user accesses the PredictEase web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The input is the data file to be analyzed, and the output is the uploaded data file sent to the server.

[0593] Step 2:

[0594] The server receives the uploaded data file and saves it to its internal storage. The data file "sales_data.csv" is saved to the server. The input is the uploaded data file, and the output is saving the data file to its internal storage.

[0595] Step 3:

[0596] The server performs a consistency check on the data file. Specifically, it checks whether the number of columns and data types match, and checks for missing or outlier values. The input is the saved data file, and the output is the data consistency check result.

[0597] Step 4:

[0598] The server imputes missing values ​​with the mean value and filters or imputes outliers, if any. The input is data that has been checked for consistency, and the output is preprocessed data.

[0599] Step 5:

[0600] The server encodes categorical variables into numeric values ​​and normalizes or standardizes numeric variables. For example, categorical variables such as "store name" are one-hot encoded, and numeric variables such as "sales amount" are processed with min-max scaling. The input is the preprocessed data, and the output is the encoded and normalized data.

[0601] Step 6:

[0602] The server trains the model by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.). The input is the encoded and normalized data, and the output is a model trained with each algorithm.

[0603] Step 7:

[0604] The server evaluates the performance of each model using cross-validation and selects the best model. For example, cross-validation evaluates the mean squared error (MSE) or coefficient of determination (R²). The input is the trained model, and the output is the best model.

[0605] Step 8:

[0606] The server optimizes the hyperparameters of the selected optimal model using grid search or random search. The input is the optimal model, and the output is the model with optimized parameters.

[0607] Step 9:

[0608] The server uses the best-fit model it has trained to make predictions or classifications for new data. The input is the new data, and the output is the prediction or classification result.

[0609] Step 10:

[0610] The server aggregates the prediction and classification results and calculates evaluation metrics such as accuracy and precision. The input is the prediction and classification results, and the output is a performance evaluation report.

[0611] Step 11:

[0612] The server visualizes the prediction and classification results and displays them on a dashboard. Users can check the results displayed in graphs and tables on the dashboard. The input is a performance evaluation report, and the output is a visualized dashboard display.

[0613] Step 12:

[0614] The server optionally makes the results available for download as CSV or Excel files, which users can download and analyze offline. The input is the visualized dashboard data, and the output is a downloadable file.

[0615] (Application example 1)

[0616] 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."

[0617] For modern self-driving vehicles, it is essential to analyze massive amounts of operational data in real time to detect and predict anomalies. However, this requires advanced knowledge of data science, which is a significant hurdle for non-expert users. Additionally, the pre-processing and modeling of operational data is complex and difficult to perform efficiently. There is a need for a system that can solve these problems and enable anyone to easily analyze data from self-driving vehicles.

[0618] 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.

[0619] In this invention, the server includes means for users to upload data files, means for the server to store the received data in an internal database, means for the server to automatically perform data consistency checks and preprocessing, means for the server to apply multiple machine learning algorithms and select the most appropriate model, means for the server to perform prediction or classification using the selected model, means for the server to visualize the results and display them on a dashboard or output them in file format, means for the server to analyze operational data of autonomous vehicles via an application installed on a smartphone and perform anomaly detection and prediction, and means for storing the trained model in the smartphone. This enables even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and prediction in real time.

[0620] "Means for users to upload data files" refers to an interface through which users submit files containing autonomous vehicle operation data to the web application.

[0621] "Means for storing the data received by the server in an internal database" refers to a processing device and storage technology for storing the received operation data in a database within the server.

[0622] "Means for the server to automatically perform data consistency checks and preprocessing" refers to technology in which the server checks operational data for missing values, abnormal values, etc., and automatically performs the necessary preprocessing (e.g., filling in missing values, cleaning data).

[0623] "A means by which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a technology in which the server tries out multiple models, evaluates their performance, and selects the model with the highest predictive accuracy.

[0624] "Means for the server to make predictions or classifications using the selected model" refers to a technology for making predictions or classifications for new operational data using the selected optimal model.

[0625] "Means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to the function by which the server displays the prediction or classification results in a visually easy-to-understand dashboard and outputs them in a file format (e.g., CSV, Excel) if necessary.

[0626] "Means of analyzing driving data of autonomous vehicles via an application installed on a smartphone to detect and predict abnormalities" refers to technology that analyzes driving data via a dedicated application on a smartphone, detects abnormal driving patterns, and predicts future driving conditions.

[0627] "Means for storing trained models within a smartphone" refers to technology that stores pre-trained machine learning models in a smartphone's storage device and makes them accessible to applications.

[0628] This invention relates to a system that enables users without specialized data science knowledge to easily analyze operational data of autonomous vehicles and perform anomaly detection and prediction. Specifically, the system is configured with the following steps.

[0629] System Overview

[0630] 1. Data Entry

[0631] Users upload autonomous vehicle operation data files (e.g., CSV or Excel files) via an application installed on their smartphones, and the uploaded data is sent to a server.

[0632] 2. Receipt and storage of data

[0633] The server stores the received operational data in an internal database, where it is quickly and securely stored and used for subsequent data analysis processes.

[0634] 3. Data integrity check and preprocessing

[0635] The server automatically performs integrity checks on the stored data, checking for missing or outlier values, and performs appropriate pre-processing, including imputing missing values ​​and cleaning the data.

[0636] 4. Model Selection

[0637] The server tries multiple machine learning algorithms and selects the best model, such as a random forest or a support vector machine. The selected model is then evaluated using cross-validation.

[0638] 5. Optimizing and training the model

[0639] The server then adjusts the hyperparameters of the selected model to achieve optimal performance, and the trained model is stored on the smartphone's memory.

[0640] 6. Prediction and Anomaly Detection

[0641] The selected model is used to predict and classify new driving data from the autonomous vehicle. The smartphone receives the analysis results and predicts abnormal driving patterns and future driving situations.

[0642] 7. Visualizing and outputting results

[0643] The server visually displays prediction and classification results on a dashboard. Users can view the results in graphs and tables through a smartphone interface. Users can also download the results in file format (e.g., CSV or Excel) if desired.

[0644] This allows even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and future predictions in real time.

[0645] Hardware and software used

[0646] Smartphone: Install the application and use it to upload data and display analysis results.

[0647] Server: Receives data, stores it, checks its integrity, preprocesses it, and selects, trains, and optimizes the model.

[0648] Python: A programming language used for data analysis and model training.

[0649] Pandas: A library used to load and preprocess data.

[0650] Scikit-learn: A library used to implement machine learning algorithms.

[0651] Joblib: A library used to save trained models.

[0652] Specific examples

[0653] For example, if a user wants to upload operational data (speed, GPS information, sensor information, etc.) from an autonomous vehicle and detect abnormal driving patterns, they can use the PredictDrive system via a smartphone application. The server receives the data, performs preprocessing, selects the optimal machine learning model, and trains it. The trained model is stored on the smartphone, and when new operational data is entered, it detects abnormalities in real time and displays the prediction results on a dashboard.

[0654] Prompt Sentence Examples

[0655] "Please tell us your next use case: Create a Python script that uses past driving data to predict whether the next driving pattern will be anomalous. The driving data includes speed, GPS information, and various sensor information. Use a random forest as the model, and create a visualization plot to present the results. Include detailed missing value handling and data standardization methods."

[0656] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0657] Step 1:

[0658] The user uploads the autonomous vehicle's operation data file (e.g., CSV format) using an application installed on their smartphone. This input data includes speed, GPS information, and sensor information. The data is then sent to the server via the smartphone.

[0659] Step 2:

[0660] The server stores the received operational data in an internal database. The stored data is used for subsequent processing within the server. The input here is the uploaded operational data, and the output is the data stored in the database.

[0661] Step 3:

[0662] The server performs consistency checks on the stored data. Specifically, it checks for missing values ​​and outliers and performs data cleaning. For example, missing values ​​are imputed with the mean value. The input of this step is the stored data, and the output is the cleaned data.

[0663] Step 4:

[0664] The server preprocesses the data by encoding categorical variables and normalizing or standardizing numerical variables. Encoding converts categorical data into a numerical representation, and normalization is achieved by scaling the data range. The input is the cleaned data, and the output is the preprocessed data.

[0665] Step 5:

[0666] The server tries multiple machine learning algorithms and applies and trains models such as random forests and support vector machines. It evaluates the performance of each model through cross-validation and selects the most suitable model. The input is the preprocessed data, and the output is the selected optimal model.

[0667] Step 6:

[0668] The server adjusts the hyperparameters of the selected model and tunes it to achieve optimal performance. This process involves an iterative process to optimize the model parameters. The input is the selected model, and the output is the optimized model.

[0669] Step 7:

[0670] The trained optimization model is stored locally on the smartphone, which is then used for real-time prediction and classification. The input here is the optimization model, and the output is the model stored locally on the smartphone.

[0671] Step 8:

[0672] The smartphone uses the stored model to make predictions and detect anomalies in real time whenever new driving data is input. For example, if an abnormal driving pattern is detected, an alert is sent to the user. The input is new driving data, and the output is the prediction or anomaly detection result and a user notification.

[0673] Step 9:

[0674] The server visually displays the prediction and classification results on a dashboard. Users can check the results on their smartphones and, if necessary, download the results in file format (e.g., CSV or Excel). The input here is the prediction or classification results, and the output is the visualized dashboard and file output.

[0675] 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.

[0676] This invention combines the PredictEase system, which performs data prediction and classification, with an emotion engine that recognizes user emotions. This system is designed as a no-code tool that can be easily used by users without data science expertise, and provides appropriate feedback and preprocessing adjustments according to the user's emotional state.

[0677] System Overview

[0678] Data Entry

[0679] 1. The user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[0680] 2. The server receives the uploaded data file, checks its format, and then stores it in its internal database.

[0681] Data Preprocessing

[0682] 1. The server performs data integrity checks to detect missing or outlier values ​​and handles them in an appropriate way.

[0683] 2. The server encodes categorical variables and normalizes or standardizes numeric variables.

[0684] Using the Emotion Engine

[0685] 1. The server passes the user's input and actions on the interface to the emotion engine to recognize the user's emotions.

[0686] 2. The server uses the emotion engine to adjust data preprocessing and model selection strategies according to the user's emotional state. For example, if the user is impatient, it can shorten the preprocessing time or provide easy-to-understand feedback to the user.

[0687] Model Selection

[0688] 1. The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[0689] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[0690] Model Optimization

[0691] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[0692] Prediction and Classification

[0693] 1. The server uses the trained model to make a prediction or classification.

[0694] 2. The server aggregates the prediction and classification results and evaluates their performance.

[0695] Output of results

[0696] 1. The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[0697] 2. The user accesses the dashboard and checks the prediction and classification results displayed in graphs and tables.

[0698] 3. If desired, the user can download the results as a CSV or Excel file.

[0699] Specific examples

[0700] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[0701] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[0702] 2. The server stores the received sales data in an internal database and performs data consistency checks, such as filling in missing values ​​with the average value and handling outliers appropriately.

[0703] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[0704] 4. The server uses an emotion engine to recognize the user's emotional state based on their input and actions. For example, if the server determines that the user is impatient, it summarizes the processing results and provides quick feedback.

[0705] 5. The server applies multiple machine learning algorithms to train each model, for example, linear regression, random forest, decision tree, etc.

[0706] 6. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[0707] 7. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[0708] 8. The user can check the forecast results displayed in graphs and tables on the dashboard and download the results in CSV format if necessary.

[0709] This allows PredictEase to provide advanced data analysis and predictions while taking into account the user's emotional state. Even users without data science knowledge can use this system to obtain highly accurate data analysis results and make strategic decisions.

[0710] The processing flow will be explained below.

[0711] Step 1:

[0712] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[0713] Step 2:

[0714] The server receives the uploaded data file, checks whether it is in the correct format, and then stores it in the internal database. If the format is incorrect, an error message is displayed to the user.

[0715] Step 3:

[0716] The server checks the integrity of the data file, detecting missing and abnormal values ​​(e.g., extreme values ​​and outliers) and performing other processing such as filling in the missing values ​​with the mean.

[0717] Step 4:

[0718] The server performs data preprocessing, such as label encoding and one-hot encoding for categorical variables and standardization and normalization for numerical variables, to convert the data into a form suitable for use in machine learning models.

[0719] Step 5:

[0720] The server uses an emotion engine to recognize the user's emotional state, for example by analyzing the video captured by the user's webcam, mouse movements, typing speed, etc.

[0721] Step 6:

[0722] The server adjusts its data preprocessing and model selection strategies based on the results of the emotion engine: for example, if the user is impatient, it will shorten the preprocessing time and present the results in an easy-to-understand format.

[0723] Step 7:

[0724] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It splits the data into a training dataset and a test dataset and proceeds with learning for each model.

[0725] Step 8:

[0726] The server performs cross-validation and evaluates the performance of each model, calculating evaluation metrics such as accuracy, F1 score, and R^2 value, and selecting the best model.

[0727] Step 9:

[0728] The server performs hyperparameter tuning for the selected model, for example, using grid search or random search to find the optimal parameter set.

[0729] Step 10:

[0730] The server uses the optimized model to make predictions and classifications, and performs final evaluation of the model using test data.

[0731] Step 11:

[0732] The server visualizes the prediction and classification results, converting them into a format that can be displayed in graphs and tables on a dashboard.

[0733] Step 12:

[0734] Users can access the dashboard to see the results and can click on graphs and tables to access more detailed information.

[0735] Step 13:

[0736] If desired, users can download the results as CSV or Excel files, which can then be saved for further analysis and reporting.

[0737] Through these specific processing steps, PredictEase incorporates the user's emotional state while automating advanced data analysis, enabling highly accurate prediction and classification without specialized knowledge.

[0738] Example 2

[0739] 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."

[0740] Conventional data analysis systems are difficult for users without specialized knowledge to use, and require a lot of time and effort for data preprocessing and machine learning model selection. Furthermore, the systems operate without taking into account the user's emotional state, which can be stressful for users. Furthermore, it is difficult to perform highly accurate predictions and classifications while maintaining data quality.

