system

Abstract

Provide a system. A means for collecting past examination data; A means for inputting information of new subscribers and collecting the data; A means of analyzing the collected data and identifying bottlenecks in the review process; A means to set up a new review flow and simulate it, A means of predicting performance one and two years into a set review flow; a means for generating reports based on the simulation and prediction results; A means of reviewing the generated report; 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] The current screening process for new subscribers faces problems such as prolonged screening times, delays due to incomplete documentation, and reduced efficiency at each screening step. In particular, if the screening flow is not sufficiently improved, the lengthy screening process can lead to reduced user satisfaction and increased operational costs. Furthermore, there are insufficient means of predicting future effects, making it difficult to evaluate the effectiveness of improvement measures in advance. To solve these issues, it is necessary to identify bottlenecks in the screening process using past data and to simulate and predict new flows. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means.

[0006] The system includes a means for collecting past screening data, which retrieves and organizes screening data from the past five years from a database. It also includes a means for inputting information about new applicants and collecting that data, and sends the subscriber information to a server in real time. It also includes a means for analyzing the collected data and identifying bottlenecks in the screening process, analyzing the average screening time and the frequency of document deficiencies, etc. It also includes a means for setting up a new screening flow and simulating it, which allows for advance evaluation of the efficiency and effectiveness of the set screening flow. It also includes a means for predicting performance one and two years from now based on the set screening flow, using a machine learning algorithm to predict future performance. It also includes a means for generating reports based on the simulation and prediction results, which are formatted and sent. It also includes a means for reviewing the generated reports, allowing users to review the results and provide feedback. This can shorten the screening time, improve customer satisfaction, and reduce costs.

[0007] "Past review data" refers to all past records and information of the review process, including application documents, review results, review periods, reasons for rejection, etc.

[0008] "Information of new subscribers" refers to all information provided by individuals or organizations who wish to join, including name, address, desired plan, required documents, etc.

[0009] "Means for collecting data" refers to methods, devices and systems for obtaining, organizing and storing the necessary data from various sources.

[0010] "Analyzing data" refers to the process of extracting useful information from collected data using statistical analysis, machine learning, and other analytical methods.

[0011] "Means for identifying bottlenecks in the review process" refers to methods and functions for detecting and identifying factors and obstacles that reduce efficiency in the review process.

[0012] The "new review flow" refers to new processes and procedures that have been revised and improved upon from previous review procedures.

[0013] "Means for simulation" refers to methods, devices, and systems for virtually reproducing and executing real processes and predicting and evaluating their results.

[0014] "Means for predicting performance" means methods, devices, or systems for calculating or predicting future efficiency, effectiveness, or other performance indicators.

[0015] "Means for generating reports" refers to methods and functions for compiling analysis and simulation results into documents and formatting them in a specific format.

[0016] "Means for reviewing reports" refers to the interface and functionality that allows users to view generated reports and provide evaluations and feedback. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7]FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[0039] System Overview

[0040] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[0041] Program processing

[0042] 1. Data Collection

[0043] The server connects to the database and collects the past five years of review data, including application documents, review results, review periods, and reasons for rejection.

[0044] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[0045] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[0046] 2. Data Analysis

[0047] The server analyzes the collected past review data, performs statistical analysis, and calculates the average and standard deviation of the review period, the rate of defects in each document, and the frequency of rejection reasons.

[0048] The server provides the functionality to identify and visualize bottlenecks in the review process through data analysis.

[0049] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[0050] 3. Improvement Simulation

[0051] The terminal provides an interface that allows users to set up new screening flows (for example, introducing pre-document checks).

[0052] The server performs a comparative simulation of the old and new review flows and evaluates their efficiency and effectiveness.

[0053] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[0054] 4. Predictive model generation

[0055] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance of the new review process one and two years from now.

[0056] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[0057] 5. Generate and review reports

[0058] The server generates a report based on the simulation and prediction results, formats it in a specific format, and sends it to the terminal.

[0059] The terminal displays the generated report to the user for the user to review.

[0060] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[0061] In this way, the system of the present invention can effectively optimize the new customer screening process and predict future results, which can be expected to shorten screening times, improve customer satisfaction, and reduce operational costs.

[0062] The processing flow will be explained below.

[0063] Step 1:

[0064] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[0065] Step 2:

[0066] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[0067] Step 3:

[0068] The server performs statistical analysis based on the past review data collected, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects. Next, the analysis results are used to identify bottlenecks in the review process, and the visualized analysis results are sent to the terminal.

[0069] Step 4:

[0070] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[0071] Step 5:

[0072] The server runs a simulation based on the new audit flow that has been set up. It compares the old audit flow with the new one and calculates the efficiency and effectiveness of each (for example, the rate of reduction in audit time and cost reduction). The simulation results are visualized and sent to the terminal.

[0073] Step 6:

[0074] The server uses a machine learning algorithm to build a predictive model based on the simulation results and past data, and uses the built model to predict the performance (e.g., review efficiency, customer satisfaction) of the new screening process one and two years from now.

[0075] Step 7:

[0076] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[0077] Step 8:

[0078] The user can view the generated report via the terminal, which includes the simulation results and the output of the predictive model, providing the user with information to make decisions about future improvements to the review process.

[0079] Step 9:

[0080] Users enter feedback, which is sent from their devices to a server that records it and uses it to improve future simulations and predictive models.

[0081] Example 1

[0082] 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."

[0083] The traditional new customer application screening process involves a large number of documents and information, which can lead to processing delays and bottlenecks. It is also difficult to simulate and predict the effects of an improved screening process in advance. This makes it challenging to improve the efficiency of screening and customer satisfaction.

[0084] 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.

[0085] In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting that data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for providing an interface for a terminal to set a new screening flow, and means for the server to use a machine learning algorithm when analyzing the data. This improves the efficiency of the screening process and enables the design and implementation of a predictable screening flow.

[0086] "Means for collecting past audit data" refers to devices or programs that can collect information related to audits conducted in the past and store it on a server or database.

[0087] The "means for inputting information about new subscribers and collecting that data" refers to an interface and device for inputting personal information and application forms of users who wish to join, and sending them to the server for storage.

[0088] "Means for analyzing collected data and identifying bottlenecks in the review process" refers to analytical tools and algorithms used to analyze collected review data and identify causes of delays and inefficiencies during the review process.

[0089] The "means for setting up a new review flow and simulating it" is a system that provides an interface that allows users to set up a new review flow, and simulates that flow to evaluate its effectiveness.

[0090] "Means for predicting one-year and two-year performance of a configured review flow" refers to a system that runs machine learning algorithms and predictive models to predict future performance of a newly configured review flow.

[0091] The "means for generating a report based on the results of simulation and prediction" refers to a device or program that automatically generates a report based on the results of simulation and prediction, and puts it in a format that can be checked by the user.

[0092] The "means for checking the generated report" refers to an interface and device that displays the generated report to the user and enables the user to check the contents.

[0093] The "means for providing an interface for a terminal to set up a new screening flow" refers to a device and a program for providing an interface for a terminal to allow a user to input settings for a new screening flow.

[0094] "Means for using machine learning algorithms when the server analyzes data" refers to a system for using machine learning algorithms when the server analyzes collected data and for identifying bottlenecks in the review process.

[0095] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[0096] System Overview

[0097] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[0098] Data collection and input

[0099] The server connects to the database and collects screening data from the past five years. This data includes application documents, screening results, screening periods, and reasons for rejection. The terminal provides a function for entering information about applicants through a user interface. The user enters the applicant's information into the terminal, and this information is sent to the server in real time.

[0100] Specific examples

[0101] The user enters Mr. Tanaka's information (name, address, application documents, etc.) into the terminal. This information is sent to the server and stored in the database.

[0102] Data analysis and bottleneck identification

[0103] The server analyzes the collected past review data. Statistical analysis is performed to calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. Through data analysis, the server identifies bottlenecks in the review process and visualizes them.

[0104] Specific examples

[0105] The server analyzes the review data using statistical analysis tools (e.g., Python's Pandas and NumPy). Using the R ggplot2 library, it identifies the main causes of review delays and generates a visualization graph showing that these are incomplete documents.

[0106] Improvement Simulation

[0107] The terminal provides an interface for setting up a new screening flow (for example, introducing a pre-document check). The user sets up the new screening flow, and the server performs a comparative simulation of the old and new screening flows to evaluate their efficiency and effectiveness.

[0108] Specific examples

[0109] The user configures the pre-document check using an HTML form on the terminal. The server runs a Monte Carlo simulation to compare the old and new flows. The new flow shows that the review period is reduced by 20%.

[0110] Predictive Model Generation

[0111] The server builds a predictive model based on the simulation results and past data, using machine learning algorithms such as SciKit-Learn and TENSORFLOW® to predict the performance of the new review process one and two years in the future.

[0112] Specific examples

[0113] The server builds a predictive model and finds that one year later, review efficiency will improve by 15% and customer satisfaction will increase by 10 points.

[0114] Report generation and review

[0115] The server generates a report based on the simulation and prediction results, formats it in a specific format (e.g., PDF, Excel), and sends it to the terminal. The terminal displays the generated report to the user, who can then check the contents.

[0116] Specific examples

[0117] The server generates a PDF report using the ReportLab library and sends it to the terminal, which displays the report to the user using a PDF viewer.The user can then decide to officially implement the new review flow based on the contents of the report.

[0118] Prompt Sentence Examples

[0119] Here is an example of a prompt to use for a generative AI model:

[0120] 1. "This system provides the ability to analyze historical data and identify bottlenecks in the review process. Please explain how data collection and analysis are performed. Include specific tools and methods."

[0121] 2. Please explain in detail the process of comparing the simulation results of the new review flow with the conventional flow. Please also provide specific examples of improvements that result from this.

[0122] 3. "Please describe how the server generates a predictive model and uses it to assess the efficiency of future review processes and customer satisfaction. Please also describe the algorithms and tools used."

[0123] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0124] Step 1:

[0125] The server accesses the database and collects past review data. The server executes SQL queries to retrieve data such as application documents, review results, review periods, and rejection reasons from the past five years. The data is saved in CSV or JSON format and temporarily stored for later analysis. The input to this data collection process is the database connection information to the server, and the output is past review data.

[0126] Specific behavior:

[0127] The server uses an SQL query to extract past review data from the database and save it as a file. Example: SELECT FROM review_data WHERE year >= CURRENT_DATE - INTERVAL '5 years';

[0128] Step 2:

[0129] The terminal provides a user interface, and the user enters information about the new member (e.g., name, address, application form). The terminal converts this information into JSON format and sends it to the server. The input is the information about the new member entered by the user, and the output is JSON format data sent to the server.

[0130] Specific behavior:

[0131] When a user enters the information of the prospective subscriber into the form and presses the submit button, the terminal converts the information into JSON format and sends it to the server as an HTTP POST request.

[0132] Step 3:

[0133] The server saves the collected data of new applicants to a database. The server validates the data format and inserts the data into the appropriate fields. The input is the JSON data of new applicants sent from the terminal, and the output is the information of the applicants saved in the database.

[0134] Specific behavior:

[0135] The server parses the received JSON data and inserts the data into the corresponding table. Example: INSERT INTO applicant_data (name, address, documents) VALUES (json_data["name"], json_data["address"], json_data["documents"]);

[0136] Step 4:

[0137] The server analyzes past review data. It uses statistical analysis tools (e.g., Python's Pandas and NumPy) to analyze the data and calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. The input is past review data, and the output is the analysis results.

[0138] Specific behavior:

[0139] The server runs a Python script, creates a data frame using the Pandas library, and performs statistical analysis. For example: import pandas as pd; data = pd.read_csv('review_data.csv'); mean_duration = data['duration'].mean();

[0140] Step 5:

[0141] The server identifies and visualizes bottlenecks in the review process based on the analysis results. It generates graphs using the R ggplot2 library to identify specific bottlenecks. The input is the analysis results, and the output is the visualized graph and the identified bottlenecks.

[0142] Specific behavior:

[0143] The server runs the R script and generates the graph using the ggplot2 library. For example: library(ggplot2); ggplot(data, aes(x=duration, y=freq)) + geom_bar(stat='identity');

[0144] Step 6:

[0145] The terminal provides an interface for setting up a new screening flow (e.g., introducing a pre-document check). The user sets up a new screening flow, which is then sent to the server. The input is the information on the new screening flow set up by the user, and the output is the flow setting data sent to the server.

[0146] Specific behavior:

[0147] When a user enters a new review flow into the form and presses the submit button, the terminal sends the content to the server.

[0148] Step 7:

[0149] The server runs a simulation comparing the old and new screening flows to evaluate their efficiency and effectiveness. The evaluation is performed using Monte Carlo simulation, etc. The input is information about the old and new screening flows, and the output is the simulation results for each flow.

[0150] Specific behavior:

[0151] The server runs the simulation program and evaluates the results. Example: monte_carlo_simulation(old_flow, new_flow);

[0152] Step 8:

[0153] The server builds a predictive model based on the simulation results and past data. It generates the predictive model using machine learning algorithms such as SciKit-Learn and TensorFlow. The input is the simulation results and past review data, and the output is the generated predictive model.

[0154] Specific behavior:

[0155] The server builds a predictive model using machine learning libraries. Example: from sklearn.linear_model import LinearRegression; model = LinearRegression().fit(X, y);

[0156] Step 9:

[0157] The server generates reports based on the simulation and prediction results and formats them in a specific format (e.g. PDF, Excel). The report is created using libraries such as the ReportLab. The input is the prediction model and the simulation results, and the output is the generated report.

[0158] Specific behavior:

[0159] The server generates a PDF report using ReportLab. Example: from reportlab.lib.pagesizes import letter; from reportlab.pdfgen import canvas; c = canvas.Canvas("output.pdf", pagesize=letter); c.drAWS®string(100, 750, "Simulation and Prediction Report"); c.save();

[0160] Step 10:

[0161] The terminal displays the generated report to the user, allowing the user to confirm the contents. The input is the report sent from the server, and the output is the report confirmed by the user.

[0162] Specific behavior:

[0163] The device opens a PDF viewer and displays the report to the user. Example: The device displays the report using the PDF viewer in the browser.

[0164] (Application example 1)

[0165] 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."

[0166] In the conventional screening process, it is difficult to efficiently identify bottlenecks and improve the screening flow using past data.In addition, there is a lack of means to optimize the screening process for new applicants using mobile devices such as smartphones, so it is difficult to sufficiently shorten the screening time or reduce false positives.

[0167] 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.

[0168] In this invention, the server includes a means for collecting past screening data, a means for inputting information on new applicants and collecting that data, and a means for analyzing the collected data and identifying bottlenecks in the screening process. This enables efficient optimization and improvement of the screening process. Furthermore, by including a means for providing an application that is installed on a smartphone and optimizes the screening process for new applicants, screening time can be shortened and customer satisfaction can be improved.

[0169] "Past review data" refers to information related to the review process, such as past application documents, review results, review periods, and reasons for rejection.

[0170] "Information of new applicants" refers to personal information such as the name, address, and application documents of applicants.

[0171] A "bottleneck" refers to a factor or obstruction that delays progress in the review process.

[0172] "Review flow" refers to the series of steps and processes for reviewing an applicant's information and arriving at a result.

[0173] "Simulation" refers to a virtual experiment or trial conducted based on set conditions and flow.

[0174] A "smartphone" refers to a mobile device that has multifunctional data communication capabilities in addition to mobile phone functions.

[0175] An "application" refers to a software program that is installed on a device such as a smartphone and provides specific functions.

[0176] As an embodiment of the present invention, the following system is embodied.

[0177] 1. System Overview

[0178] This system collects past screening data and inputs and collects information on new applicants to efficiently identify bottlenecks in the screening process and simulate and predict new screening flows. In particular, it aims to optimize the screening process through an application installed on smartphones.

[0179] 2. Hardware and Software Used

[0180] Hardware: smartphones, tablets, servers

[0181] software:

[0182] Python and pandas are used for data collection and analysis.

[0183] For simulation and prediction, we use the random forest algorithm from scikit-learn.

[0184] Data visualization will be done using matplotlib and seaborn.

[0185] Python is used to generate reports and output them in PDF format, etc.

[0186] 3. Data processing and calculation flow

[0187] The server retrieves past screening data from the database and analyzes it to identify bottlenecks in the screening process. For example, it can identify that a specific document deficiency is the main cause of screening delays. The device (smartphone or tablet) provides an interface for entering information about new applicants and has the function of sending the collected information to the server.

[0188] The server has the function of setting up a new review flow (e.g., introducing a pre-document check) and runs a simulation to evaluate its effectiveness. This can result in results such as a shortened review period and improved customer satisfaction. The server generates a report based on the simulation results and prediction model and sends it to the terminal in a specific format such as PDF. The user can check the generated report through the terminal and evaluate its effectiveness.

[0189] 4. Specific examples and prompts for generative AI models

[0190] As a concrete example, the following simulation results can be obtained. For example, if a new review flow is introduced, it is predicted that the review period will be shortened by 20% and review efficiency will improve by 15% after one year. These results are provided to the user as a PDF report.

[0191] An example of a prompt to be input to the generative AI model is as follows:

[0192] Please provide a detailed explanation of the correlation between the rate of document deficiencies and the review period based on the review data from the past five years.

[0193] "Please predict the performance improvement if we implement a new review flow."

[0194] In this way, the system of the present invention effectively optimizes the screening process for new subscribers, shortening the screening period and improving customer satisfaction.

[0195] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0196] Step 1:

[0197] The server collects past review data from the database. Specifically, it connects to the review database and extracts information such as application documents, review results, review periods, and rejection reasons over the past five years. The input to this data collection process is the database connection information, and the output is the collected data in a data frame format.