[0741] 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.

[0742] In this invention, the server includes: [means for a user to upload a data file to be analyzed;] [means for the server to store the uploaded data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model; and [means for the server to visualize the results of prediction or classification and display them on a dashboard or output them in file format.] This makes it possible to easily perform highly accurate data analysis and prediction while taking into account the user's emotional state.

[0743] The "means for users to upload data files to be analyzed" refers to an interface that allows users to submit data files saved in formats such as CSV or Excel to the web application.

[0744] "Means by which the server stores uploaded data in its internal database" refers to the process by which the server receives data files uploaded by users, organizes and verifies the data, and stores it in its internal database.

[0745] "Means for the server to automatically perform data consistency checks and preprocessing" refers to a function that enables the server to detect missing or outlier values ​​in the data, complete or delete them, and automatically perform data preprocessing such as encoding categorical variables and normalizing numerical variables.

[0746] "Means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection" refers to the process in which the server analyzes the user's interactions such as keyboard input and mouse clicks based on an emotion engine, recognizes the user's emotional state, and dynamically adjusts the data preprocessing and machine learning model selection method based on this.

[0747] "Means in which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a method in which the server applies multiple machine learning algorithms, such as linear regression, random forest, decision tree, and support vector machine, trains and evaluates each model, and then selects the most appropriate model.

[0748] The "means for the server to make predictions or classifications using the selected model" is a function that makes predictions or classifications for new data using the optimal machine learning model that the server has previously selected.

[0749] "Means for the server to visualize the prediction or classification results and display them on a dashboard or output them in a file format" means the ability of the server to visualize the prediction or classification results obtained by the machine learning model in graph or table format and display them on a dashboard that can be accessed by the user, or output them in a file format such as CSV or Excel.

[0750] MODE FOR CARRYING OUT THE INVENTION

[0751] This invention relates to a system that streamlines data analysis, recognizes user emotions, and adjusts feedback. This system is based on the no-code data analysis tool "PredictEase," and can be used intuitively even by beginners.

[0752] In an example, a user first accesses the PredictEase web application through a web browser and uploads the data file to be analyzed, which can be in CSV or Excel format.

[0753] The server receives, validates, and stores data using a backend built in Python. Specifically, the received data is checked for consistency using the pandas library. Missing values ​​and outliers are detected and then appropriately imputed or deleted. This ensures the quality of the data.

[0754] Next, the data is preprocessed using scikit-learn: categorical variables are one-hot encoded and numerical variables are normalized or standardized. This preprocessing enables efficient and effective training with multiple machine learning algorithms.

[0755] Furthermore, the system incorporates an emotion engine that recognizes emotions by analyzing user behavior on the interface. For example, a user who frequently sees error messages can be determined to be impatient. Based on the analysis results of the emotion engine, the server dynamically adjusts data preprocessing and model selection strategies. This process reduces user stress and provides a comfortable user experience.

[0756] For model selection, the server uses scikit-learn and TensorFlow to apply multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains models for each algorithm. It evaluates the performance of each model using cross-validation and selects the best model. This training and evaluation process maximizes the accuracy of predictions and classifications.

[0757] After the best model is selected, hyperparameter tuning is performed using GridSearchCV to build the best performing model. The server then uses this optimized model to make predictions or classifications on new data.

[0758] Finally, the server visualizes the prediction or classification results and outputs them in a dashboard format that is easy for users to understand, using visualization tools such as matplotlib and Plotly. Users can check the results through the dashboard and download them in CSV or Excel format if necessary.

[0759] Specific examples

[0760] For example, if a marketing manager at a company wants to use monthly sales data to predict next month's sales, he or she can use PredictEase as follows:

[0761] 1. The user prepares past sales data in CSV format and accesses the PredictEase web application to upload it.

[0762] 2. The server receives the data and uses the pandas library to check the data for consistency, impute missing values, and handle outliers appropriately.

[0763] 3. Use scikit-learn to perform data preprocessing, one-hot encoding categorical variables, and normalizing or standardizing numerical variables.

[0764] 4. The server uses an emotion engine to recognize the user's emotional state and adjusts pre-processing and feedback strategies according to the user's level of impatience.

[0765] 5. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow and selects the best model using cross-validation.

[0766] 6. Use GridSearchCV to tune the hyperparameters of the selected model and build the optimal model.

[0767] 7. The server uses this optimal model to predict next month's sales, visualizes the prediction results, and displays them on a dashboard.

[0768] 8. Users can check the prediction results on the dashboard and download them in CSV format if necessary.

[0769] This allows PredictEase to significantly reduce the effort required for data analysis and provide highly accurate predictions while taking into account the user's emotional state.

[0770] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0771] PredictEase System Program Processing Steps

[0772] Step 1: Upload your data files

[0773] Input: User-uploaded data file in CSV or Excel format.

[0774] Specific behavior:

[0775] 1. A user accesses the web application and selects a data file.

[0776] 2. The user clicks the "Upload" button.

[0777] Data processing: None at this time.

[0778] Output: The uploaded file is sent to the server.

[0779] Step 2: Receiving and storing data

[0780] Input: User uploaded data file.

[0781] Specific behavior:

[0782] 1. The server receives the uploaded file.

[0783] 2. The server checks the file format and stores it in its internal database.

[0784] Data processing: Check the format of the received file.

[0785] Output: The path of the data file stored in the database.

[0786] Step 3: Data integrity check and preprocessing

[0787] Input: Data files stored in the internal database.

[0788] Specific behavior:

[0789] 1. The server uses the pandas library to check the integrity of the data.

[0790] 2. The server detects missing values ​​and fills them with DataFrame.fillna().

[0791] 3. The server detects outliers and handles them appropriately.

[0792] Data processing: missing value imputation, outlier treatment.

[0793] Output: Consistency checked and preprocessed data.

[0794] Step 4: Data Encoding and Normalization

[0795] Input: Data that has been integrity checked and preprocessed.

[0796] Specific behavior:

[0797] 1. The server encodes categorical variables using scikit-learn's OneHotEncoder.

[0798] 2. The server normalizes the numerical variables using scikit-learn's StandardScaler.

[0799] Data processing: categorical data encoding, numerical data normalization.

[0800] Output: The dataset after encoding and normalization.

[0801] Step 5: Recognizing the user's emotional state

[0802] Input: Data about user actions on the interface (clicks, keystrokes, etc.).

[0803] Specific behavior:

[0804] 1. The server passes the behavioral data to the emotion engine.

[0805] 2. The server receives the analysis results of the emotion engine.

[0806] Data calculation: User behavior data analysis.

[0807] Output: User's emotional state (stressed, anxious, etc.).

[0808] Step 6: Preprocessing and model selection adjustments

[0809] Input: User emotional state and the encoded and normalized dataset.

[0810] Specific behavior:

[0811] 1. The server adjusts data preprocessing and model selection methods based on the emotional state.

[0812] 2. The server simplifies the data preprocessing step.

[0813] Data processing: Adjusting data preprocessing strategies.

[0814] Output: Preprocessed data reflecting emotional states.

[0815] Step 7: Train the machine learning model

[0816] Input: Preprocessed data.

[0817] Specific behavior:

[0818] 1. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow.

[0819] 2. The server trains each model and performs cross-validation.

[0820] Data computation: model training, cross-validation.

[0821] Output: Performance evaluation results for each machine learning model.

[0822] Step 8: Selecting the best model

[0823] Input: Performance evaluation results for each model.

[0824] Specific behavior:

[0825] 1. The server selects the optimal model based on performance evaluation.

[0826] Data calculation: Comparison of performance evaluation results.

[0827] Output: The best model selected.

[0828] Step 9: Hyperparameter tuning

[0829] Input: The selected best model.

[0830] Specific behavior:

[0831] 1. The server optimizes the hyperparameters using GridSearchCV.

[0832] Data computation: Hyperparameter optimization.

[0833] Output: An optimized machine learning model.

[0834] Step 10: Perform prediction and classification

[0835] Input: Optimized machine learning model and new data.

[0836] Specific behavior:

[0837] 1. The server uses the optimized model to make predictions or classifications on new data.

[0838] Data calculations: Predicting or classifying new data.

[0839] Output: Prediction or classification results.

[0840] Step 11: Visualizing and outputting results

[0841] Input: Prediction or classification results.

[0842] Specific behavior:

[0843] 1. The server visualizes the results using matplotlib, Plotly, etc.

[0844] 2. The server visualizes the results and displays them on a dashboard.

[0845] 3. Export the results in CSV or Excel format as needed.

[0846] Data arithmetic: visualization and format transformation of results.

[0847] Output: Dashboard view and CSV / Excel format files.

[0848] (Application example 2)

[0849] 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."

[0850] Conventional data analysis systems do not take into account the user's emotional state, resulting in situations where users feel stressed or where data preprocessing and model selection are suboptimal. Furthermore, product recommendation systems on online shopping sites do not reflect user emotions and are limited to uniform recommendation accuracy. This prevents improvements to the user experience and increases in purchasing motivation.

[0851] 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.

[0852] In this invention, the server includes: [means for a user to upload a data file;] [means for the server to store the received data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to use an emotion engine that recognizes the user's emotional state;] [means for the server to adjust preprocessing and model selection based on the user's emotional state;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model;] [means for the server to visualize the results and display them on a dashboard or output them in file format;] [means for collecting user emotion data via a smartphone or other device; and [means for the server to recommend products.] This enables real-time recognition of the user's emotional state, corresponding data preprocessing and model selection, and even emotion-based product recommendations.

[0853] The "means for users to upload data files" refers to the means by which users add data files to be analyzed to the system.

[0854] The "means for storing data received by the server in an internal database" refers to the means for the server to store data files received from users in a database.

[0855] "Means for the server to automatically check data consistency and preprocess" refers to automated means for the server to check data consistency and handle missing values ​​and outliers.

[0856] The "means for the server to use an emotion engine to recognize the user's emotional state" refers to a means for utilizing an engine to detect and analyze the user's emotions and identify a specific emotional state.

[0857] The "means for the server to adjust preprocessing and model selection based on the emotional state of the user" refers to a means for changing the process of data preprocessing and appropriate model selection in accordance with the emotional state of the user detected using the emotion engine.

[0858] "Means in which the server applies multiple machine learning algorithms and selects the most appropriate model" refers to a means in which data is analyzed using multiple machine learning technologies and the optimal algorithm is selected from among them.

[0859] The "means for the server to make predictions or classifications using the selected model" refers to the means for making predictions or classifications of data using the optimal trained model.

[0860] The "means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to a means for displaying the results of prediction or classification in a visually easy-to-understand manner on a dashboard or outputting them as a file.

[0861] "Means for collecting user emotional data via smartphones or other devices" refers to means for collecting user emotional information in real time using smartphones or other sensor devices.

[0862] The "means for the server to recommend products" refers to a means for recommending products suitable for a user based on collected data and the user's emotional state.

[0863] This invention is an emotion-adaptive product recommendation system that applies the PredictEase system, which incorporates an emotion engine that recognizes the user's emotional state, to an online shopping site. This system can be accessed by users via smartphones or other devices.

[0864] Data Entry and Preprocessing

[0865] Users access the online shopping site's application and upload the necessary data files. The server receives the data and stores it in an internal database. At that time, the server checks the data for consistency and automatically processes missing or outlier values.

[0866] Recognition of emotional states

[0867] The server collects user emotional data via smartphones and other devices (cameras, microphones, emotion-sensing devices, etc.). The emotion engine analyzes this data and identifies the user's emotional state. For example, it uses facial recognition and voice analysis technology to determine whether the user is feeling stressed.

[0868] Model selection and tuning

[0869] Based on the user's emotional state identified by the emotion engine, the server adaptively adjusts the data preprocessing and model selection processes, thereby shortening processing time and providing easy-to-understand feedback to users experiencing stress.

[0870] The server analyzes the data by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, etc.) to select the optimal model, and then performs hyperparameter tuning of the selected model to achieve optimal performance.

[0871] Predictions and product recommendations

[0872] The server uses the best-fit model trained on the system to make predictions or classifications based on the user's data. It also considers the user's emotional state and purchasing history to recommend suitable products. The results are visualized and displayed intuitively on the user's application screen. Users can also download the results as a file if desired.

[0873] Hardware and software used

[0874] Hardware: Smartphone (camera, microphone), emotion sensing device (e.g., EEG headset)

[0875] Software: Sentiment analysis software (e.g., Sentiment API), machine learning libraries (e.g., TensorFlow, Scikit-learn)

[0876] Examples and prompts

[0877] For example, if a user is feeling stressed, the "Emotion-Adaptive Smart Recommender" will recommend products using prompts like the following:

[0878] User Emotion: Stress

[0879] Past purchase history: Aroma candles, relaxation CDs, herbal tea

[0880] Recommended Products:

[0881] 1. Lavender aroma oil

[0882] 2. Yoga mat

[0883] 3. Relaxation Cafe Gift Card

[0884] This system can recommend optimal products by taking into account the user's emotional state in real time, which is expected to improve the user experience and satisfaction.

[0885] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0886] Step 1:

[0887] The user accesses the application of the online shopping site and uploads the data file to be analyzed (e.g., a CSV file).

[0888] Input: Data file uploaded by the user.

[0889] Output: The data file that is transferred to the server.

[0890] Specific operation: The user uses a smartphone or personal computer to select a data file from the file selection screen and click the upload button.

[0891] Step 2:

[0892] The server stores the received data files in an internal database, performs data consistency checks, and automatically handles missing and outlier values.

[0893] Input: Data file received from user.

[0894] Output: Preprocessed data that has passed integrity checks.

[0895] What it does: The server stores the data in a database and runs scripts to impute missing values ​​(e.g., impute with the mean) and remove / correct outliers.

[0896] Step 3:

[0897] The server collects user emotional data via the smartphone's camera, microphone, and emotion sensing device.

[0898] Input: User facial recognition data, voice data, and input from other sensors.

[0899] Output: Data on the user's emotional state.

[0900] Specific operation: The server performs facial expression analysis from camera images using an emotion recognition API, and performs emotion analysis from microphone audio.