[0198] Step 2:

[0199] The terminal provides an interface for entering information about prospective new subscribers. The user uses the terminal interface to enter their name, address, and application documents. The entered information is sent in real time to the server, where it is stored in a database. The input to this process is the information entered by the user, and the output is the new subscriber's information stored in the server database.

[0200] Step 3:

[0201] The server analyzes the collected past data. As a specific example of the analysis, the server performs statistical analysis to calculate the average and standard deviation of the review period, the defect rate of each document, and the frequency of rejection reasons. The input of this step is the collected past review data, and the output is a report containing the analysis results.

[0202] Step 4:

[0203] Based on the results of the data analysis, the server identifies bottlenecks in the review process and visualizes them. For example, the server may identify that incomplete documents are the main cause of review delays and generate the results in a visualized format such as a graph. The input to this process is the analysis results, and the output is visualized data on the bottlenecks.

[0204] Step 5:

[0205] The terminal provides an interface for setting up a new screening flow. The user sets up a new flow, such as introducing a pre-document check, through this interface. The set information is sent to the server. The input of this step is the information for the new screening flow set up by the user, and the output is the data for the new flow setting sent to the server.

[0206] Step 6:

[0207] The server simulates the old and new review flows and evaluates their efficiency and effectiveness. Specifically, the server uses the collected data and new flow information to run machine learning algorithms such as RandomForestClassifier to simulate the new flow. The inputs to this step are past data and new flow information, and the output is performance evaluation data as the simulation results.

[0208] Step 7:

[0209] The server builds a predictive model based on the simulation results and past data. For example, it uses a machine learning algorithm to predict how a new screening process will perform one and two years from now. The inputs to this process are the simulation results and past data, and the output is the results of the predictive model.

[0210] Step 8:

[0211] The server generates a report based on the prediction results and sends it to the terminal. The report is formatted in a specific format (e.g., PDF) and includes visualization data. The user can view the generated report through the terminal and evaluate the effectiveness. The input of this step is the results from the predictive model, and the output is the generated report.

[0212] The above steps realize a concrete system for optimizing the screening process for new subscribers.

[0213] 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.

[0214] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[0215] System Overview

[0216] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows. Furthermore, by combining it with an emotion engine that recognizes user emotions, the system can further improve the system based on emotion analysis of user feedback.

[0217] Program processing

[0218] 1. Data Collection

[0219] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[0220] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[0221] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[0222] 2. Data Analysis

[0223] The server performs statistical analysis based on the collected past review data, calculating the average and standard deviation of review times, the frequency of each rejection reason, and the rate of application document defects.

[0224] The server identifies bottlenecks in the review process through data analysis and transmits them to the terminal in a visualized form.

[0225] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[0226] 3. Improvement Simulation

[0227] The terminal provides an interface that allows users to set up a new screening flow (for example, introducing a pre-document check process). When the user inputs new settings, the terminal sends the setting information to the server.

[0228] The server runs a simulation based on the new screening flow. It compares the old and new screening flows and evaluates their efficiency and effectiveness (reduced screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[0229] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[0230] 4. Predictive model generation

[0231] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[0232] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[0233] 5. Generate and review reports

[0234] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[0235] The terminal displays the generated report to the user and provides an interface for the user to review the report and enter feedback.

[0236] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[0237] 6. Emotion Recognition by Emotion Engine

[0238] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[0239] Example: While a user is viewing a report, the emotion engine recognizes the user's emotions (e.g., satisfaction, dissatisfaction, excitement, disappointment) via a webcam or microphone and transmits them to the server in real time.

[0240] 7. Emotional Data Analysis and Feedback

[0241] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[0242] Example: The server analyzes sentiment data and user feedback to identify, for example, which specific process steps are causing user frustration. Based on this, further flow improvements are suggested.

[0243] In this way, the system of the present invention not only effectively optimizes the new customer screening process, but also takes into account the user's emotional state to suggest further improvements, thereby further shortening the screening period, improving customer satisfaction, and reducing operational costs.

[0244] The processing flow will be explained below.

[0245] Step 1:

[0246] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[0247] Step 2:

[0248] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[0249] Step 3:

[0250] Statistical analysis is performed based on past review data collected by the server, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects.

[0251] Step 4:

[0252] The server uses the analysis results to identify bottlenecks in the review process and transmits the visualized analysis results to the terminal, using graphs and heat maps.

[0253] Step 5:

[0254] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[0255] Step 6:

[0256] The server runs a simulation based on the new screening flow that has been set up. The old screening flow is compared with the new one, and the efficiency and effectiveness of each (review period reduction rate, cost reduction effect) are evaluated. The simulation results are visualized in graphs and tables and sent to the terminal.

[0257] Step 7:

[0258] The server builds a predictive model based on the simulation results and past data, and uses machine learning algorithms (such as linear regression or random forest) to predict performance (screening efficiency, customer satisfaction) one and two years into the future.

[0259] Step 8:

[0260] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[0261] Step 9:

[0262] The user checks the generated report through the terminal, evaluates the proposed improvements to the review process based on the report contents, and makes a final decision.

[0263] Step 10:

[0264] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[0265] Step 11:

[0266] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[0267] Example 2

[0268] 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."

[0269] In the conventional review process, bottlenecks were identified by analyzing past data and the effectiveness of new flows was simulated, but there were limitations to improving the efficiency of the overall review process or customer satisfaction. Furthermore, it was not possible to reflect user sentiment in real time, and the quality of feedback was insufficient. As a result, the reduction of review time and operational costs was not fully achieved.

[0270] 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.

[0271] A means of collecting past audit data;

[0272] A means for inputting information of new subscribers and collecting the data;

[0273] A means of analyzing the collected data and identifying bottlenecks in the review process;

[0274] A means to set up a new review flow and simulate it,

[0275] A means of predicting performance one and two years into a set review flow;

[0276] a means for generating reports based on the simulation and prediction results;

[0277] A means of reviewing the generated report;

[0278] A means for recognizing user emotions in real time and collecting the results;

[0279] A means to analyze the collected sentiment data and suggest further improvements to the review process; and

[0280] This will make the screening process more efficient, improve customer satisfaction, and reduce operational costs.

[0281] "Review Data" refers to all data related to the review process, including past application documents, review results, review periods, and reasons for rejection.

[0282] "Information of prospective subscribers" refers to all information provided by users who wish to sign up, such as the name, address, and application documents of users who wish to sign up.

[0283] A "bottleneck" refers to an item or step in the review process that tends to slow down processing and reduce efficiency.

[0284] "Review flow" refers to the order and method of each step and procedure in the review process.

[0285] "Simulation" refers to a computer-based analytical method used to virtually implement a new review process and predict its effects.

[0286] "Performance prediction" refers to using machine learning algorithms to predict review efficiency and customer satisfaction one and two years after the introduction of a new review process.

[0287] "Report" refers to a written report summarizing the simulation and prediction results.

[0288] "Emotion recognition" refers to a technology that analyzes a user's facial expressions and voice to identify the user's emotional state in real time.

[0289] "Emotion Data" refers to data regarding a user's emotional state collected through emotion recognition technology.

[0290] "Recommendations for improving the review process" refers to specific changes and improvements proposed to make the review process more efficient, based on analysis of emotional data and user feedback.

[0291] To implement this invention, a system is used that combines a terminal for collecting past screening data and inputting information on new applicants, a server for analyzing the data and setting up and simulating new screening flows, and an emotion engine that recognizes user emotions in real time.

[0292] System Configuration

[0293] This system consists of multiple pieces of hardware and software, as follows:

[0294] 1. Server: Performs large-volume data processing, analysis, and simulation.

[0295] 2. Client Device: Provides an interface for users to enter information about new subscribers and confirm the data.

[0296] 3. Emotion Recognition Engine: A module that analyzes the user's facial expressions and voice to recognize emotions.

[0297] Hardware and Software

[0298] The servers are equipped with high-performance CPUs and GPUs, and a comprehensive database management system (e.g., MySQL (registered trademark), PostgreSQL). Data analysis uses Python and related packages (e.g., pandas, numpy, scikit-learn).

[0299] The terminals used are PCs and tablets, which allow users to access the system through a web browser. Web applications are built using HTML, CSS, and JavaScript (registered trademark) (e.g., React.js).

[0300] The emotion engine uses input devices such as cameras and microphones, as well as software (e.g., OpenCV, TensorFlow) that controls and analyzes them.

[0301] Program processing flow

[0302] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). This data is then converted into a format that can be used for analysis after missing values ​​are imputed and data normalization is performed.

[0303] The terminal provides a function to input information of new subscribers through a user interface. When a user inputs information of a subscriber, the information is sent to the server in real time and stored in a database.

[0304] The server performs statistical analysis of the collected audit data to identify bottlenecks. The identified bottlenecks are visualized and sent to the terminal. The server then simulates the new audit flow and compares it with the conventional flow to evaluate its efficiency and effectiveness.

[0305] A predictive model is built based on the simulation results and past data, and performance one and two years after the new flow is implemented is predicted. Finally, a report is generated based on the simulation and prediction results and sent to the terminal in PDF format or other format.

[0306] While users are reviewing the report, the emotion engine recognizes their emotions in real time and sends the data to the server, which analyzes the emotion data and provides user feedback to suggest further improvements to the review process.

[0307] Specific examples

[0308] For example, when a user enters information about a prospective new member into a terminal (such as name, address, and application documents), this information is sent to the server and stored in a database. The server analyzes this data and, based on past screening data, identifies incomplete documents as a major bottleneck. The user sets up a new screening flow from the terminal (for example, introducing a pre-document check process), and the server simulates it, showing that the screening period can be reduced by 20%.

[0309] Example prompts from a generative AI model:

[0310] "Please enter the information of the person wishing to sign up. We will propose the optimal screening process based on the analysis of screening data from the past five years."

[0311] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0312] Step 1: Data collection

[0313] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is then filled in with missing values, normalized, and converted into a format that can be used for analysis. Specifically, SQL queries are executed to extract data from the database and preprocessed using Python.

[0314] Input: Database connection information and SQL query

[0315] Output: Preprocessed audit dataset

[0316] Step 2: Enter information about new members

[0317] The terminal provides a function that allows users to input information about prospective subscribers through a user interface. When a user inputs information such as the prospective subscriber's name, address, and application documents into the terminal, this information is sent to the server in real time. The terminal is built using a web application (e.g., React.js), and the input information is sent to the server via an AJAX request.

[0318] Input: Subscriber information entered by the user into the terminal

[0319] Output: Subscriber information sent to the server

[0320] Step 3: Data analysis

[0321] Statistical analysis is performed based on past review data collected by the server. Specifically, data frames are manipulated using Python's pandas and numpy to calculate the average and standard deviation of review periods, the frequency of each reason for rejection, and the rate of incomplete application documents.

[0322] Input: Preprocessed audit dataset

[0323] Output: Calculated statistics (average review period, frequency of rejection reasons, etc.)

[0324] Step 4: Identify bottlenecks

[0325] The server analyzes the data to identify bottlenecks in the review process. Based on this statistical data, the server identifies specific items (e.g., incomplete documents) or specific process steps (e.g., review delays). The results are visualized in graphs and charts and sent to the terminal.

[0326] Input: Calculated statistic value

[0327] Output: Visualized data of bottleneck identification results

[0328] Step 5: Setting up and simulating a new review flow

[0329] The terminal provides the user with an interface that allows them to set up a new screening flow. The user sets up a new flow (for example, the introduction of a pre-document check process), and the information is sent to the server. Based on this information, the server simulates the old and new screening flows and evaluates the efficiency and effectiveness of each flow (reduction in screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[0330] Input: Information about the new screening flow configured by the user

[0331] Output: Visualized data of the simulation results of the old and new screening flows

[0332] Step 6: Generate a predictive model

[0333] The server generates a predictive model based on the simulation results and past data. Specifically, it uses a Python machine learning package (e.g., scikit-learn) to build a model that predicts the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[0334] Input: Simulation results and past audit data

[0335] Output: Prediction model and results (prediction performance after 1 year and 2 years)

[0336] Step 7: Generate and review the report

[0337] The server automatically generates a detailed report based on the simulation and prediction results. The report is formatted in PDF format or similar and sent to the terminal. The terminal displays the generated report to the user and provides an interface where the user can review the report and enter feedback.

[0338] Input: Prediction model and its results

[0339] Output: Report in PDF format

[0340] Step 8: Emotion Recognition with the Emotion Engine

[0341] The emotion engine recognizes the user's emotions in real time as they check the report. Specifically, it analyzes data collected through the camera and microphone using OpenCV and TensorFlow to generate emotion data, which is then sent to the server.

[0342] Input: Real-time data collected from cameras and microphones

[0343] Output: Emotion data

[0344] Step 9: Emotional Data Analysis and Feedback

[0345] The server analyzes the emotion data sent from the emotion engine, evaluates it comprehensively with user feedback, and proposes further improvements to the review process. For analysis, Python analysis libraries (e.g., pandas, numpy) are used.

[0346] Input: Emotion data and user feedback

[0347] Output: Improvement proposal data and feedback report

[0348] Through the above processing steps, it is possible to improve the efficiency of the new subscriber screening process and customer satisfaction.

[0349] (Application example 2)

[0350] 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."

[0351] In the conventional new subscriber screening process, collecting and analyzing application data and improving the screening flow took time, resulting in problems that reduced user satisfaction. Furthermore, bottlenecks in the screening process were often not clearly identified, preventing effective improvements. Furthermore, the screening process proceeded without taking into consideration the user's feelings, potentially increasing user dissatisfaction and stress. A system that solves these problems and realizes an efficient screening process with high user satisfaction is needed.

[0352] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting the data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for recognizing user emotions in real time using an emotion recognition engine, and means for providing feedback based on the recognized emotion data. This enables an efficient and accurate screening process for new applicants and the provision of feedback that takes user emotions into consideration.

[0353] "Past screening data" refers to historical information such as information on the screening process for past new applicants, application documents, screening results, screening period, and reasons for rejection.

[0354] "Information on new subscribers" refers to personal information such as the name, address, and application documents of users who wish to sign up for a new membership, as well as the contents of that application.

[0355] A "bottleneck" refers to a major obstacle or problem that causes delays or inefficiencies in the new member screening process.

[0356] "New review flow" refers to the newly established review procedures and processes to improve, streamline and optimize the existing review flow.

[0357] "Simulation" refers to a method for virtually simulating a new review flow and evaluating its effectiveness and efficiency.

[0358] "Performance" refers to the efficiency resulting from the operation of the established review flow, the rate at which the review period is shortened, the cost reduction effect, and customer satisfaction, etc.

[0359] "Report" refers to a detailed report prepared based on the simulation and prediction results.

[0360] An "emotion recognition engine" refers to a software module that identifies and analyzes emotions from a user's facial expressions and voice in real time.

[0361] "Feedback" refers to information and advice to suggest further improvements and modifications based on user sentiment and reactions to the actual review process.

[0362] "Data analysis" refers to a method of statistically analyzing collected past screening data and information on new applicants to identify problems in the screening process.

[0363] "Visualization" refers to a method of representing analysis and simulation results in a visual format such as a graph or chart.

[0364] "Format" refers to the rules and style for organizing and formatting reports and data according to a particular format or standard.

[0365] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[0366] The system configuration is as follows:

[0367] 1. Server: This is the central component for data collection, analysis, simulation, prediction, report generation, and sentiment data analysis.

[0368] 2. Terminal: Provides an interface for new subscribers to enter information, check reports, and enter feedback.

[0369] 3. User: A person who wishes to join or an operator in charge of improving the screening process.

[0370] 4. Emotion engine: A software module for recognizing the user's emotions in real time and transmitting them to the server.

[0371] Program processing overview

[0372] server

[0373] The server connects to a database to collect past review data. Specifically, it collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis. The server then analyzes the collected data and identifies bottlenecks in the review process. Furthermore, the server simulates a new review flow and predicts performance one and two years from now based on the results. A report is generated based on the prediction and simulation results. Emotion data sent from the emotion engine is also analyzed and reflected as feedback.

[0374] Terminal

[0375] The terminal provides an interface where users can input information about prospective subscribers, which is then transmitted to the server in real time. The terminal also provides generated reports to the users and provides an interface for inputting feedback.

[0376] User

[0377] The user inputs the information of the new subscriber through the terminal, checks the generated report, and also inputs feedback, which is sent to the server.

[0378] Emotion Engine

[0379] The emotion engine recognizes emotions in real time while the user is checking the report and sends the results to a server, such as Google® Cloud Natural Language API or Microsoft® Azure® Text Analytics.

[0380] Hardware and software used

[0381] Hardware: Smartphone (iOS / ANDROID (registered trademark)), server (AWS, GCP, Azure, etc.)

[0382] software:

[0383] Google Cloud Natural Language API or Microsoft Azure Text Analytics: Emotion Recognition Engine

[0384] PostgreSQL or MySQL: Database Management Systems

[0385] Specific examples

[0386] For example, when a prospective new member enters their information using a smartphone, that information is saved in real time on a server. The server then conducts a review based on that information and past data, simulating a new review flow. The results show that adopting a specific flow can shorten the review period. Furthermore, as the user reviews the report, an emotion engine recognizes emotions in real time, and emotions such as dissatisfaction or satisfaction are sent to the server. Based on this, further improvements to the flow are proposed.

[0387] Prompt Sentence Examples

[0388] "Simulate new review flows and evaluate their effectiveness. Identify areas for improvement based on past review data and use simulation results to generate predictive models. Also, recognize user sentiment data in real time and use it to suggest further improvements."