[0901] Step 4:

[0902] The server uses an emotion engine to analyze the collected emotion data and identify the user's emotional state.

[0903] Input: Collected emotion data.

[0904] Output: The user's specific emotional state (e.g., stress, joy, excitement).

[0905] Specific operation: The server runs the emotion engine, analyzes the data, and outputs the emotional state as a number or category.

[0906] Step 5:

[0907] The server adaptively adjusts the data preprocessing and model selection processes based on the user's emotional state.

[0908] Input: The user's emotional state.

[0909] Output: Adjusted preprocessing parameters and model selection strategy.

[0910] What it does: The server considers the emotion data and performs simplified pre-processing to reduce processing time or generates easy-to-understand feedback.

[0911] Step 6:

[0912] The server applies multiple machine learning algorithms and selects the most suitable model.

[0913] Input: Preprocessed data.

[0914] Output: Best model via cross-validation.

[0915] What it does: The server applies different algorithms (e.g., linear regression, random forest) and selects the best model based on the results of cross-validation.

[0916] Step 7:

[0917] The server tunes the hyperparameters of the selected model to achieve optimal performance.

[0918] Input: A cross-validated model.

[0919] Output: The optimal tuned model.

[0920] Specific operation: Set the range of hyperparameters and search for the optimal parameters using grid search or random search.

[0921] Step 8:

[0922] The server uses the best trained model to make predictions and product recommendations.

[0923] Input: User data and emotional state.

[0924] Output: A list of product recommendations.

[0925] What it does: Apply the model to rank and recommend products based on user sentiment and purchase history.

[0926] Step 9:

[0927] The server visualizes the recommendation results and displays them on the user's application screen.

[0928] Input: A list of recommended products.

[0929] Output: Visualized dashboard or file format.

[0930] What it does: Generates results in the form of graphs and lists, displays them in a dashboard that is intuitive and easy for users to understand, and provides the results as a downloadable file if required.

[0931] 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.

[0932] 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.

[0933] 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.

[0934] [Third embodiment]

[0935] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0936] 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.

[0937] 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).

[0938] 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.

[0939] 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.

[0940] 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).

[0941] 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.

[0942] 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.

[0943] 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.

[0944] 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.

[0945] 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.

[0946] 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."

[0947] This invention relates to PredictEase, a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. Specifically, the system is composed of the following steps:

[0948] System Overview

[0949] Data Entry

[0950] 1. A user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[0951] 2. The server receives the uploaded data file and stores it in its internal database.

[0952] Data Preprocessing

[0953] 1. The server performs a consistency check on the data file to determine if there are any missing or outlier values.

[0954] 2. The server handles missing and outliers in an appropriate way and converts the data into a form that can be applied to the machine learning model.

[0955] 3. The server encodes categorical variables and normalizes or standardizes numeric variables.

[0956] Model Selection

[0957] 1. The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[0958] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[0959] Model Optimization

[0960] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[0961] Prediction and Classification

[0962] 1. The server uses the trained model to make a prediction or classification.

[0963] 2. The server aggregates the prediction and classification results and evaluates their performance.

[0964] Output of results

[0965] 1. The server visually displays prediction and classification results on a dashboard.

[0966] 2. Display the results in an interface so that users can view them in graphs and tables.

[0967] 3. The server makes the results available for download as a CSV or Excel file, if desired.

[0968] Specific examples

[0969] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[0970] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[0971] 2. The server stores the received sales data in an internal database and starts checking the data for consistency. For example, if there are missing values ​​in the sales data, it fills them with the average value and handles any outliers appropriately.

[0972] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[0973] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[0974] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[0975] 6. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[0976] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[0977] This makes it easy for users without data science knowledge to perform advanced data analysis and predictions, making PredictEase a powerful tool for supporting data-driven decision-making.

[0978] The processing flow will be explained below.

[0979] Step 1:

[0980] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[0981] Step 2:

[0982] The server receives the uploaded data file, checks whether it is in the correct format, and then stores the data file in its internal database.

[0983] Step 3:

[0984] The server performs data integrity checks, specifically detecting missing and outlier values ​​and handling them in an appropriate way (e.g., imputing with the mean, removing outliers).

[0985] Step 4:

[0986] The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables to convert the data into a format that can be applied to machine learning models.

[0987] Step 5:

[0988] The server analyzes the features in the dataset and evaluates correlations and the importance of variables, thereby selecting important features and eliminating unnecessary features.

[0989] Step 6:

[0990] The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) to train each model.

[0991] Step 7:

[0992] The server evaluates the performance of each model using cross-validation, calculates evaluation metrics such as accuracy and F1 score for each model, and selects the optimal model.

[0993] Step 8:

[0994] The server tunes the hyperparameters of the selected model, using grid search or random search to find the optimal hyperparameters and build the final version of the model.

[0995] Step 9:

[0996] The server performs prediction or classification using the prepared test data, and then aggregates the prediction and classification results to evaluate their performance.

[0997] Step 10:

[0998] The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[0999] Step 11:

[1000] Users access the dashboard and view prediction and classification results displayed in graphs and tables.

[1001] Step 12:

[1002] If desired, users can download the results as a CSV or Excel file.

[1003] Through this series of steps, PredictEase enables users to easily perform highly accurate data analysis and predictions, even without data science knowledge.

[1004] Example 1

[1005] 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."

[1006] Current data analysis and prediction systems are difficult to use effectively unless users have specialized data science knowledge. They also require complex procedures such as data preprocessing, optimal model selection, and hyperparameter tuning, so there is a need for a system that can perform these processes consistently and automatically. Furthermore, there is a need for a system that can easily visualize and output the results.

[1007] 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.

[1008] In this invention, the server includes means for users to upload data files, means for the computer to store the received data in an internal storage device, means for the computer to automatically check the consistency of the data and preprocess it, means for the computer to apply multiple machine learning algorithms and select the most suitable model, means for the computer to make predictions or classifications using the selected model, means for the computer to visualize the results and display them on a display device or output them in file format, and means for users to check the results and make them downloadable. This allows even users without specialized knowledge to easily and efficiently perform processes from uploading data to obtaining prediction results.

[1009] "User" refers to a person who uses this system, uploads data, and obtains analysis and prediction results.

[1010] A "data file" refers to a collection of digital data, such as CSV or Excel format, that is input into a system for analysis or prediction.

[1011] "Computer" refers to a collection of hardware and software for receiving, processing, analyzing data, and outputting results.

[1012] "Internal storage device" refers to a storage medium that a computer uses to temporarily or permanently store data files.

[1013] "Integrity check" refers to the process of verifying that a data file conforms to the expected format and structure and that there are no missing or outlier values.

[1014] "Preprocessing" refers to a series of steps that transform, filter, and encode data to improve the accuracy of analysis or prediction.

[1015] "Machine learning algorithms" refer to mathematical models and methods that automatically learn patterns and knowledge from data and make predictions and classifications.

[1016] A "model" is a computational system trained by machine learning algorithms that has the ability to make predictions or classifications based on input data.

[1017] "Hyperparameter tuning" refers to the process of adjusting the configuration parameters of an algorithm to optimize the performance of a machine learning model.

[1018] "Visualization" refers to the process of representing prediction and classification results as graphs or charts so that they can be understood intuitively.

[1019] "Display device" refers to hardware such as a monitor or display that presents visualized results to a user.

[1020] "Downloading" refers to the act of a user transferring and saving a data file of prediction or classification results from the system to their own device.

[1021] This invention relates to a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. In this system, users upload data files, and the server automatically preprocesses the data, selects and optimizes models, and predicts and visualizes the results.

[1022] System Overview

[1023] Data Entry

[1024] The process begins when a user accesses the web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The server receives the uploaded data file and stores it in its internal storage.

[1025] Data Preprocessing

[1026] The server performs a consistency check on the data files to check for missing or outliers. It then imputes missing values ​​with the mean and handles outliers appropriately. It also encodes categorical variables and normalizes or standardizes numerical variables. This process is important as a preprocessing step for prediction and classification.

[1027] Model Selection and Optimization

[1028] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It evaluates the performance of each model using cross-validation and selects the best model. It then optimizes the hyperparameters of the selected model using grid search or random search. This step improves the model's predictive accuracy.

[1029] Prediction and visualization

[1030] The server then uses the best trained model to make predictions and classifications for new data. The results are visualized and displayed on a dashboard, presented to users in the form of graphs and tables. Users can also download the results as CSV or Excel files for offline analysis.

[1031] Specific examples

[1032] For example, if a user in a company's planning department wants to forecast monthly sales data, they can use the system as follows:

[1033] 1. The user prepares past sales data in CSV file format and accesses the web application to upload the file.

[1034] 2. The server stores the received sales data in its internal storage device and starts checking the data for consistency. If missing values ​​are found, they are filled in with the average value, and any outliers are dealt with appropriately.

[1035] 3. The server preprocesses the data, encoding categorical variables and normalizing numerical variables.

[1036] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[1037] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[1038] 6. The server uses the optimal model to predict sales for the next month and displays the forecast results on the dashboard.

[1039] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[1040] Prompt Sentence Examples

[1041] "Use the following sales data to predict next month's sales. Use linear regression and random forest as your models. Impute missing values ​​with the mean and treat outliers appropriately."

[1042] In this way, the system efficiently processes everything from data input to output, providing users with highly accurate prediction results. This process eliminates the need for specialized knowledge, resulting in a system that is convenient and user-friendly.

[1043] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1044] Step 1:

[1045] A user accesses the PredictEase web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The input is the data file to be analyzed, and the output is the uploaded data file sent to the server.

[1046] Step 2:

[1047] The server receives the uploaded data file and saves it to its internal storage. The data file "sales_data.csv" is saved to the server. The input is the uploaded data file, and the output is saving the data file to its internal storage.

[1048] Step 3:

[1049] The server performs a consistency check on the data file. Specifically, it checks whether the number of columns and data types match, and checks for missing or outlier values. The input is the saved data file, and the output is the data consistency check result.

[1050] Step 4:

[1051] The server imputes missing values ​​with the mean value and filters or imputes outliers, if any. The input is data that has been checked for consistency, and the output is preprocessed data.

[1052] Step 5:

[1053] The server encodes categorical variables into numeric values ​​and normalizes or standardizes numeric variables. For example, categorical variables such as "store name" are one-hot encoded, and numeric variables such as "sales amount" are processed with min-max scaling. The input is the preprocessed data, and the output is the encoded and normalized data.

[1054] Step 6:

[1055] The server trains the model by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.). The input is the encoded and normalized data, and the output is a model trained with each algorithm.

[1056] Step 7:

[1057] The server evaluates the performance of each model using cross-validation and selects the best model. For example, cross-validation evaluates the mean squared error (MSE) or coefficient of determination (R²). The input is the trained model, and the output is the best model.

[1058] Step 8:

[1059] The server optimizes the hyperparameters of the selected optimal model using grid search or random search. The input is the optimal model, and the output is the model with optimized parameters.

[1060] Step 9:

[1061] The server uses the best-fit model it has trained to make predictions or classifications for new data. The input is the new data, and the output is the prediction or classification result.

[1062] Step 10:

[1063] The server aggregates the prediction and classification results and calculates evaluation metrics such as accuracy and precision. The input is the prediction and classification results, and the output is a performance evaluation report.

[1064] Step 11:

[1065] The server visualizes the prediction and classification results and displays them on a dashboard. Users can check the results displayed in graphs and tables on the dashboard. The input is a performance evaluation report, and the output is a visualized dashboard display.

[1066] Step 12:

[1067] The server optionally makes the results available for download as CSV or Excel files, which users can download and analyze offline. The input is the visualized dashboard data, and the output is a downloadable file.

[1068] (Application example 1)

[1069] 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."

[1070] For modern self-driving vehicles, it is essential to analyze massive amounts of operational data in real time to detect and predict anomalies. However, this requires advanced knowledge of data science, which is a significant hurdle for non-expert users. Additionally, the pre-processing and modeling of operational data is complex and difficult to perform efficiently. There is a need for a system that can solve these problems and enable anyone to easily analyze data from self-driving vehicles.

[1071] 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.

[1072] In this invention, the server includes means for users to upload data files, means for the server to store the received data in an internal database, means for the server to automatically perform data consistency checks and preprocessing, means for the server to apply multiple machine learning algorithms and select the most appropriate model, means for the server to perform prediction or classification using the selected model, means for the server to visualize the results and display them on a dashboard or output them in file format, means for the server to analyze operational data of autonomous vehicles via an application installed on a smartphone and perform anomaly detection and prediction, and means for storing the trained model in the smartphone. This enables even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and prediction in real time.

[1073] "Means for users to upload data files" refers to an interface through which users submit files containing autonomous vehicle operation data to the web application.

[1074] "Means for storing the data received by the server in an internal database" refers to a processing device and storage technology for storing the received operation data in a database within the server.

[1075] "Means for the server to automatically perform data consistency checks and preprocessing" refers to technology in which the server checks operational data for missing values, abnormal values, etc., and automatically performs the necessary preprocessing (e.g., filling in missing values, cleaning data).

[1076] "A means by which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a technology in which the server tries out multiple models, evaluates their performance, and selects the model with the highest predictive accuracy.

[1077] "Means for the server to make predictions or classifications using the selected model" refers to a technology for making predictions or classifications for new operational data using the selected optimal model.

[1078] "Means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to the function by which the server displays the prediction or classification results in a visually easy-to-understand dashboard and outputs them in a file format (e.g., CSV, Excel) if necessary.

[1079] "Means of analyzing driving data of autonomous vehicles via an application installed on a smartphone to detect and predict abnormalities" refers to technology that analyzes driving data via a dedicated application on a smartphone, detects abnormal driving patterns, and predicts future driving conditions.

[1080] "Means for storing trained models within a smartphone" refers to technology that stores pre-trained machine learning models in a smartphone's storage device and makes them accessible to applications.

[1081] This invention relates to a system that enables users without specialized data science knowledge to easily analyze operational data of autonomous vehicles and perform anomaly detection and prediction. Specifically, the system is configured with the following steps.