[0389] In this way, the present invention effectively optimizes the screening process for new subscribers and provides feedback that takes into account the user's emotional state, thereby shortening the screening period, improving customer satisfaction, and reducing operating costs.

[0390] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0391] Step 1:

[0392] The server collects past review data from a database. It collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) and performs preprocessing, including missing value imputation and data normalization. The input is the past review data in the database, and the output is data converted into a format that can be used for analysis.

[0393] Step 2:

[0394] The terminal provides a function for inputting information about new members through a user interface. When a user inputs information such as name, address, and application documents into the terminal, this information is sent to the server in real time. The input is the information about new members entered by the user, and the output is the information about new members sent to the server.

[0395] Step 3:

[0396] The server performs statistical analysis based on the collected historical data and information on new applicants. The analysis involves calculating the average and standard deviation of the review period, the frequency of each rejection reason, and the rate of application document defects. The input is the preprocessed historical data and information on new applicants, and the output is the analysis results.

[0397] Step 4:

[0398] The server identifies bottlenecks in the review process through data analysis, visualizes them, and sends them to the terminal. For example, it identifies that incomplete documents are the main cause of review delays. The input is the data analysis results, and the output is visualized bottleneck information.

[0399] Step 5:

[0400] The terminal provides an interface that allows the user to set up a new screening flow (for example, the introduction of a pre-document check process). When the user inputs the new settings, the terminal sends the setting information to the server. The input is the information of the new screening flow set by the user, and the output is the information of the new screening flow sent to the server.

[0401] Step 6:

[0402] The server runs a simulation based on the new review flow. It compares it with the conventional review flow and evaluates the efficiency and effectiveness of each (review period reduction rate, cost reduction effect). The simulation results are visualized and sent to the terminal. The input is information about the new review flow, and the output is the simulation results.

[0403] Step 7:

[0404] The server builds a predictive model based on the simulation results and past data. It uses a machine learning algorithm to predict the performance of the new screening process one and two years from now. The inputs are the simulation results and past data, and the output is the results of the predictive model.

[0405] Step 8:

[0406] The server automatically generates detailed reports based on the simulation and prediction results. The reports are formatted in a specific format (e.g., PDF) and sent to the terminal. The input is the result of the prediction model, and the output is the generated report.

[0407] Step 9:

[0408] The terminal displays the generated report to the user and provides an interface for inputting feedback, where the input is the generated report and the output is the user's feedback information.

[0409] Step 10:

[0410] The emotion engine recognizes the user's emotions in real time as they check the report. The recognition results are sent to the server for analysis. The input is the user's voice and facial expression data, and the output is the emotion recognition results.

[0411] Step 11:

[0412] The server analyzes the emotion data and proposes further improvements to the review process based on the user's emotional state. The inputs are emotion recognition results and user feedback information, and the output is improvement suggestions.

[0413] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0414] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (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.

[0415] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0416] [Second embodiment]

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

[0418] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0419] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0420] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0421] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0422] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0423] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0424] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0425] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0426] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0427] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0428] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0429] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[0430] System Overview

[0431] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[0432] Program processing

[0433] 1. Data Collection

[0434] The server connects to the database and collects the past five years of review data, including application documents, review results, review periods, and reasons for rejection.

[0435] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[0436] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[0437] 2. Data Analysis

[0438] The server analyzes the collected past review data, performs statistical analysis, and calculates the average and standard deviation of the review period, the rate of defects in each document, and the frequency of rejection reasons.

[0439] The server provides the functionality to identify and visualize bottlenecks in the review process through data analysis.

[0440] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[0441] 3. Improvement Simulation

[0442] The terminal provides an interface that allows users to set up new screening flows (for example, introducing pre-document checks).

[0443] The server performs a comparative simulation of the old and new review flows and evaluates their efficiency and effectiveness.

[0444] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[0445] 4. Predictive model generation

[0446] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance of the new review process one and two years from now.

[0447] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[0448] 5. Generate and review reports

[0449] The server generates a report based on the simulation and prediction results, formats it in a specific format, and sends it to the terminal.

[0450] The terminal displays the generated report to the user for the user to review.

[0451] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[0452] In this way, the system of the present invention can effectively optimize the new customer screening process and predict future results, which can be expected to shorten screening times, improve customer satisfaction, and reduce operational costs.

[0453] The processing flow will be explained below.

[0454] Step 1:

[0455] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[0456] Step 2:

[0457] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[0458] Step 3:

[0459] The server performs statistical analysis based on the past review data collected, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects. Next, the analysis results are used to identify bottlenecks in the review process, and the visualized analysis results are sent to the terminal.

[0460] Step 4:

[0461] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[0462] Step 5:

[0463] The server runs a simulation based on the new audit flow that has been set up. It compares the old audit flow with the new one and calculates the efficiency and effectiveness of each (for example, the rate of reduction in audit time and cost reduction). The simulation results are visualized and sent to the terminal.

[0464] Step 6:

[0465] The server uses a machine learning algorithm to build a predictive model based on the simulation results and past data, and uses the built model to predict the performance (e.g., review efficiency, customer satisfaction) of the new screening process one and two years from now.

[0466] Step 7:

[0467] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[0468] Step 8:

[0469] The user can view the generated report via the terminal, which includes the simulation results and the output of the predictive model, providing the user with information to make decisions about future improvements to the review process.

[0470] Step 9:

[0471] Users enter feedback, which is sent from their devices to a server that records it and uses it to improve future simulations and predictive models.

[0472] Example 1

[0473] 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."

[0474] The traditional new customer application screening process involves a large number of documents and information, which can lead to processing delays and bottlenecks. It is also difficult to simulate and predict the effects of an improved screening process in advance. This makes it challenging to improve the efficiency of screening and customer satisfaction.

[0475] 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.

[0476] In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting that data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for providing an interface for a terminal to set a new screening flow, and means for the server to use a machine learning algorithm when analyzing the data. This improves the efficiency of the screening process and enables the design and implementation of a predictable screening flow.

[0477] "Means for collecting past audit data" refers to devices or programs that can collect information related to audits conducted in the past and store it on a server or database.

[0478] The "means for inputting information about new subscribers and collecting that data" refers to an interface and device for inputting personal information and application forms of users who wish to join, and sending them to the server for storage.

[0479] "Means for analyzing collected data and identifying bottlenecks in the review process" refers to analytical tools and algorithms used to analyze collected review data and identify causes of delays and inefficiencies during the review process.

[0480] The "means for setting up a new review flow and simulating it" is a system that provides an interface that allows users to set up a new review flow, and simulates that flow to evaluate its effectiveness.

[0481] "Means for predicting one-year and two-year performance of a configured review flow" refers to a system that runs machine learning algorithms and predictive models to predict future performance of a newly configured review flow.

[0482] The "means for generating a report based on the results of simulation and prediction" refers to a device or program that automatically generates a report based on the results of simulation and prediction, and puts it in a format that can be checked by the user.

[0483] The "means for checking the generated report" refers to an interface and device that displays the generated report to the user and enables the user to check the contents.

[0484] The "means for providing an interface for a terminal to set up a new screening flow" refers to a device and a program for providing an interface for a terminal to allow a user to input settings for a new screening flow.

[0485] "Means for using machine learning algorithms when the server analyzes data" refers to a system for using machine learning algorithms when the server analyzes collected data and for identifying bottlenecks in the review process.

[0486] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[0487] System Overview

[0488] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[0489] Data collection and input

[0490] The server connects to the database and collects screening data from the past five years. This data includes application documents, screening results, screening periods, and reasons for rejection. The terminal provides a function for entering information about applicants through a user interface. The user enters the applicant's information into the terminal, and this information is sent to the server in real time.

[0491] Specific examples

[0492] The user enters Mr. Tanaka's information (name, address, application documents, etc.) into the terminal. This information is sent to the server and stored in the database.

[0493] Data analysis and bottleneck identification

[0494] The server analyzes the collected past review data. Statistical analysis is performed to calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. Through data analysis, the server identifies bottlenecks in the review process and visualizes them.

[0495] Specific examples

[0496] The server analyzes the review data using statistical analysis tools (e.g., Python's Pandas and NumPy). Using the R ggplot2 library, it identifies the main causes of review delays and generates a visualization graph showing that these are incomplete documents.

[0497] Improvement Simulation

[0498] The terminal provides an interface for setting up a new screening flow (for example, introducing a pre-document check). The user sets up the new screening flow, and the server performs a comparative simulation of the old and new screening flows to evaluate their efficiency and effectiveness.

[0499] Specific examples

[0500] The user configures the pre-document check using an HTML form on the terminal. The server runs a Monte Carlo simulation to compare the old and new flows. The new flow shows that the review period is reduced by 20%.

[0501] Predictive Model Generation

[0502] The server builds a predictive model based on the simulation results and past data, using machine learning algorithms such as SciKit-Learn and TensorFlow to predict how the new screening process will perform one and two years from now.

[0503] Specific examples

[0504] The server builds a predictive model and finds that one year later, review efficiency will improve by 15% and customer satisfaction will increase by 10 points.

[0505] Report generation and review

[0506] The server generates a report based on the simulation and prediction results, formats it in a specific format (e.g., PDF, Excel), and sends it to the terminal. The terminal displays the generated report to the user, who can then check the contents.

[0507] Specific examples

[0508] The server generates a PDF report using the ReportLab library and sends it to the terminal, which displays the report to the user using a PDF viewer.The user can then decide to officially implement the new review flow based on the contents of the report.

[0509] Prompt Sentence Examples

[0510] Here is an example of a prompt to use for a generative AI model:

[0511] 1. "This system provides the ability to analyze historical data and identify bottlenecks in the review process. Please explain how data collection and analysis are performed. Include specific tools and methods."

[0512] 2. Please explain in detail the process of comparing the simulation results of the new review flow with the conventional flow. Please also provide specific examples of improvements that result from this.

[0513] 3. "Please describe how the server generates a predictive model and uses it to assess the efficiency of future review processes and customer satisfaction. Please also describe the algorithms and tools used."

[0514] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0515] Step 1:

[0516] The server accesses the database and collects past review data. The server executes SQL queries to retrieve data such as application documents, review results, review periods, and rejection reasons from the past five years. The data is saved in CSV or JSON format and temporarily stored for later analysis. The input to this data collection process is the database connection information to the server, and the output is past review data.

[0517] Specific behavior:

[0518] The server uses an SQL query to extract past review data from the database and save it as a file. Example: SELECT FROM review_data WHERE year >= CURRENT_DATE - INTERVAL '5 years';

[0519] Step 2:

[0520] The terminal provides a user interface, and the user enters information about the new member (e.g., name, address, application form). The terminal converts this information into JSON format and sends it to the server. The input is the information about the new member entered by the user, and the output is JSON format data sent to the server.

[0521] Specific behavior:

[0522] When a user enters the information of the prospective subscriber into the form and presses the submit button, the terminal converts the information into JSON format and sends it to the server as an HTTP POST request.

[0523] Step 3:

[0524] The server saves the collected data of new applicants to a database. The server validates the data format and inserts the data into the appropriate fields. The input is the JSON data of new applicants sent from the terminal, and the output is the information of the applicants saved in the database.

[0525] Specific behavior:

[0526] The server parses the received JSON data and inserts the data into the corresponding table. Example: INSERT INTO applicant_data (name, address, documents) VALUES (json_data["name"], json_data["address"], json_data["documents"]);

[0527] Step 4:

[0528] The server analyzes past review data. It uses statistical analysis tools (e.g., Python's Pandas and NumPy) to analyze the data and calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. The input is past review data, and the output is the analysis results.

[0529] Specific behavior:

[0530] The server runs a Python script, creates a data frame using the Pandas library, and performs statistical analysis. For example: import pandas as pd; data = pd.read_csv('review_data.csv'); mean_duration = data['duration'].mean();

[0531] Step 5:

[0532] The server identifies and visualizes bottlenecks in the review process based on the analysis results. It generates graphs using the R ggplot2 library to identify specific bottlenecks. The input is the analysis results, and the output is the visualized graph and the identified bottlenecks.

[0533] Specific behavior:

[0534] The server runs the R script and generates the graph using the ggplot2 library. For example: library(ggplot2); ggplot(data, aes(x=duration, y=freq)) + geom_bar(stat='identity');

[0535] Step 6:

[0536] The terminal provides an interface for setting up a new screening flow (e.g., introducing a pre-document check). The user sets up a new screening flow, which is then sent to the server. The input is the information on the new screening flow set up by the user, and the output is the flow setting data sent to the server.

[0537] Specific behavior:

[0538] When a user enters a new review flow into the form and presses the submit button, the terminal sends the content to the server.

[0539] Step 7:

[0540] The server runs a simulation comparing the old and new screening flows to evaluate their efficiency and effectiveness. The evaluation is performed using Monte Carlo simulation, etc. The input is information about the old and new screening flows, and the output is the simulation results for each flow.

[0541] Specific behavior:

[0542] The server runs the simulation program and evaluates the results. Example: monte_carlo_simulation(old_flow, new_flow);

[0543] Step 8:

[0544] The server builds a predictive model based on the simulation results and past data. It generates the predictive model using machine learning algorithms such as SciKit-Learn and TensorFlow. The input is the simulation results and past review data, and the output is the generated predictive model.

[0545] Specific behavior:

[0546] The server builds a predictive model using machine learning libraries. Example: from sklearn.linear_model import LinearRegression; model = LinearRegression().fit(X, y);

[0547] Step 9:

[0548] The server generates reports based on the simulation and prediction results and formats them in a specific format (e.g. PDF, Excel). The report is created using libraries such as the ReportLab. The input is the prediction model and the simulation results, and the output is the generated report.

[0549] Specific behavior:

[0550] The server generates a PDF report using ReportLab. Example: from reportlab.lib.pagesizes import letter; from reportlab.pdfgen import canvas; c = canvas.Canvas("output.pdf", pagesize=letter); c.drawString(100, 750, "Simulation and Prediction Report"); c.save();

[0551] Step 10:

[0552] The terminal displays the generated report to the user, allowing the user to confirm the contents. The input is the report sent from the server, and the output is the report confirmed by the user.

[0553] Specific behavior:

[0554] The device opens a PDF viewer and displays the report to the user. Example: The device displays the report using the PDF viewer in the browser.

[0555] (Application example 1)

[0556] 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."

[0557] In the conventional screening process, it is difficult to efficiently identify bottlenecks and improve the screening flow using past data.In addition, there is a lack of means to optimize the screening process for new applicants using mobile devices such as smartphones, so it is difficult to sufficiently shorten the screening time or reduce false positives.

[0558] 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.

[0559] In this invention, the server includes a means for collecting past screening data, a means for inputting information on new applicants and collecting that data, and a means for analyzing the collected data and identifying bottlenecks in the screening process. This enables efficient optimization and improvement of the screening process. Furthermore, by including a means for providing an application that is installed on a smartphone and optimizes the screening process for new applicants, screening time can be shortened and customer satisfaction can be improved.

[0560] "Past review data" refers to information related to the review process, such as past application documents, review results, review periods, and reasons for rejection.

[0561] "Information of new applicants" refers to personal information such as the name, address, and application documents of applicants.

[0562] A "bottleneck" refers to a factor or obstruction that delays progress in the review process.

[0563] "Review flow" refers to the series of steps and processes for reviewing an applicant's information and arriving at a result.

[0564] "Simulation" refers to a virtual experiment or trial conducted based on set conditions and flow.

[0565] A "smartphone" refers to a mobile device that has multifunctional data communication capabilities in addition to mobile phone functions.

[0566] An "application" refers to a software program that is installed on a device such as a smartphone and provides specific functions.

[0567] As an embodiment of the present invention, the following system is embodied.

[0568] 1. System Overview

[0569] This system collects past screening data and inputs and collects information on new applicants to efficiently identify bottlenecks in the screening process and simulate and predict new screening flows. In particular, it aims to optimize the screening process through an application installed on smartphones.

[0570] 2. Hardware and Software Used

[0571] Hardware: smartphones, tablets, servers

[0572] software:

[0573] Python and pandas are used for data collection and analysis.

[0574] For simulation and prediction, we use the random forest algorithm from scikit-learn.

[0575] Data visualization will be done using matplotlib and seaborn.

[0576] Python is used to generate reports and output them in PDF format, etc.

[0577] 3. Data processing and calculation flow

[0578] The server retrieves past screening data from the database and analyzes it to identify bottlenecks in the screening process. For example, it can identify that a specific document deficiency is the main cause of screening delays. The device (smartphone or tablet) provides an interface for entering information about new applicants and has the function of sending the collected information to the server.

[0579] The server has the function of setting up a new review flow (e.g., introducing a pre-document check) and runs a simulation to evaluate its effectiveness. This can result in results such as a shortened review period and improved customer satisfaction. The server generates a report based on the simulation results and prediction model and sends it to the terminal in a specific format such as PDF. The user can check the generated report through the terminal and evaluate its effectiveness.

[0580] 4. Specific examples and prompts for generative AI models

[0581] As a concrete example, the following simulation results can be obtained. For example, if a new review flow is introduced, it is predicted that the review period will be shortened by 20% and review efficiency will improve by 15% after one year. These results are provided to the user as a PDF report.

[0582] An example of a prompt to be input to the generative AI model is as follows:

[0583] Please provide a detailed explanation of the correlation between the rate of document deficiencies and the review period based on the review data from the past five years.

[0584] "Please predict the performance improvement if we implement a new review flow."

[0585] In this way, the system of the present invention effectively optimizes the screening process for new subscribers, shortening the screening period and improving customer satisfaction.

[0586] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0587] Step 1:

[0588] The server collects past review data from the database. Specifically, it connects to the review database and extracts information such as application documents, review results, review periods, and rejection reasons over the past five years. The input to this data collection process is the database connection information, and the output is the collected data in a data frame format.