[1082] System Overview

[1083] 1. Data Entry

[1084] Users upload autonomous vehicle operation data files (e.g., CSV or Excel files) via an application installed on their smartphones, and the uploaded data is sent to a server.

[1085] 2. Receipt and storage of data

[1086] The server stores the received operational data in an internal database, where it is quickly and securely stored and used for subsequent data analysis processes.

[1087] 3. Data integrity check and preprocessing

[1088] The server automatically performs integrity checks on the stored data, checking for missing or outlier values, and performs appropriate pre-processing, including imputing missing values ​​and cleaning the data.

[1089] 4. Model Selection

[1090] The server tries multiple machine learning algorithms and selects the best model, such as a random forest or a support vector machine. The selected model is then evaluated using cross-validation.

[1091] 5. Optimizing and training the model

[1092] The server then adjusts the hyperparameters of the selected model to achieve optimal performance, and the trained model is stored on the smartphone's memory.

[1093] 6. Prediction and Anomaly Detection

[1094] The selected model is used to predict and classify new driving data from the autonomous vehicle. The smartphone receives the analysis results and predicts abnormal driving patterns and future driving situations.

[1095] 7. Visualizing and outputting results

[1096] The server visually displays prediction and classification results on a dashboard. Users can view the results in graphs and tables through a smartphone interface. Users can also download the results in file format (e.g., CSV or Excel) if desired.

[1097] This allows even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and future predictions in real time.

[1098] Hardware and software used

[1099] Smartphone: Install the application and use it to upload data and display analysis results.

[1100] Server: Receives data, stores it, checks its integrity, preprocesses it, and selects, trains, and optimizes the model.

[1101] Python: A programming language used for data analysis and model training.

[1102] Pandas: A library used to load and preprocess data.

[1103] Scikit-learn: A library used to implement machine learning algorithms.

[1104] Joblib: A library used to save trained models.

[1105] Specific examples

[1106] For example, if a user wants to upload operational data (speed, GPS information, sensor information, etc.) from an autonomous vehicle and detect abnormal driving patterns, they can use the PredictDrive system via a smartphone application. The server receives the data, performs preprocessing, selects the optimal machine learning model, and trains it. The trained model is stored on the smartphone, and when new operational data is entered, it detects abnormalities in real time and displays the prediction results on a dashboard.

[1107] Prompt Sentence Examples

[1108] "Please tell us your next use case: Create a Python script that uses past driving data to predict whether the next driving pattern will be anomalous. The driving data includes speed, GPS information, and various sensor information. Use a random forest as the model, and create a visualization plot to present the results. Include detailed missing value handling and data standardization methods."

[1109] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1110] Step 1:

[1111] The user uploads the autonomous vehicle's operation data file (e.g., CSV format) using an application installed on their smartphone. This input data includes speed, GPS information, and sensor information. The data is then sent to the server via the smartphone.

[1112] Step 2:

[1113] The server stores the received operational data in an internal database. The stored data is used for subsequent processing within the server. The input here is the uploaded operational data, and the output is the data stored in the database.

[1114] Step 3:

[1115] The server performs consistency checks on the stored data. Specifically, it checks for missing values ​​and outliers and performs data cleaning. For example, missing values ​​are imputed with the mean value. The input of this step is the stored data, and the output is the cleaned data.

[1116] Step 4:

[1117] The server preprocesses the data by encoding categorical variables and normalizing or standardizing numerical variables. Encoding converts categorical data into a numerical representation, and normalization is achieved by scaling the data range. The input is the cleaned data, and the output is the preprocessed data.

[1118] Step 5:

[1119] The server tries multiple machine learning algorithms and applies and trains models such as random forests and support vector machines. It evaluates the performance of each model through cross-validation and selects the most suitable model. The input is the preprocessed data, and the output is the selected optimal model.

[1120] Step 6:

[1121] The server adjusts the hyperparameters of the selected model and tunes it to achieve optimal performance. This process involves an iterative process to optimize the model parameters. The input is the selected model, and the output is the optimized model.

[1122] Step 7:

[1123] The trained optimization model is stored locally on the smartphone, which is then used for real-time prediction and classification. The input here is the optimization model, and the output is the model stored locally on the smartphone.

[1124] Step 8:

[1125] The smartphone uses the stored model to make predictions and detect anomalies in real time whenever new driving data is input. For example, if an abnormal driving pattern is detected, an alert is sent to the user. The input is new driving data, and the output is the prediction or anomaly detection result and a user notification.

[1126] Step 9:

[1127] The server visually displays the prediction and classification results on a dashboard. Users can check the results on their smartphones and, if necessary, download the results in file format (e.g., CSV or Excel). The input here is the prediction or classification results, and the output is the visualized dashboard and file output.

[1128] 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.

[1129] This invention combines the PredictEase system, which performs data prediction and classification, with an emotion engine that recognizes user emotions. This system is designed as a no-code tool that can be easily used by users without data science expertise, and provides appropriate feedback and preprocessing adjustments according to the user's emotional state.

[1130] System Overview

[1131] Data Entry

[1132] 1. The user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[1133] 2. The server receives the uploaded data file, checks its format, and then stores it in its internal database.

[1134] Data Preprocessing

[1135] 1. The server performs data integrity checks to detect missing or outlier values ​​and handles them in an appropriate way.

[1136] 2. The server encodes categorical variables and normalizes or standardizes numeric variables.

[1137] Using the Emotion Engine

[1138] 1. The server passes the user's input and actions on the interface to the emotion engine to recognize the user's emotions.

[1139] 2. The server uses the emotion engine to adjust data preprocessing and model selection strategies according to the user's emotional state. For example, if the user is impatient, it can shorten the preprocessing time or provide easy-to-understand feedback to the user.

[1140] Model Selection

[1141] 1. The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[1142] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[1143] Model Optimization

[1144] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[1145] Prediction and Classification

[1146] 1. The server uses the trained model to make a prediction or classification.

[1147] 2. The server aggregates the prediction and classification results and evaluates their performance.

[1148] Output of results

[1149] 1. The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[1150] 2. The user accesses the dashboard and checks the prediction and classification results displayed in graphs and tables.

[1151] 3. If desired, the user can download the results as a CSV or Excel file.

[1152] Specific examples

[1153] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[1154] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[1155] 2. The server stores the received sales data in an internal database and performs data consistency checks, such as filling in missing values ​​with the average value and handling outliers appropriately.

[1156] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[1157] 4. The server uses an emotion engine to recognize the user's emotional state based on their input and actions. For example, if the server determines that the user is impatient, it summarizes the processing results and provides quick feedback.

[1158] 5. The server applies multiple machine learning algorithms to train each model, for example, linear regression, random forest, decision tree, etc.

[1159] 6. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[1160] 7. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[1161] 8. The user can check the forecast results displayed in graphs and tables on the dashboard and download the results in CSV format if necessary.

[1162] This allows PredictEase to provide advanced data analysis and predictions while taking into account the user's emotional state. Even users without data science knowledge can use this system to obtain highly accurate data analysis results and make strategic decisions.

[1163] The processing flow will be explained below.

[1164] Step 1:

[1165] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[1166] Step 2:

[1167] The server receives the uploaded data file, checks whether it is in the correct format, and then stores it in the internal database. If the format is incorrect, an error message is displayed to the user.

[1168] Step 3:

[1169] The server checks the integrity of the data file, detecting missing and abnormal values ​​(e.g., extreme values ​​and outliers) and performing other processing such as filling in the missing values ​​with the mean.

[1170] Step 4:

[1171] The server performs data preprocessing, such as label encoding and one-hot encoding for categorical variables and standardization and normalization for numerical variables, to convert the data into a form suitable for use in machine learning models.

[1172] Step 5:

[1173] The server uses an emotion engine to recognize the user's emotional state, for example by analyzing the video captured by the user's webcam, mouse movements, typing speed, etc.

[1174] Step 6:

[1175] The server adjusts its data preprocessing and model selection strategies based on the results of the emotion engine: for example, if the user is impatient, it will shorten the preprocessing time and present the results in an easy-to-understand format.

[1176] Step 7:

[1177] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It splits the data into a training dataset and a test dataset and proceeds with learning for each model.

[1178] Step 8:

[1179] The server performs cross-validation and evaluates the performance of each model, calculating evaluation metrics such as accuracy, F1 score, and R^2 value, and selecting the best model.

[1180] Step 9:

[1181] The server performs hyperparameter tuning for the selected model, for example, using grid search or random search to find the optimal parameter set.

[1182] Step 10:

[1183] The server uses the optimized model to make predictions and classifications, and performs final evaluation of the model using test data.

[1184] Step 11:

[1185] The server visualizes the prediction and classification results, converting them into a format that can be displayed in graphs and tables on a dashboard.

[1186] Step 12:

[1187] Users can access the dashboard to see the results and can click on graphs and tables to access more detailed information.

[1188] Step 13:

[1189] If desired, users can download the results as CSV or Excel files, which can then be saved for further analysis and reporting.

[1190] Through these specific processing steps, PredictEase incorporates the user's emotional state while automating advanced data analysis, enabling highly accurate prediction and classification without specialized knowledge.

[1191] Example 2

[1192] 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."

[1193] Conventional data analysis systems are difficult for users without specialized knowledge to use, and require a lot of time and effort for data preprocessing and machine learning model selection. Furthermore, the systems operate without taking into account the user's emotional state, which can be stressful for users. Furthermore, it is difficult to perform highly accurate predictions and classifications while maintaining data quality.

[1194] 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.

[1195] In this invention, the server includes: [means for a user to upload a data file to be analyzed;] [means for the server to store the uploaded data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model; and [means for the server to visualize the results of prediction or classification and display them on a dashboard or output them in file format.] This makes it possible to easily perform highly accurate data analysis and prediction while taking into account the user's emotional state.

[1196] The "means for users to upload data files to be analyzed" refers to an interface that allows users to submit data files saved in formats such as CSV or Excel to the web application.

[1197] "Means by which the server stores uploaded data in its internal database" refers to the process by which the server receives data files uploaded by users, organizes and verifies the data, and stores it in its internal database.

[1198] "Means for the server to automatically perform data consistency checks and preprocessing" refers to a function that enables the server to detect missing or outlier values ​​in the data, complete or delete them, and automatically perform data preprocessing such as encoding categorical variables and normalizing numerical variables.

[1199] "Means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection" refers to the process in which the server analyzes the user's interactions such as keyboard input and mouse clicks based on an emotion engine, recognizes the user's emotional state, and dynamically adjusts the data preprocessing and machine learning model selection method based on this.

[1200] "Means in which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a method in which the server applies multiple machine learning algorithms, such as linear regression, random forest, decision tree, and support vector machine, trains and evaluates each model, and then selects the most appropriate model.

[1201] The "means for the server to make predictions or classifications using the selected model" is a function that makes predictions or classifications for new data using the optimal machine learning model that the server has previously selected.

[1202] "Means for the server to visualize the prediction or classification results and display them on a dashboard or output them in a file format" means the ability of the server to visualize the prediction or classification results obtained by the machine learning model in graph or table format and display them on a dashboard that can be accessed by the user, or output them in a file format such as CSV or Excel.

[1203] MODE FOR CARRYING OUT THE INVENTION

[1204] This invention relates to a system that streamlines data analysis, recognizes user emotions, and adjusts feedback. This system is based on the no-code data analysis tool "PredictEase," and can be used intuitively even by beginners.

[1205] In an example, a user first accesses the PredictEase web application through a web browser and uploads the data file to be analyzed, which can be in CSV or Excel format.

[1206] The server receives, validates, and stores data using a backend built in Python. Specifically, the received data is checked for consistency using the pandas library. Missing values ​​and outliers are detected and then appropriately imputed or deleted. This ensures the quality of the data.

[1207] Next, the data is preprocessed using scikit-learn: categorical variables are one-hot encoded and numerical variables are normalized or standardized. This preprocessing enables efficient and effective training with multiple machine learning algorithms.

[1208] Furthermore, the system incorporates an emotion engine that recognizes emotions by analyzing user behavior on the interface. For example, a user who frequently sees error messages can be determined to be impatient. Based on the analysis results of the emotion engine, the server dynamically adjusts data preprocessing and model selection strategies. This process reduces user stress and provides a comfortable user experience.

[1209] For model selection, the server uses scikit-learn and TensorFlow to apply multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains models for each algorithm. It evaluates the performance of each model using cross-validation and selects the best model. This training and evaluation process maximizes the accuracy of predictions and classifications.

[1210] After the best model is selected, hyperparameter tuning is performed using GridSearchCV to build the best performing model. The server then uses this optimized model to make predictions or classifications on new data.

[1211] Finally, the server visualizes the prediction or classification results and outputs them in a dashboard format that is easy for users to understand, using visualization tools such as matplotlib and Plotly. Users can check the results through the dashboard and download them in CSV or Excel format if necessary.

[1212] Specific examples

[1213] For example, if a marketing manager at a company wants to use monthly sales data to predict next month's sales, he or she can use PredictEase as follows:

[1214] 1. The user prepares past sales data in CSV format and accesses the PredictEase web application to upload it.

[1215] 2. The server receives the data and uses the pandas library to check the data for consistency, impute missing values, and handle outliers appropriately.

[1216] 3. Use scikit-learn to perform data preprocessing, one-hot encoding categorical variables, and normalizing or standardizing numerical variables.

[1217] 4. The server uses an emotion engine to recognize the user's emotional state and adjusts pre-processing and feedback strategies according to the user's level of impatience.

[1218] 5. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow and selects the best model using cross-validation.

[1219] 6. Use GridSearchCV to tune the hyperparameters of the selected model and build the optimal model.

[1220] 7. The server uses this optimal model to predict next month's sales, visualizes the prediction results, and displays them on a dashboard.

[1221] 8. Users can check the prediction results on the dashboard and download them in CSV format if necessary.

[1222] This allows PredictEase to significantly reduce the effort required for data analysis and provide highly accurate predictions while taking into account the user's emotional state.

[1223] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1224] PredictEase System Program Processing Steps

[1225] Step 1: Upload your data files

[1226] Input: User-uploaded data file in CSV or Excel format.