[0589] Step 2:

[0590] The terminal provides an interface for entering information about prospective new subscribers. The user uses the terminal interface to enter their name, address, and application documents. The entered information is sent in real time to the server, where it is stored in a database. The input to this process is the information entered by the user, and the output is the new subscriber's information stored in the server database.

[0591] Step 3:

[0592] The server analyzes the collected past data. As a specific example of the analysis, the server performs statistical analysis to calculate the average and standard deviation of the review period, the defect rate of each document, and the frequency of rejection reasons. The input of this step is the collected past review data, and the output is a report containing the analysis results.

[0593] Step 4:

[0594] Based on the results of the data analysis, the server identifies bottlenecks in the review process and visualizes them. For example, the server may identify that incomplete documents are the main cause of review delays and generate the results in a visualized format such as a graph. The input to this process is the analysis results, and the output is visualized data on the bottlenecks.

[0595] Step 5:

[0596] The terminal provides an interface for setting up a new screening flow. The user sets up a new flow, such as introducing a pre-document check, through this interface. The set information is sent to the server. The input of this step is the information for the new screening flow set up by the user, and the output is the data for the new flow setting sent to the server.

[0597] Step 6:

[0598] The server simulates the old and new review flows and evaluates their efficiency and effectiveness. Specifically, the server uses the collected data and new flow information to run machine learning algorithms such as RandomForestClassifier to simulate the new flow. The inputs to this step are past data and new flow information, and the output is performance evaluation data as the simulation results.

[0599] Step 7:

[0600] The server builds a predictive model based on the simulation results and past data. For example, it uses a machine learning algorithm to predict how a new screening process will perform one and two years from now. The inputs to this process are the simulation results and past data, and the output is the results of the predictive model.

[0601] Step 8:

[0602] The server generates a report based on the prediction results and sends it to the terminal. The report is formatted in a specific format (e.g., PDF) and includes visualization data. The user can view the generated report through the terminal and evaluate the effectiveness. The input of this step is the results from the predictive model, and the output is the generated report.

[0603] The above steps realize a concrete system for optimizing the screening process for new subscribers.

[0604] 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.

[0605] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[0606] System Overview

[0607] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows. Furthermore, by combining it with an emotion engine that recognizes user emotions, the system can further improve the system based on emotion analysis of user feedback.

[0608] Program processing

[0609] 1. Data Collection

[0610] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[0611] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[0612] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[0613] 2. Data Analysis

[0614] The server performs statistical analysis based on the collected past review data, calculating the average and standard deviation of review times, the frequency of each rejection reason, and the rate of application document defects.

[0615] The server identifies bottlenecks in the review process through data analysis and transmits them to the terminal in a visualized form.

[0616] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[0617] 3. Improvement Simulation

[0618] The terminal provides an interface that allows users to set up a new screening flow (for example, introducing a pre-document check process). When the user inputs new settings, the terminal sends the setting information to the server.

[0619] The server runs a simulation based on the new screening flow. It compares the old and new screening flows and evaluates their efficiency and effectiveness (reduced screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[0620] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[0621] 4. Predictive model generation

[0622] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[0623] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[0624] 5. Generate and review reports

[0625] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[0626] The terminal displays the generated report to the user and provides an interface for the user to review the report and enter feedback.

[0627] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[0628] 6. Emotion Recognition by Emotion Engine

[0629] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[0630] Example: While a user is viewing a report, the emotion engine recognizes the user's emotions (e.g., satisfaction, dissatisfaction, excitement, disappointment) via a webcam or microphone and transmits them to the server in real time.

[0631] 7. Emotional Data Analysis and Feedback

[0632] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[0633] Example: The server analyzes sentiment data and user feedback to identify, for example, which specific process steps are causing user frustration. Based on this, further flow improvements are suggested.

[0634] In this way, the system of the present invention not only effectively optimizes the new customer screening process, but also takes into account the user's emotional state to suggest further improvements, thereby further shortening the screening period, improving customer satisfaction, and reducing operational costs.

[0635] The processing flow will be explained below.

[0636] Step 1:

[0637] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[0638] Step 2:

[0639] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[0640] Step 3:

[0641] Statistical analysis is performed based on past review data collected by the server, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects.

[0642] Step 4:

[0643] The server uses the analysis results to identify bottlenecks in the review process and transmits the visualized analysis results to the terminal, using graphs and heat maps.

[0644] Step 5:

[0645] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[0646] Step 6:

[0647] The server runs a simulation based on the new screening flow that has been set up. The old screening flow is compared with the new one, and the efficiency and effectiveness of each (review period reduction rate, cost reduction effect) are evaluated. The simulation results are visualized in graphs and tables and sent to the terminal.

[0648] Step 7:

[0649] The server builds a predictive model based on the simulation results and past data, and uses machine learning algorithms (such as linear regression or random forest) to predict performance (screening efficiency, customer satisfaction) one and two years into the future.

[0650] Step 8:

[0651] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[0652] Step 9:

[0653] The user checks the generated report through the terminal, evaluates the proposed improvements to the review process based on the report contents, and makes a final decision.

[0654] Step 10:

[0655] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[0656] Step 11:

[0657] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[0658] Example 2

[0659] 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."

[0660] In the conventional review process, bottlenecks were identified by analyzing past data and the effectiveness of new flows was simulated, but there were limitations to improving the efficiency of the overall review process or customer satisfaction. Furthermore, it was not possible to reflect user sentiment in real time, and the quality of feedback was insufficient. As a result, the reduction of review time and operational costs was not fully achieved.

[0661] 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.

[0662] A means of collecting past audit data;

[0663] A means for inputting information of new subscribers and collecting the data;

[0664] A means of analyzing the collected data and identifying bottlenecks in the review process;

[0665] A means to set up a new review flow and simulate it,

[0666] A means of predicting performance one and two years into a set review flow;

[0667] a means for generating reports based on the simulation and prediction results;

[0668] A means of reviewing the generated report;

[0669] A means for recognizing user emotions in real time and collecting the results;

[0670] A means to analyze the collected sentiment data and suggest further improvements to the review process; and

[0671] This will make the screening process more efficient, improve customer satisfaction, and reduce operational costs.

[0672] "Review Data" refers to all data related to the review process, including past application documents, review results, review periods, and reasons for rejection.

[0673] "Information of prospective subscribers" refers to all information provided by users who wish to sign up, such as the name, address, and application documents of users who wish to sign up.

[0674] A "bottleneck" refers to an item or step in the review process that tends to slow down processing and reduce efficiency.

[0675] "Review flow" refers to the order and method of each step and procedure in the review process.

[0676] "Simulation" refers to a computer-based analytical method used to virtually implement a new review process and predict its effects.

[0677] "Performance prediction" refers to using machine learning algorithms to predict review efficiency and customer satisfaction one and two years after the introduction of a new review process.

[0678] "Report" refers to a written report summarizing the simulation and prediction results.

[0679] "Emotion recognition" refers to a technology that analyzes a user's facial expressions and voice to identify the user's emotional state in real time.

[0680] "Emotion Data" refers to data regarding a user's emotional state collected through emotion recognition technology.

[0681] "Recommendations for improving the review process" refers to specific changes and improvements proposed to make the review process more efficient, based on analysis of emotional data and user feedback.

[0682] To implement this invention, a system is used that combines a terminal for collecting past screening data and inputting information on new applicants, a server for analyzing the data and setting up and simulating new screening flows, and an emotion engine that recognizes user emotions in real time.

[0683] System Configuration

[0684] This system consists of multiple pieces of hardware and software, as follows:

[0685] 1. Server: Performs large-volume data processing, analysis, and simulation.

[0686] 2. Client Device: Provides an interface for users to enter information about new subscribers and confirm the data.

[0687] 3. Emotion Recognition Engine: A module that analyzes the user's facial expressions and voice to recognize emotions.

[0688] Hardware and Software

[0689] The servers are equipped with high-performance CPUs and GPUs, and feature a comprehensive database management system (e.g., MySQL, PostgreSQL). Data analysis uses Python and related packages (e.g., pandas, numpy, scikit-learn).

[0690] The terminals used are PCs and tablets, which allow users to access the system through a web browser. Web applications are built using HTML, CSS, and JavaScript (e.g., React.js).

[0691] The emotion engine uses input devices such as cameras and microphones, as well as software (e.g., OpenCV, TensorFlow) that controls and analyzes them.

[0692] Program processing flow

[0693] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). This data is then converted into a format that can be used for analysis after missing values ​​are imputed and data normalization is performed.

[0694] The terminal provides a function to input information of new subscribers through a user interface. When a user inputs information of a subscriber, the information is sent to the server in real time and stored in a database.

[0695] The server performs statistical analysis of the collected audit data to identify bottlenecks. The identified bottlenecks are visualized and sent to the terminal. The server then simulates the new audit flow and compares it with the conventional flow to evaluate its efficiency and effectiveness.