[1227] Specific behavior:

[1228] 1. A user accesses the web application and selects a data file.

[1229] 2. The user clicks the "Upload" button.

[1230] Data processing: None at this time.

[1231] Output: The uploaded file is sent to the server.

[1232] Step 2: Receiving and storing data

[1233] Input: User uploaded data file.

[1234] Specific behavior:

[1235] 1. The server receives the uploaded file.

[1236] 2. The server checks the file format and stores it in its internal database.

[1237] Data processing: Check the format of the received file.

[1238] Output: The path of the data file stored in the database.

[1239] Step 3: Data integrity check and preprocessing

[1240] Input: Data files stored in the internal database.

[1241] Specific behavior:

[1242] 1. The server uses the pandas library to check the integrity of the data.

[1243] 2. The server detects missing values ​​and fills them with DataFrame.fillna().

[1244] 3. The server detects outliers and handles them appropriately.

[1245] Data processing: missing value imputation, outlier treatment.

[1246] Output: Consistency checked and preprocessed data.

[1247] Step 4: Data Encoding and Normalization

[1248] Input: Data that has been integrity checked and preprocessed.

[1249] Specific behavior:

[1250] 1. The server encodes categorical variables using scikit-learn's OneHotEncoder.

[1251] 2. The server normalizes the numerical variables using scikit-learn's StandardScaler.

[1252] Data processing: categorical data encoding, numerical data normalization.

[1253] Output: The dataset after encoding and normalization.

[1254] Step 5: Recognizing the user's emotional state

[1255] Input: Data about user actions on the interface (clicks, keystrokes, etc.).

[1256] Specific behavior:

[1257] 1. The server passes the behavioral data to the emotion engine.

[1258] 2. The server receives the analysis results of the emotion engine.

[1259] Data calculation: User behavior data analysis.

[1260] Output: User's emotional state (stressed, anxious, etc.).

[1261] Step 6: Preprocessing and model selection adjustments

[1262] Input: User emotional state and the encoded and normalized dataset.

[1263] Specific behavior:

[1264] 1. The server adjusts data preprocessing and model selection methods based on the emotional state.

[1265] 2. The server simplifies the data preprocessing step.

[1266] Data processing: Adjusting data preprocessing strategies.

[1267] Output: Preprocessed data reflecting emotional states.

[1268] Step 7: Train the machine learning model

[1269] Input: Preprocessed data.

[1270] Specific behavior:

[1271] 1. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow.

[1272] 2. The server trains each model and performs cross-validation.

[1273] Data computation: model training, cross-validation.

[1274] Output: Performance evaluation results for each machine learning model.

[1275] Step 8: Selecting the best model

[1276] Input: Performance evaluation results for each model.

[1277] Specific behavior:

[1278] 1. The server selects the optimal model based on performance evaluation.

[1279] Data calculation: Comparison of performance evaluation results.

[1280] Output: The best model selected.

[1281] Step 9: Hyperparameter tuning

[1282] Input: The selected best model.

[1283] Specific behavior:

[1284] 1. The server optimizes the hyperparameters using GridSearchCV.

[1285] Data computation: Hyperparameter optimization.

[1286] Output: An optimized machine learning model.

[1287] Step 10: Perform prediction and classification

[1288] Input: Optimized machine learning model and new data.

[1289] Specific behavior:

[1290] 1. The server uses the optimized model to make predictions or classifications on new data.

[1291] Data calculations: Predicting or classifying new data.

[1292] Output: Prediction or classification results.

[1293] Step 11: Visualizing and outputting results

[1294] Input: Prediction or classification results.

[1295] Specific behavior:

[1296] 1. The server visualizes the results using matplotlib, Plotly, etc.

[1297] 2. The server visualizes the results and displays them on a dashboard.

[1298] 3. Export the results in CSV or Excel format as needed.

[1299] Data arithmetic: visualization and format transformation of results.

[1300] Output: Dashboard view and CSV / Excel format files.

[1301] (Application example 2)

[1302] 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."

[1303] Conventional data analysis systems do not take into account the user's emotional state, resulting in situations where users feel stressed or where data preprocessing and model selection are suboptimal. Furthermore, product recommendation systems on online shopping sites do not reflect user emotions and are limited to uniform recommendation accuracy. This prevents improvements to the user experience and increases in purchasing motivation.

[1304] 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.

[1305] In this invention, the server includes: [means for a user to upload a data file;] [means for the server to store the received data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to use an emotion engine that recognizes the user's emotional state;] [means for the server to adjust preprocessing and model selection based on the user's emotional state;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model;] [means for the server to visualize the results and display them on a dashboard or output them in file format;] [means for collecting user emotion data via a smartphone or other device; and [means for the server to recommend products.] This enables real-time recognition of the user's emotional state, corresponding data preprocessing and model selection, and even emotion-based product recommendations.

[1306] The "means for users to upload data files" refers to the means by which users add data files to be analyzed to the system.

[1307] The "means for storing data received by the server in an internal database" refers to the means for the server to store data files received from users in a database.

[1308] "Means for the server to automatically check data consistency and preprocess" refers to automated means for the server to check data consistency and handle missing values ​​and outliers.

[1309] The "means for the server to use an emotion engine to recognize the user's emotional state" refers to a means for utilizing an engine to detect and analyze the user's emotions and identify a specific emotional state.

[1310] The "means for the server to adjust preprocessing and model selection based on the emotional state of the user" refers to a means for changing the process of data preprocessing and appropriate model selection in accordance with the emotional state of the user detected using the emotion engine.

[1311] "Means in which the server applies multiple machine learning algorithms and selects the most appropriate model" refers to a means in which data is analyzed using multiple machine learning technologies and the optimal algorithm is selected from among them.

[1312] The "means for the server to make predictions or classifications using the selected model" refers to the means for making predictions or classifications of data using the optimal trained model.

[1313] The "means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to a means for displaying the results of prediction or classification in a visually easy-to-understand manner on a dashboard or outputting them as a file.

[1314] "Means for collecting user emotional data via smartphones or other devices" refers to means for collecting user emotional information in real time using smartphones or other sensor devices.

[1315] The "means for the server to recommend products" refers to a means for recommending products suitable for a user based on collected data and the user's emotional state.

[1316] This invention is an emotion-adaptive product recommendation system that applies the PredictEase system, which incorporates an emotion engine that recognizes the user's emotional state, to an online shopping site. This system can be accessed by users via smartphones or other devices.

[1317] Data Entry and Preprocessing

[1318] Users access the online shopping site's application and upload the necessary data files. The server receives the data and stores it in an internal database. At that time, the server checks the data for consistency and automatically processes missing or outlier values.

[1319] Recognition of emotional states

[1320] The server collects user emotional data via smartphones and other devices (cameras, microphones, emotion-sensing devices, etc.). The emotion engine analyzes this data and identifies the user's emotional state. For example, it uses facial recognition and voice analysis technology to determine whether the user is feeling stressed.

[1321] Model selection and tuning

[1322] Based on the user's emotional state identified by the emotion engine, the server adaptively adjusts the data preprocessing and model selection processes, thereby shortening processing time and providing easy-to-understand feedback to users experiencing stress.

[1323] The server analyzes the data by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, etc.) to select the optimal model, and then performs hyperparameter tuning of the selected model to achieve optimal performance.

[1324] Predictions and product recommendations

[1325] The server uses the best-fit model trained on the system to make predictions or classifications based on the user's data. It also considers the user's emotional state and purchasing history to recommend suitable products. The results are visualized and displayed intuitively on the user's application screen. Users can also download the results as a file if desired.

[1326] Hardware and software used

[1327] Hardware: Smartphone (camera, microphone), emotion sensing device (e.g., EEG headset)

[1328] Software: Sentiment analysis software (e.g., Sentiment API), machine learning libraries (e.g., TensorFlow, Scikit-learn)

[1329] Examples and prompts

[1330] For example, if a user is feeling stressed, the "Emotion-Adaptive Smart Recommender" will recommend products using prompts like the following:

[1331] User Emotion: Stress

[1332] Past purchase history: Aroma candles, relaxation CDs, herbal tea

[1333] Recommended Products:

[1334] 1. Lavender aroma oil

[1335] 2. Yoga mat

[1336] 3. Relaxation Cafe Gift Card

[1337] This system can recommend optimal products by taking into account the user's emotional state in real time, which is expected to improve the user experience and satisfaction.

[1338] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1339] Step 1:

[1340] The user accesses the application of the online shopping site and uploads the data file to be analyzed (e.g., a CSV file).

[1341] Input: Data file uploaded by the user.

[1342] Output: The data file that is transferred to the server.

[1343] Specific operation: The user uses a smartphone or personal computer to select a data file from the file selection screen and click the upload button.

[1344] Step 2:

[1345] The server stores the received data files in an internal database, performs data consistency checks, and automatically handles missing and outlier values.

[1346] Input: Data file received from user.

[1347] Output: Preprocessed data that has passed integrity checks.

[1348] What it does: The server stores the data in a database and runs scripts to impute missing values ​​(e.g., impute with the mean) and remove / correct outliers.

[1349] Step 3:

[1350] The server collects user emotional data via the smartphone's camera, microphone, and emotion sensing device.

[1351] Input: User facial recognition data, voice data, and input from other sensors.

[1352] Output: Data on the user's emotional state.

[1353] Specific operation: The server performs facial expression analysis from camera images using an emotion recognition API, and performs emotion analysis from microphone audio.

[1354] Step 4:

[1355] The server uses an emotion engine to analyze the collected emotion data and identify the user's emotional state.

[1356] Input: Collected emotion data.

[1357] Output: The user's specific emotional state (e.g., stress, joy, excitement).

[1358] Specific operation: The server runs the emotion engine, analyzes the data, and outputs the emotional state as a number or category.

[1359] Step 5:

[1360] The server adaptively adjusts the data preprocessing and model selection processes based on the user's emotional state.

[1361] Input: The user's emotional state.

[1362] Output: Adjusted preprocessing parameters and model selection strategy.

[1363] What it does: The server considers the emotion data and performs simplified pre-processing to reduce processing time or generates easy-to-understand feedback.

[1364] Step 6:

[1365] The server applies multiple machine learning algorithms and selects the most suitable model.

[1366] Input: Preprocessed data.

[1367] Output: Best model via cross-validation.

[1368] What it does: The server applies different algorithms (e.g., linear regression, random forest) and selects the best model based on the results of cross-validation.

[1369] Step 7:

[1370] The server tunes the hyperparameters of the selected model to achieve optimal performance.

[1371] Input: A cross-validated model.

[1372] Output: The optimal tuned model.

[1373] Specific operation: Set the range of hyperparameters and search for the optimal parameters using grid search or random search.

[1374] Step 8:

[1375] The server uses the best trained model to make predictions and product recommendations.

[1376] Input: User data and emotional state.

[1377] Output: A list of product recommendations.

[1378] What it does: Apply the model to rank and recommend products based on user sentiment and purchase history.

[1379] Step 9:

[1380] The server visualizes the recommendation results and displays them on the user's application screen.

[1381] Input: A list of recommended products.

[1382] Output: Visualized dashboard or file format.

[1383] What it does: Generates results in the form of graphs and lists, displays them in a dashboard that is intuitive and easy for users to understand, and provides the results as a downloadable file if required.

[1384] 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.

[1385] 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.

[1386] 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.

[1387] [Fourth embodiment]

[1388] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1389] 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.

[1390] 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).

[1391] 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.

[1392] 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.

[1393] 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).

[1394] 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.

[1395] 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.

[1396] 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.

[1397] 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.

[1398] 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.

[1399] 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.

[1400] 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."

[1401] This invention relates to PredictEase, a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. Specifically, the system is composed of the following steps:

[1402] System Overview

[1403] Data Entry

[1404] 1. A user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[1405] 2. The server receives the uploaded data file and stores it in its internal database.

[1406] Data Preprocessing

[1407] 1. The server performs a consistency check on the data file to determine if there are any missing or outlier values.

[1408] 2. The server handles missing and outliers in an appropriate way and converts the data into a form that can be applied to the machine learning model.

[1409] 3. The server encodes categorical variables and normalizes or standardizes numeric variables.

[1410] Model Selection

[1411] 1. The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[1412] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[1413] Model Optimization

[1414] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[1415] Prediction and Classification

[1416] 1. The server uses the trained model to make a prediction or classification.

[1417] 2. The server aggregates the prediction and classification results and evaluates their performance.

[1418] Output of results

[1419] 1. The server visually displays prediction and classification results on a dashboard.

[1420] 2. Display the results in an interface so that users can view them in graphs and tables.

[1421] 3. The server makes the results available for download as a CSV or Excel file, if desired.

[1422] Specific examples

[1423] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[1424] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[1425] 2. The server stores the received sales data in an internal database and starts checking the data for consistency. For example, if there are missing values ​​in the sales data, it fills them with the average value and handles any outliers appropriately.

[1426] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[1427] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[1428] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[1429] 6. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[1430] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[1431] This makes it easy for users without data science knowledge to perform advanced data analysis and predictions, making PredictEase a powerful tool for supporting data-driven decision-making.

[1432] The processing flow will be explained below.

[1433] Step 1:

[1434] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[1435] Step 2:

[1436] The server receives the uploaded data file, checks whether it is in the correct format, and then stores the data file in its internal database.

[1437] Step 3:

[1438] The server performs data integrity checks, specifically detecting missing and outlier values ​​and handling them in an appropriate way (e.g., imputing with the mean, removing outliers).

[1439] Step 4:

[1440] The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables to convert the data into a format that can be applied to machine learning models.

[1441] Step 5:

[1442] The server analyzes the features in the dataset and evaluates correlations and the importance of variables, thereby selecting important features and eliminating unnecessary features.

[1443] Step 6:

[1444] The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) to train each model.

[1445] Step 7:

[1446] The server evaluates the performance of each model using cross-validation, calculates evaluation metrics such as accuracy and F1 score for each model, and selects the optimal model.

[1447] Step 8:

[1448] The server tunes the hyperparameters of the selected model, using grid search or random search to find the optimal hyperparameters and build the final version of the model.

[1449] Step 9:

[1450] The server performs prediction or classification using the prepared test data, and then aggregates the prediction and classification results to evaluate their performance.

[1451] Step 10:

[1452] The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[1453] Step 11:

[1454] Users access the dashboard and view prediction and classification results displayed in graphs and tables.

[1455] Step 12:

[1456] If desired, users can download the results as a CSV or Excel file.

[1457] Through this series of steps, PredictEase enables users to easily perform highly accurate data analysis and predictions, even without data science knowledge.

[1458] Example 1

[1459] 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."

[1460] Current data analysis and prediction systems are difficult to use effectively unless users have specialized data science knowledge. They also require complex procedures such as data preprocessing, optimal model selection, and hyperparameter tuning, so there is a need for a system that can perform these processes consistently and automatically. Furthermore, there is a need for a system that can easily visualize and output the results.

[1461] 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.

[1462] In this invention, the server includes means for users to upload data files, means for the computer to store the received data in an internal storage device, means for the computer to automatically check the consistency of the data and preprocess it, means for the computer to apply multiple machine learning algorithms and select the most suitable model, means for the computer to make predictions or classifications using the selected model, means for the computer to visualize the results and display them on a display device or output them in file format, and means for users to check the results and make them downloadable. This allows even users without specialized knowledge to easily and efficiently perform processes from uploading data to obtaining prediction results.

[1463] "User" refers to a person who uses this system, uploads data, and obtains analysis and prediction results.

[1464] A "data file" refers to a collection of digital data, such as CSV or Excel format, that is input into a system for analysis or prediction.

[1465] "Computer" refers to a collection of hardware and software for receiving, processing, analyzing data, and outputting results.

[1466] "Internal storage device" refers to a storage medium that a computer uses to temporarily or permanently store data files.

[1467] "Integrity check" refers to the process of verifying that a data file conforms to the expected format and structure and that there are no missing or outlier values.

[1468] "Preprocessing" refers to a series of steps that transform, filter, and encode data to improve the accuracy of analysis or prediction.

[1469] "Machine learning algorithms" refer to mathematical models and methods that automatically learn patterns and knowledge from data and make predictions and classifications.

[1470] A "model" is a computational system trained by machine learning algorithms that has the ability to make predictions or classifications based on input data.

[1471] "Hyperparameter tuning" refers to the process of adjusting the configuration parameters of an algorithm to optimize the performance of a machine learning model.

[1472] "Visualization" refers to the process of representing prediction and classification results as graphs or charts so that they can be understood intuitively.

[1473] "Display device" refers to hardware such as a monitor or display that presents visualized results to a user.

[1474] "Downloading" refers to the act of a user transferring and saving a data file of prediction or classification results from the system to their own device.

[1475] This invention relates to a no-code tool that enables users without specialized data science knowledge to easily predict and classify data. In this system, users upload data files, and the server automatically preprocesses the data, selects and optimizes models, and predicts and visualizes the results.

[1476] System Overview

[1477] Data Entry

[1478] The process begins when a user accesses the web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The server receives the uploaded data file and stores it in its internal storage.

[1479] Data Preprocessing

[1480] The server performs a consistency check on the data files to check for missing or outliers. It then imputes missing values ​​with the mean and handles outliers appropriately. It also encodes categorical variables and normalizes or standardizes numerical variables. This process is important as a preprocessing step for prediction and classification.

[1481] Model Selection and Optimization

[1482] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It evaluates the performance of each model using cross-validation and selects the best model. It then optimizes the hyperparameters of the selected model using grid search or random search. This step improves the model's predictive accuracy.

[1483] Prediction and visualization

[1484] The server then uses the best trained model to make predictions and classifications for new data. The results are visualized and displayed on a dashboard, presented to users in the form of graphs and tables. Users can also download the results as CSV or Excel files for offline analysis.

[1485] Specific examples

[1486] For example, if a user in a company's planning department wants to forecast monthly sales data, they can use the system as follows:

[1487] 1. The user prepares past sales data in CSV file format and accesses the web application to upload the file.

[1488] 2. The server stores the received sales data in its internal storage device and starts checking the data for consistency. If missing values ​​are found, they are filled in with the average value, and any outliers are dealt with appropriately.

[1489] 3. The server preprocesses the data, encoding categorical variables and normalizing numerical variables.

[1490] 4. The server applies multiple machine learning algorithms to train each model, such as linear regression, random forests, and decision trees.

[1491] 5. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[1492] 6. The server uses the optimal model to predict sales for the next month and displays the forecast results on the dashboard.

[1493] 7. Users can view the forecast results displayed in graphs and tables on the dashboard to help them make strategic decisions, and can also download the forecast results in CSV format if necessary.

[1494] Prompt Sentence Examples

[1495] "Use the following sales data to predict next month's sales. Use linear regression and random forest as your models. Impute missing values ​​with the mean and treat outliers appropriately."

[1496] In this way, the system efficiently processes everything from data input to output, providing users with highly accurate prediction results. This process eliminates the need for specialized knowledge, resulting in a system that is convenient and user-friendly.

[1497] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1498] Step 1:

[1499] A user accesses the PredictEase web application, selects and uploads the data file to be analyzed (e.g., a CSV or Excel file). The input is the data file to be analyzed, and the output is the uploaded data file sent to the server.

[1500] Step 2:

[1501] The server receives the uploaded data file and saves it to its internal storage. The data file "sales_data.csv" is saved to the server. The input is the uploaded data file, and the output is saving the data file to its internal storage.

[1502] Step 3:

[1503] The server performs a consistency check on the data file. Specifically, it checks whether the number of columns and data types match, and checks for missing or outlier values. The input is the saved data file, and the output is the data consistency check result.

[1504] Step 4:

[1505] The server imputes missing values ​​with the mean value and filters or imputes outliers, if any. The input is data that has been checked for consistency, and the output is preprocessed data.

[1506] Step 5:

[1507] The server encodes categorical variables into numeric values ​​and normalizes or standardizes numeric variables. For example, categorical variables such as "store name" are one-hot encoded, and numeric variables such as "sales amount" are processed with min-max scaling. The input is the preprocessed data, and the output is the encoded and normalized data.

[1508] Step 6:

[1509] The server trains the model by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.). The input is the encoded and normalized data, and the output is a model trained with each algorithm.

[1510] Step 7:

[1511] The server evaluates the performance of each model using cross-validation and selects the best model. For example, cross-validation evaluates the mean squared error (MSE) or coefficient of determination (R²). The input is the trained model, and the output is the best model.

[1512] Step 8:

[1513] The server optimizes the hyperparameters of the selected optimal model using grid search or random search. The input is the optimal model, and the output is the model with optimized parameters.

[1514] Step 9:

[1515] The server uses the best-fit model to predict or classify new data. The input is the new data, and the output is the prediction or classification result.

[1516] Step 10:

[1517] The server aggregates the prediction and classification results and calculates evaluation metrics such as accuracy and precision. The input is the prediction and classification results, and the output is a performance evaluation report.

[1518] Step 11:

[1519] The server visualizes the prediction and classification results and displays them on a dashboard. Users can check the results displayed in graphs and tables on the dashboard. The input is a performance evaluation report, and the output is a visualized dashboard display.

[1520] Step 12:

[1521] The server optionally makes the results available for download as CSV or Excel files, which users can download and analyze offline. The input is the visualized dashboard data, and the output is a downloadable file.

[1522] (Application example 1)

[1523] 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."

[1524] For modern self-driving vehicles, it is essential to analyze massive amounts of operational data in real time to detect and predict anomalies. However, this requires advanced knowledge of data science, which is a significant hurdle for non-expert users. Additionally, the pre-processing and modeling of operational data is complex and difficult to perform efficiently. There is a need for a system that can solve these problems and enable anyone to easily analyze data from self-driving vehicles.

[1525] 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.

[1526] In this invention, the server includes means for users to upload data files, means for the server to store the received data in an internal database, means for the server to automatically perform data consistency checks and preprocessing, means for the server to apply multiple machine learning algorithms and select the most appropriate model, means for the server to perform prediction or classification using the selected model, means for the server to visualize the results and display them on a dashboard or output them in file format, means for the server to analyze operational data of autonomous vehicles via an application installed on a smartphone and perform anomaly detection and prediction, and means for storing the trained model in the smartphone. This enables even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and prediction in real time.

[1527] "Means for users to upload data files" refers to an interface through which users submit files containing autonomous vehicle operation data to the web application.

[1528] "Means for storing the data received by the server in an internal database" refers to a processing device and storage technology for storing the received operation data in a database within the server.

[1529] "Means for the server to automatically perform data consistency checks and preprocessing" refers to technology in which the server checks operational data for missing values, abnormal values, etc., and automatically performs the necessary preprocessing (e.g., filling in missing values, cleaning data).

[1530] "A means by which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a technology in which the server tries out multiple models, evaluates their performance, and selects the model with the highest predictive accuracy.

[1531] "Means for the server to make predictions or classifications using the selected model" refers to a technology for making predictions or classifications for new operational data using the selected optimal model.

[1532] "Means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to the function by which the server displays the prediction or classification results in a visually easy-to-understand dashboard and outputs them in a file format (e.g., CSV, Excel) if necessary.

[1533] "Means of analyzing driving data of autonomous vehicles via an application installed on a smartphone to detect and predict abnormalities" refers to technology that analyzes driving data via a dedicated application on a smartphone, detects abnormal driving patterns, and predicts future driving conditions.

[1534] "Means for storing trained models within a smartphone" refers to technology that stores pre-trained machine learning models in a smartphone's storage device and makes them accessible to applications.

[1535] This invention relates to a system that enables users without specialized data science knowledge to easily analyze operational data of autonomous vehicles and perform anomaly detection and prediction. Specifically, the system is configured with the following steps.

[1536] System Overview

[1537] 1. Data Entry

[1538] Users upload autonomous vehicle operation data files (e.g., CSV or Excel files) via an application installed on their smartphones, and the uploaded data is sent to a server.

[1539] 2. Receipt and storage of data

[1540] The server stores the received operational data in an internal database, where it is quickly and securely stored and used for subsequent data analysis processes.

[1541] 3. Data integrity check and preprocessing

[1542] The server automatically performs integrity checks on the stored data, checking for missing or outlier values, and performs appropriate pre-processing, including imputing missing values ​​and cleaning the data.

[1543] 4. Model Selection

[1544] The server tries multiple machine learning algorithms and selects the best model, such as a random forest or a support vector machine. The selected model is then evaluated using cross-validation.

[1545] 5. Optimizing and training the model

[1546] The server then adjusts the hyperparameters of the selected model to achieve optimal performance, and the trained model is stored on the smartphone's memory.

[1547] 6. Prediction and Anomaly Detection

[1548] The selected model is used to predict and classify new driving data from the autonomous vehicle. The smartphone receives the analysis results and predicts abnormal driving patterns and future driving situations.

[1549] 7. Visualizing and outputting results

[1550] The server visually displays prediction and classification results on a dashboard. Users can view the results in graphs and tables through a smartphone interface. Users can also download the results in file format (e.g., CSV or Excel) if desired.

[1551] This allows even non-expert users to easily analyze data from autonomous vehicles and perform anomaly detection and future predictions in real time.

[1552] Hardware and software used

[1553] Smartphone: Install the application and use it to upload data and display analysis results.

[1554] Server: Receives data, stores it, checks its integrity, preprocesses it, and selects, trains, and optimizes the model.

[1555] Python: A programming language used for data analysis and model training.

[1556] Pandas: A library used to load and preprocess data.

[1557] Scikit-learn: A library used to implement machine learning algorithms.

[1558] Joblib: A library used to save trained models.

[1559] Specific examples

[1560] For example, if a user wants to upload operational data (speed, GPS information, sensor information, etc.) from an autonomous vehicle and detect abnormal driving patterns, they can use the PredictDrive system via a smartphone application. The server receives the data, performs preprocessing, selects the optimal machine learning model, and trains it. The trained model is stored on the smartphone, and when new operational data is entered, it detects abnormalities in real time and displays the prediction results on a dashboard.

[1561] Prompt Sentence Examples

[1562] "Please tell us your next use case: Create a Python script that uses past driving data to predict whether the next driving pattern will be anomalous. The driving data includes speed, GPS information, and various sensor information. Use a random forest as the model, and create a visualization plot to present the results. Include detailed missing value handling and data standardization methods."

[1563] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1564] Step 1:

[1565] The user uploads the autonomous vehicle's operation data file (e.g., CSV format) using an application installed on their smartphone. This input data includes speed, GPS information, and sensor information. The data is then sent to the server via the smartphone.

[1566] Step 2:

[1567] The server stores the received operational data in an internal database. The stored data is used for subsequent processing within the server. The input here is the uploaded operational data, and the output is the data stored in the database.

[1568] Step 3:

[1569] The server performs consistency checks on the stored data. Specifically, it checks for missing values ​​and outliers and performs data cleaning. For example, missing values ​​are imputed with the mean value. The input of this step is the stored data, and the output is the cleaned data.

[1570] Step 4:

[1571] The server preprocesses the data by encoding categorical variables and normalizing or standardizing numerical variables. Encoding converts categorical data into a numerical representation, and normalization is achieved by scaling the data range. The input is the cleaned data, and the output is the preprocessed data.

[1572] Step 5:

[1573] The server tries multiple machine learning algorithms and applies and trains models such as random forests and support vector machines. It evaluates the performance of each model through cross-validation and selects the most suitable model. The input is the preprocessed data, and the output is the selected optimal model.

[1574] Step 6:

[1575] The server adjusts the hyperparameters of the selected model and tunes it to achieve optimal performance. This process involves an iterative process to optimize the model parameters. The input is the selected model, and the output is the optimized model.

[1576] Step 7:

[1577] The trained optimization model is stored locally on the smartphone, which is then used for real-time prediction and classification. The input here is the optimization model, and the output is the model stored locally on the smartphone.

[1578] Step 8:

[1579] The smartphone uses the stored model to make predictions and detect anomalies in real time whenever new driving data is input. For example, if an abnormal driving pattern is detected, an alert is sent to the user. The input is new driving data, and the output is the prediction or anomaly detection result and a user notification.

[1580] Step 9:

[1581] The server visually displays the prediction and classification results on a dashboard. Users can check the results on their smartphones and, if necessary, download the results in file format (e.g., CSV or Excel). The input here is the prediction or classification results, and the output is the visualized dashboard and file output.

[1582] 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.

[1583] This invention combines the PredictEase system, which performs data prediction and classification, with an emotion engine that recognizes user emotions. This system is designed as a no-code tool that can be easily used by users without data science expertise, and provides appropriate feedback and preprocessing adjustments according to the user's emotional state.

[1584] System Overview

[1585] Data Entry

[1586] 1. The user accesses the PredictEase web application and selects and uploads the data file to be analyzed (e.g., a CSV or Excel file).

[1587] 2. The server receives the uploaded data file, checks its format, and then stores it in its internal database.

[1588] Data Preprocessing

[1589] 1. The server performs data integrity checks to detect missing or outlier values ​​and handles them in an appropriate way.

[1590] 2. The server encodes categorical variables and normalizes or standardizes numeric variables.

[1591] Using the Emotion Engine

[1592] 1. The server passes the user's input and actions on the interface to the emotion engine to recognize the user's emotions.

[1593] 2. The server uses the emotion engine to adjust data preprocessing and model selection strategies according to the user's emotional state. For example, if the user is impatient, it can shorten the preprocessing time or provide easy-to-understand feedback to the user.

[1594] Model Selection

[1595] 1. The server applies multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model.

[1596] 2. The server evaluates the performance of each model using cross-validation and selects the best model.

[1597] Model Optimization

[1598] 1. The server tunes the hyperparameters of the selected model and builds a model that achieves optimal performance.

[1599] Prediction and Classification

[1600] 1. The server uses the trained model to make a prediction or classification.

[1601] 2. The server aggregates the prediction and classification results and evaluates their performance.

[1602] Output of results

[1603] 1. The server visualizes the prediction and classification results and converts them into a format for display on a dashboard.

[1604] 2. The user accesses the dashboard and checks the prediction and classification results displayed in graphs and tables.

[1605] 3. If desired, the user can download the results as a CSV or Excel file.

[1606] Specific examples

[1607] For example, if a user in a company's planning department wants to predict monthly sales data, they can use PredictEase as follows:

[1608] 1. The user prepares past sales data in CSV file format and accesses the PredictEase web application to upload the file.

[1609] 2. The server stores the received sales data in an internal database and performs data consistency checks, such as filling in missing values ​​with the average value and handling outliers appropriately.

[1610] 3. The server performs data preprocessing, encoding categorical variables and normalizing or standardizing numerical variables.

[1611] 4. The server uses an emotion engine to recognize the user's emotional state based on their input and actions. For example, if the server determines that the user is impatient, it summarizes the processing results and provides quick feedback.

[1612] 5. The server applies multiple machine learning algorithms to train each model, for example, linear regression, random forest, decision tree, etc.

[1613] 6. The server evaluates the performance of the models using cross-validation, selects the optimal model, and tunes the hyperparameters of the selected model to build the optimal model.

[1614] 7. The server uses the best model to predict next month's sales and displays the forecast results on the dashboard.

[1615] 8. The user can check the forecast results displayed in graphs and tables on the dashboard and download the results in CSV format if necessary.

[1616] This allows PredictEase to provide advanced data analysis and predictions while taking into account the user's emotional state. Even users without data science knowledge can use this system to obtain highly accurate data analysis results and make strategic decisions.

[1617] The processing flow will be explained below.

[1618] Step 1:

[1619] The user accesses the PredictEase web application, selects the data file to be analyzed (such as a CSV or Excel file), and uploads it.

[1620] Step 2:

[1621] The server receives the uploaded data file, checks whether it is in the correct format, and then stores it in the internal database. If the format is incorrect, an error message is displayed to the user.

[1622] Step 3:

[1623] The server checks the integrity of the data file, detecting missing and abnormal values ​​(e.g., extreme values ​​and outliers) and performing other processing such as filling in the missing values ​​with the mean.

[1624] Step 4:

[1625] The server performs data preprocessing, such as label encoding and one-hot encoding for categorical variables and standardization and normalization for numerical variables, to convert the data into a form suitable for use in machine learning models.

[1626] Step 5:

[1627] The server uses an emotion engine to recognize the user's emotional state, for example by analyzing the video captured by the user's webcam, mouse movements, typing speed, etc.

[1628] Step 6:

[1629] The server adjusts its data preprocessing and model selection strategies based on the results of the emotion engine: for example, if the user is impatient, it will shorten the preprocessing time and present the results in an easy-to-understand format.

[1630] Step 7:

[1631] The server tries multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains each model. It splits the data into a training dataset and a test dataset and proceeds with learning for each model.

[1632] Step 8:

[1633] The server performs cross-validation and evaluates the performance of each model, calculating evaluation metrics such as accuracy, F1 score, and R^2 value, and selecting the best model.

[1634] Step 9:

[1635] The server performs hyperparameter tuning for the selected model, for example, using grid search or random search to find the optimal parameter set.

[1636] Step 10:

[1637] The server uses the optimized model to make predictions and classifications, and performs final evaluation of the model using test data.

[1638] Step 11:

[1639] The server visualizes the prediction and classification results, converting them into a format that can be displayed in graphs and tables on a dashboard.

[1640] Step 12:

[1641] Users can access the dashboard to see the results and can click on graphs and tables to access more detailed information.

[1642] Step 13:

[1643] If desired, users can download the results as CSV or Excel files, which can then be saved for further analysis and reporting.

[1644] Through these specific processing steps, PredictEase incorporates the user's emotional state while automating advanced data analysis, enabling highly accurate prediction and classification without specialized knowledge.

[1645] Example 2

[1646] 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."

[1647] Conventional data analysis systems are difficult for users without specialized knowledge to use, and require a lot of time and effort for data preprocessing and machine learning model selection. Furthermore, the systems operate without taking into account the user's emotional state, which can be stressful for users. Furthermore, it is difficult to perform highly accurate predictions and classifications while maintaining data quality.

[1648] 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.

[1649] In this invention, the server includes: [means for a user to upload a data file to be analyzed;] [means for the server to store the uploaded data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model; and [means for the server to visualize the results of prediction or classification and display them on a dashboard or output them in file format.] This makes it possible to easily perform highly accurate data analysis and prediction while taking into account the user's emotional state.

[1650] The "means for users to upload data files to be analyzed" refers to an interface that allows users to submit data files saved in formats such as CSV or Excel to the web application.

[1651] "Means by which the server stores uploaded data in its internal database" refers to the process by which the server receives data files uploaded by users, organizes and verifies the data, and stores it in its internal database.

[1652] "Means for the server to automatically perform data consistency checks and preprocessing" refers to a function that enables the server to detect missing or outlier values ​​in the data, complete or delete them, and automatically perform data preprocessing such as encoding categorical variables and normalizing numerical variables.

[1653] "Means for the server to recognize the user's actions on the interface using an emotion engine and adjust preprocessing and model selection" refers to the process in which the server analyzes the user's interactions such as keyboard input and mouse clicks based on an emotion engine, recognizes the user's emotional state, and dynamically adjusts the data preprocessing and machine learning model selection method based on this.

[1654] "Means in which the server applies multiple machine learning algorithms to select the most appropriate model" refers to a method in which the server applies multiple machine learning algorithms, such as linear regression, random forest, decision tree, and support vector machine, trains and evaluates each model, and then selects the most appropriate model.

[1655] The "means for the server to make predictions or classifications using the selected model" is a function that makes predictions or classifications for new data using the optimal machine learning model that the server has previously selected.

[1656] "Means for the server to visualize the prediction or classification results and display them on a dashboard or output them in a file format" means the ability of the server to visualize the prediction or classification results obtained by the machine learning model in graph or table format and display them on a dashboard that can be accessed by the user, or output them in a file format such as CSV or Excel.

[1657] MODE FOR CARRYING OUT THE INVENTION

[1658] This invention relates to a system that streamlines data analysis, recognizes user emotions, and adjusts feedback. This system is based on the no-code data analysis tool "PredictEase," and can be used intuitively even by beginners.

[1659] In an example, a user first accesses the PredictEase web application through a web browser and uploads the data file to be analyzed, which can be in CSV or Excel format.

[1660] The server receives, validates, and stores data using a backend built in Python. Specifically, the received data is checked for consistency using the pandas library. Missing values ​​and outliers are detected and then appropriately imputed or deleted. This ensures the quality of the data.

[1661] Next, the data is preprocessed using scikit-learn: categorical variables are one-hot encoded and numerical variables are normalized or standardized. This preprocessing enables efficient and effective training with multiple machine learning algorithms.

[1662] Furthermore, the system incorporates an emotion engine that recognizes emotions by analyzing user behavior on the interface. For example, a user who frequently sees error messages can be determined to be impatient. Based on the analysis results of the emotion engine, the server dynamically adjusts data preprocessing and model selection strategies. This process reduces user stress and provides a comfortable user experience.

[1663] For model selection, the server uses scikit-learn and TensorFlow to apply multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, support vector machine, etc.) and trains models for each algorithm. It evaluates the performance of each model using cross-validation and selects the best model. This training and evaluation process maximizes the accuracy of predictions and classifications.

[1664] After the best model is selected, hyperparameter tuning is performed using GridSearchCV to build the best performing model. The server then uses this optimized model to make predictions or classifications on new data.

[1665] Finally, the server visualizes the prediction or classification results and outputs them in a dashboard format that is easy for users to understand, using visualization tools such as matplotlib and Plotly. Users can check the results through the dashboard and download them in CSV or Excel format if necessary.

[1666] Specific examples

[1667] For example, if a marketing manager at a company wants to use monthly sales data to predict next month's sales, he or she can use PredictEase as follows:

[1668] 1. The user prepares past sales data in CSV format and accesses the PredictEase web application to upload it.

[1669] 2. The server receives the data and uses the pandas library to check the data for consistency, impute missing values, and handle outliers appropriately.

[1670] 3. Use scikit-learn to perform data preprocessing, one-hot encoding categorical variables, and normalizing or standardizing numerical variables.

[1671] 4. The server uses an emotion engine to recognize the user's emotional state and adjusts pre-processing and feedback strategies according to the user's level of impatience.

[1672] 5. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow and selects the best model using cross-validation.

[1673] 6. Use GridSearchCV to tune the hyperparameters of the selected model and build the optimal model.

[1674] 7. The server uses this optimal model to predict next month's sales, visualizes the prediction results, and displays them on a dashboard.

[1675] 8. Users can check the prediction results on the dashboard and download them in CSV format if necessary.

[1676] This allows PredictEase to significantly reduce the effort required for data analysis and provide highly accurate predictions while taking into account the user's emotional state.

[1677] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1678] PredictEase System Program Processing Steps

[1679] Step 1: Upload your data files

[1680] Input: User-uploaded data file in CSV or Excel format.

[1681] Specific behavior:

[1682] 1. A user accesses the web application and selects a data file.

[1683] 2. The user clicks the "Upload" button.

[1684] Data processing: None at this time.

[1685] Output: The uploaded file is sent to the server.

[1686] Step 2: Receiving and storing data

[1687] Input: User uploaded data file.

[1688] Specific behavior:

[1689] 1. The server receives the uploaded file.

[1690] 2. The server checks the file format and stores it in its internal database.

[1691] Data processing: Check the format of the received file.

[1692] Output: The path of the data file stored in the database.

[1693] Step 3: Data integrity check and preprocessing

[1694] Input: Data files stored in the internal database.

[1695] Specific behavior:

[1696] 1. The server uses the pandas library to check the integrity of the data.

[1697] 2. The server detects missing values ​​and fills them with DataFrame.fillna().

[1698] 3. The server detects outliers and handles them appropriately.

[1699] Data processing: missing value imputation, outlier treatment.

[1700] Output: Consistency checked and preprocessed data.

[1701] Step 4: Data Encoding and Normalization

[1702] Input: Data that has been integrity checked and preprocessed.

[1703] Specific behavior:

[1704] 1. The server encodes categorical variables using scikit-learn's OneHotEncoder.

[1705] 2. The server normalizes the numerical variables using scikit-learn's StandardScaler.

[1706] Data processing: categorical data encoding, numerical data normalization.

[1707] Output: The dataset after encoding and normalization.

[1708] Step 5: Recognizing the user's emotional state

[1709] Input: Data about user actions on the interface (clicks, keystrokes, etc.).

[1710] Specific behavior:

[1711] 1. The server passes the behavioral data to the emotion engine.

[1712] 2. The server receives the analysis results of the emotion engine.

[1713] Data calculation: User behavior data analysis.

[1714] Output: User's emotional state (stressed, anxious, etc.).

[1715] Step 6: Preprocessing and model selection adjustments

[1716] Input: User emotional state and the encoded and normalized dataset.

[1717] Specific behavior:

[1718] 1. The server adjusts data preprocessing and model selection methods based on the emotional state.

[1719] 2. The server simplifies the data preprocessing step.

[1720] Data processing: Adjusting data preprocessing strategies.

[1721] Output: Preprocessed data reflecting emotional states.

[1722] Step 7: Train the machine learning model

[1723] Input: Preprocessed data.

[1724] Specific behavior:

[1725] 1. The server applies multiple machine learning algorithms using scikit-learn and TensorFlow.

[1726] 2. The server trains each model and performs cross-validation.

[1727] Data computation: model training, cross-validation.

[1728] Output: Performance evaluation results for each machine learning model.

[1729] Step 8: Selecting the best model

[1730] Input: Performance evaluation results for each model.

[1731] Specific behavior:

[1732] 1. The server selects the optimal model based on performance evaluation.

[1733] Data calculation: Comparison of performance evaluation results.

[1734] Output: The best model selected.

[1735] Step 9: Hyperparameter tuning

[1736] Input: The selected best model.

[1737] Specific behavior:

[1738] 1. The server optimizes the hyperparameters using GridSearchCV.

[1739] Data computation: Hyperparameter optimization.

[1740] Output: An optimized machine learning model.

[1741] Step 10: Perform prediction and classification

[1742] Input: Optimized machine learning model and new data.

[1743] Specific behavior:

[1744] 1. The server uses the optimized model to make predictions or classifications on new data.

[1745] Data calculations: Predicting or classifying new data.

[1746] Output: Prediction or classification results.

[1747] Step 11: Visualizing and outputting results

[1748] Input: Prediction or classification results.

[1749] Specific behavior:

[1750] 1. The server visualizes the results using matplotlib, Plotly, etc.

[1751] 2. The server visualizes the results and displays them on a dashboard.

[1752] 3. Export the results in CSV or Excel format as needed.

[1753] Data arithmetic: visualization and format transformation of results.

[1754] Output: Dashboard view and CSV / Excel format files.

[1755] (Application example 2)

[1756] 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."

[1757] Conventional data analysis systems do not take into account the user's emotional state, resulting in situations where users feel stressed or where data preprocessing and model selection are suboptimal. Furthermore, product recommendation systems on online shopping sites do not reflect user emotions and are limited to uniform recommendation accuracy. This prevents improvements to the user experience and increases in purchasing motivation.

[1758] 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.

[1759] In this invention, the server includes: [means for a user to upload a data file;] [means for the server to store the received data in an internal database;] [means for the server to automatically perform data consistency checks and preprocessing;] [means for the server to use an emotion engine that recognizes the user's emotional state;] [means for the server to adjust preprocessing and model selection based on the user's emotional state;] [means for the server to apply multiple machine learning algorithms to select the most appropriate model;] [means for the server to perform prediction or classification using the selected model;] [means for the server to visualize the results and display them on a dashboard or output them in file format;] [means for collecting user emotion data via a smartphone or other device; and [means for the server to recommend products.] This enables real-time recognition of the user's emotional state, corresponding data preprocessing and model selection, and even emotion-based product recommendations.

[1760] The "means for users to upload data files" refers to the means by which users add data files to be analyzed to the system.

[1761] The "means for storing data received by the server in an internal database" refers to the means for the server to store data files received from users in a database.

[1762] "Means for the server to automatically check data consistency and preprocess" refers to automated means for the server to check data consistency and handle missing values ​​and outliers.

[1763] The "means for the server to use an emotion engine to recognize the user's emotional state" refers to a means for utilizing an engine to detect and analyze the user's emotions and identify a specific emotional state.

[1764] The "means for the server to adjust preprocessing and model selection based on the emotional state of the user" refers to a means for changing the process of data preprocessing and appropriate model selection in accordance with the emotional state of the user detected using the emotion engine.

[1765] "Means in which the server applies multiple machine learning algorithms and selects the most appropriate model" refers to a means in which data is analyzed using multiple machine learning technologies and the optimal algorithm is selected from among them.

[1766] The "means for the server to make predictions or classifications using the selected model" refers to the means for making predictions or classifications of data using the optimal trained model.

[1767] The "means for the server to visualize the results and display them on a dashboard or output them in a file format" refers to a means for displaying the results of prediction or classification in a visually easy-to-understand manner on a dashboard or outputting them as a file.

[1768] "Means for collecting user emotional data via smartphones or other devices" refers to means for collecting user emotional information in real time using smartphones or other sensor devices.

[1769] The "means for the server to recommend products" refers to a means for recommending products suitable for a user based on collected data and the user's emotional state.

[1770] This invention is an emotion-adaptive product recommendation system that applies the PredictEase system, which incorporates an emotion engine that recognizes the user's emotional state, to an online shopping site. This system can be accessed by users via smartphones or other devices.

[1771] Data Entry and Preprocessing

[1772] Users access the online shopping site's application and upload the necessary data files. The server receives the data and stores it in an internal database. At that time, the server checks the data for consistency and automatically processes missing or outlier values.

[1773] Recognition of emotional states

[1774] The server collects user emotional data via smartphones and other devices (cameras, microphones, emotion-sensing devices, etc.). The emotion engine analyzes this data and identifies the user's emotional state. For example, it uses facial recognition and voice analysis technology to determine whether the user is feeling stressed.

[1775] Model selection and tuning

[1776] Based on the user's emotional state identified by the emotion engine, the server adaptively adjusts the data preprocessing and model selection processes, thereby shortening processing time and providing easy-to-understand feedback to users experiencing stress.

[1777] The server analyzes the data by applying multiple machine learning algorithms (e.g., linear regression, random forest, decision tree, etc.) to select the optimal model, and then performs hyperparameter tuning of the selected model to achieve optimal performance.

[1778] Predictions and product recommendations

[1779] The server uses the best-fit model trained on the system to make predictions or classifications based on the user's data. It also considers the user's emotional state and purchasing history to recommend suitable products. The results are visualized and displayed intuitively on the user's application screen. Users can also download the results as a file if desired.

[1780] Hardware and software used

[1781] Hardware: Smartphone (camera, microphone), emotion sensing device (e.g., EEG headset)

[1782] Software: Sentiment analysis software (e.g., Sentiment API), machine learning libraries (e.g., TensorFlow, Scikit-learn)

[1783] Examples and prompts

[1784] For example, if a user is feeling stressed, the "Emotion-Adaptive Smart Recommender" will recommend products using prompts like the following:

[1785] User Emotion: Stress

[1786] Past purchase history: Aroma candles, relaxation CDs, herbal tea

[1787] Recommended Products:

[1788] 1. Lavender aroma oil

[1789] 2. Yoga mat

[1790] 3. Relaxation Cafe Gift Card

[1791] This system can recommend optimal products by taking into account the user's emotional state in real time, which is expected to improve the user experience and satisfaction.

[1792] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1793] Step 1:

[1794] The user accesses the application of the online shopping site and uploads the data file to be analyzed (e.g., a CSV file).

[1795] Input: Data file uploaded by the user.

[1796] Output: The data file that is transferred to the server.

[1797] Specific operation: The user uses a smartphone or personal computer to select a data file from the file selection screen and click the upload button.

[1798] Step 2:

[1799] The server stores the received data files in an internal database, performs data consistency checks, and automatically handles missing and outlier values.

[1800] Input: Data file received from user.

[1801] Output: Preprocessed data that has passed integrity checks.

[1802] What it does: The server stores the data in a database and runs scripts to impute missing values ​​(e.g., impute with the mean) and remove / correct outliers.

[1803] Step 3:

[1804] The server collects user emotional data via the smartphone's camera, microphone, and emotion sensing device.

[1805] Input: User facial recognition data, voice data, and input from other sensors.

[1806] Output: Data on the user's emotional state.

[1807] Specific operation: The server performs facial expression analysis from camera images using an emotion recognition API, and performs emotion analysis from microphone audio.

[1808] Step 4:

[1809] The server uses an emotion engine to analyze the collected emotion data and identify the user's emotional state.

[1810] Input: Collected emotion data.

[1811] Output: The user's specific emotional state (e.g., stress, joy, excitement).

[1812] Specific operation: The server runs the emotion engine, analyzes the data, and outputs the emotional state as a number or category.

[1813] Step 5:

[1814] The server adaptively adjusts the data preprocessing and model selection processes based on the user's emotional state.

[1815] Input: The user's emotional state.

[1816] Output: Adjusted preprocessing parameters and model selection strategy.

[1817] What it does: The server considers the emotion data and performs simplified pre-processing to reduce processing time or generates easy-to-understand feedback.

[1818] Step 6:

[1819] The server applies multiple machine learning algorithms and selects the most suitable model.

[1820] Input: Preprocessed data.

[1821] Output: Best model via cross-validation.

[1822] What it does: The server applies different algorithms (e.g., linear regression, random forest) and selects the best model based on the results of cross-validation.

[1823] Step 7:

[1824] The server tunes the hyperparameters of the selected model to achieve optimal performance.

[1825] Input: A cross-validated model.

[1826] Output: The optimal tuned model.

[1827] Specific operation: Set the range of hyperparameters and search for the optimal parameters using grid search or random search.

[1828] Step 8:

[1829] The server uses the best trained model to make predictions and product recommendations.

[1830] Input: User data and emotional state.

[1831] Output: A list of product recommendations.

[1832] What it does: Apply the model to rank and recommend products based on user sentiment and purchase history.

[1833] Step 9:

[1834] The server visualizes the recommendation results and displays them on the user's application screen.

[1835] Input: A list of recommended products.

[1836] Output: Visualized dashboard or file format.

[1837] What it does: Generates results in the form of graphs and lists, displays them in a dashboard that is intuitive and easy for users to understand, and provides the results as a downloadable file if required.

[1838] 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.

[1839] 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.

[1840] 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.

[1841] 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.

[1842] 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.

[1843] 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.

[1844] 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).

[1845] 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.

[1846] 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."

[1847] 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.

[1848] 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).

[1849] 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.

[1850] 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.

[1851] 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.

[1852] 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.

[1853] 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.

[1854] 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.

[1855] 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.

[1856] 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.

[1857] 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.

[1858] 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.

[1859] The following is further disclosed regarding the above embodiment.

[1860] (Claim 1)

[1861] [A means for users to upload data files;

[1862] [Means for the server to store the received data in an internal database;

[1863] [Means for the server to automatically perform data integrity checks and preprocessing;

[1864] [Means for the server to apply multiple machine learning algorithms and select the most suitable model;

[1865] [Means for the server to make predictions or classifications using the selected model;

[1866] [Means for the server to visualize the results and display them on a dashboard or output them in a file format;

[1867] A system including:

[1868] (Claim 2)

[1869] [The system according to claim 1, wherein the server performs hyperparameter tuning to construct an optimal model.

[1870] (Claim 3)

[1871] The system of claim 1, wherein the server encodes categorical variables in the data and normalizes or standardizes numerical variables.

[1872] "Example 1"

[1873] (Claim 1)

[1874] [Means for users to upload data files;

[1875] [means for storing the received data in an internal storage device of the computer;

[1876] [Means for the computer to automatically check the integrity of data and perform preprocessing;

[1877] [Means for a computer to apply multiple machine learning algorithms and select the most suitable model;

[1878] [Means for a computer to make a prediction or classification using the selected model;

[1879] [Means for the computer to visualize the results and display them on a display device or output them in a file format;

[1880] [Means for users to view and download results;

[1881] A system including:

[1882] (Claim 2)

[1883] [The system of claim 1, wherein the computer performs hyperparameter tuning to construct an optimal model.

[1884] (Claim 3)

[1885] The system of claim 1, wherein the computer encodes categorical variables in the data and normalizes or standardizes numerical variables.

[1886] "Application Example 1"

[1887] (Claim 1)

[1888] [A means for users to upload data files;

[1889] [Means for the server to store the received data in an internal database;

[1890] [Means for the server to automatically perform data integrity checks and preprocessing;

[1891] [Means for the server to apply multiple machine learning algorithms and select the most suitable model;

[1892] [Means for the server to make predictions or classifications using the selected model;

[1893] [Means for the server to visualize the results and display them on a dashboard or output them in a file format;

[1894] [Means of analyzing the operation data of autonomous vehicles and detecting and predicting abnormalities via an application installed on a smartphone,

[1895] [Means to store the trained model on the smartphone,

[1896] A system including:

[1897] (Claim 2)

[1898] [The system according to claim 1, wherein the server performs hyperparameter tuning to construct an optimal model.

[1899] (Claim 3)

[1900] The system of claim 1, wherein the server encodes categorical variables in the data and normalizes or standardizes numerical variables.

[1901] "Example 2: Combining Emotion Engines"

[1902] (Claim 1)

[1903] [A means for users to upload data files to be analyzed;

[1904] [Means for the server to store the uploaded data in an internal database;

[1905] [Means for the server to automatically perform data integrity checks and preprocessing;

[1906] [Means for the server to recognize user behavior on the interface using an emotion engine and adjust preprocessing and model selection;

[1907] [Means for the server to apply multiple machine learning algorithms and select the most suitable model;

[1908] [Means for the server to make predictions or classifications using the selected model;

[1909] [Means for the server to visualize the results of prediction or classification and display them on a dashboard or output them in a file format;

[1910] A system including:

[1911] (Claim 2)

[1912] [The system according to claim 1, wherein the server performs hyperparameter tuning to construct an optimal model.

[1913] (Claim 3)

[1914] The system of claim 1, wherein the server encodes categorical variables in the data and normalizes or standardizes numerical variables.

[1915] "Application example 2 when combining emotion engines"

[1916] (Claim 1)

[1917] [A means for users to upload data files;

[1918] [Means for the server to store the received data in an internal database;

[1919] [Means for the server to automatically perform data integrity checks and preprocessing;

[1920] [means for the server to use an emotion engine to recognize the user's emotional state;

[1921] [Means for the server to adjust preprocessing and model selection based on the user's emotional state; and

[1922] [Means for the server to apply multiple machine learning algorithms and select the most suitable model;

[1923] [Means for the server to make predictions or classifications using the selected model;

[1924] [Means for the server to visualize the results and display them on a dashboard or output them in a file format;

[1925] [Means of collecting user emotional data via smartphones and other devices;

[1926] [Means for the server to recommend products;

[1927] A system including:

[1928] (Claim 2)

[1929] [The system according to claim 1, wherein the server performs hyperparameter tuning to construct an optimal model.

[1930] (Claim 3)

[1931] The system of claim 1, wherein the server encodes categorical variables in the data and normalizes or standardizes numerical variables. [Explanation of symbols]

[1932] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for users to upload data files; a means for the server to store the received data in an internal database; a means for the server to automatically perform data integrity checking and preprocessing; A means for the server to apply multiple machine learning algorithms and select the most suitable model; a means for the server to perform prediction or classification using the selected model; A means for the server to visualize the results and display them on a dashboard or output them in a file format; A system including:

2. The system according to claim 1, wherein the server performs hyperparameter tuning to construct an optimal model.

3. The system of claim 1 , wherein the server encodes categorical variables in the data and normalizes or standardizes numerical variables.

Images