[0696] A predictive model is built based on the simulation results and past data, and performance one and two years after the new flow is implemented is predicted. Finally, a report is generated based on the simulation and prediction results and sent to the terminal in PDF format or other format.

[0697] While users are reviewing the report, the emotion engine recognizes their emotions in real time and sends the data to the server, which analyzes the emotion data and provides user feedback to suggest further improvements to the review process.

[0698] Specific examples

[0699] For example, when a user enters information about a prospective new member into a terminal (such as name, address, and application documents), this information is sent to the server and stored in a database. The server analyzes this data and, based on past screening data, identifies incomplete documents as a major bottleneck. The user sets up a new screening flow from the terminal (for example, introducing a pre-document check process), and the server simulates it, showing that the screening period can be reduced by 20%.

[0700] Example prompts from a generative AI model:

[0701] "Please enter the information of the person wishing to sign up. We will propose the optimal screening process based on the analysis of screening data from the past five years."

[0702] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0703] Step 1: Data collection

[0704] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is then filled in with missing values, normalized, and converted into a format that can be used for analysis. Specifically, SQL queries are executed to extract data from the database and preprocessed using Python.

[0705] Input: Database connection information and SQL query

[0706] Output: Preprocessed audit dataset

[0707] Step 2: Enter information about new members

[0708] The terminal provides a function that allows users to input information about prospective subscribers through a user interface. When a user inputs information such as the prospective subscriber's name, address, and application documents into the terminal, this information is sent to the server in real time. The terminal is built using a web application (e.g., React.js), and the input information is sent to the server via an AJAX request.

[0709] Input: Subscriber information entered by the user into the terminal

[0710] Output: Subscriber information sent to the server

[0711] Step 3: Data analysis

[0712] Statistical analysis is performed based on past review data collected by the server. Specifically, data frames are manipulated using Python's pandas and numpy to calculate the average and standard deviation of review periods, the frequency of each reason for rejection, and the rate of incomplete application documents.

[0713] Input: Preprocessed audit dataset

[0714] Output: Calculated statistics (average review period, frequency of rejection reasons, etc.)

[0715] Step 4: Identify bottlenecks

[0716] The server analyzes the data to identify bottlenecks in the review process. Based on this statistical data, the server identifies specific items (e.g., incomplete documents) or specific process steps (e.g., review delays). The results are visualized in graphs and charts and sent to the terminal.

[0717] Input: Calculated statistic value

[0718] Output: Visualized data of bottleneck identification results

[0719] Step 5: Setting up and simulating a new review flow

[0720] The terminal provides the user with an interface that allows them to set up a new screening flow. The user sets up a new flow (for example, the introduction of a pre-document check process), and the information is sent to the server. Based on this information, the server simulates the old and new screening flows and evaluates the efficiency and effectiveness of each flow (reduction in screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[0721] Input: Information about the new screening flow configured by the user

[0722] Output: Visualized data of the simulation results of the old and new screening flows

[0723] Step 6: Generate a predictive model

[0724] The server generates a predictive model based on the simulation results and past data. Specifically, it uses a Python machine learning package (e.g., scikit-learn) to build a model that predicts the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[0725] Input: Simulation results and past audit data

[0726] Output: Prediction model and results (prediction performance after 1 year and 2 years)

[0727] Step 7: Generate and review the report

[0728] The server automatically generates a detailed report based on the simulation and prediction results. The report is formatted in PDF format or similar and sent to the terminal. The terminal displays the generated report to the user and provides an interface where the user can review the report and enter feedback.

[0729] Input: Prediction model and its results

[0730] Output: Report in PDF format

[0731] Step 8: Emotion Recognition with the Emotion Engine

[0732] The emotion engine recognizes the user's emotions in real time as they check the report. Specifically, it analyzes data collected through the camera and microphone using OpenCV and TensorFlow to generate emotion data, which is then sent to the server.

[0733] Input: Real-time data collected from cameras and microphones

[0734] Output: Emotion data

[0735] Step 9: Emotional Data Analysis and Feedback

[0736] The server analyzes the emotion data sent from the emotion engine, evaluates it comprehensively with user feedback, and proposes further improvements to the review process. For analysis, Python analysis libraries (e.g., pandas, numpy) are used.

[0737] Input: Emotion data and user feedback

[0738] Output: Improvement proposal data and feedback report

[0739] Through the above processing steps, it is possible to improve the efficiency of the new subscriber screening process and customer satisfaction.

[0740] (Application example 2)

[0741] 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."

[0742] In the conventional new subscriber screening process, collecting and analyzing application data and improving the screening flow took time, resulting in problems that reduced user satisfaction. Furthermore, bottlenecks in the screening process were often not clearly identified, preventing effective improvements. Furthermore, the screening process proceeded without taking into consideration the user's feelings, potentially increasing user dissatisfaction and stress. A system that solves these problems and realizes an efficient screening process with high user satisfaction is needed.

[0743] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting the data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for recognizing user emotions in real time using an emotion recognition engine, and means for providing feedback based on the recognized emotion data. This enables an efficient and accurate screening process for new applicants and the provision of feedback that takes user emotions into consideration.

[0744] "Past screening data" refers to historical information such as information on the screening process for past new applicants, application documents, screening results, screening period, and reasons for rejection.

[0745] "Information on new subscribers" refers to personal information such as the name, address, and application documents of users who wish to sign up for a new membership, as well as the contents of that application.

[0746] A "bottleneck" refers to a major obstacle or problem that causes delays or inefficiencies in the new member screening process.

[0747] "New review flow" refers to the newly established review procedures and processes to improve, streamline and optimize the existing review flow.

[0748] "Simulation" refers to a method for virtually simulating a new review flow and evaluating its effectiveness and efficiency.

[0749] "Performance" refers to the efficiency resulting from the operation of the established review flow, the rate at which the review period is shortened, the cost reduction effect, and customer satisfaction, etc.

[0750] "Report" refers to a detailed report prepared based on the simulation and prediction results.

[0751] An "emotion recognition engine" refers to a software module that identifies and analyzes emotions from a user's facial expressions and voice in real time.

[0752] "Feedback" refers to information and advice to suggest further improvements and modifications based on user sentiment and reactions to the actual review process.

[0753] "Data analysis" refers to a method of statistically analyzing collected past screening data and information on new applicants to identify problems in the screening process.

[0754] "Visualization" refers to a method of representing analysis and simulation results in a visual format such as a graph or chart.

[0755] "Format" refers to the rules and style for organizing and formatting reports and data according to a particular format or standard.

[0756] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[0757] The system configuration is as follows:

[0758] 1. Server: This is the central component for data collection, analysis, simulation, prediction, report generation, and sentiment data analysis.

[0759] 2. Terminal: Provides an interface for new subscribers to enter information, check reports, and enter feedback.

[0760] 3. User: A person who wishes to join or an operator in charge of improving the screening process.

[0761] 4. Emotion engine: A software module for recognizing the user's emotions in real time and transmitting them to the server.

[0762] Program processing overview

[0763] server

[0764] The server connects to a database to collect past review data. Specifically, it collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis. The server then analyzes the collected data and identifies bottlenecks in the review process. Furthermore, the server simulates a new review flow and predicts performance one and two years from now based on the results. A report is generated based on the prediction and simulation results. Emotion data sent from the emotion engine is also analyzed and reflected as feedback.

[0765] Terminal

[0766] The terminal provides an interface where users can input information about prospective subscribers, which is then transmitted to the server in real time. The terminal also provides generated reports to the users and provides an interface for inputting feedback.

[0767] User

[0768] The user inputs the information of the new subscriber through the terminal, checks the generated report, and also inputs feedback, which is sent to the server.

[0769] Emotion Engine

[0770] The emotion engine recognizes emotions in real time while the user is reviewing the report and sends the results to a server, such as Google Cloud Natural Language API or Microsoft Azure Text Analytics.

[0771] Hardware and software used

[0772] Hardware: Smartphones (iOS / Android), servers (AWS, GCP, Azure, etc.)

[0773] software:

[0774] Google Cloud Natural Language API or Microsoft Azure Text Analytics: Emotion Recognition Engine

[0775] PostgreSQL or MySQL: Database Management Systems

[0776] Specific examples

[0777] For example, when a prospective new member enters their information using a smartphone, that information is saved in real time on a server. The server then conducts a review based on that information and past data, simulating a new review flow. The results show that adopting a specific flow can shorten the review period. Furthermore, as the user reviews the report, an emotion engine recognizes emotions in real time, and emotions such as dissatisfaction or satisfaction are sent to the server. Based on this, further improvements to the flow are proposed.

[0778] Prompt Sentence Examples

[0779] "Simulate new review flows and evaluate their effectiveness. Identify areas for improvement based on past review data and use simulation results to generate predictive models. Also, recognize user sentiment data in real time and use it to suggest further improvements."

[0780] In this way, the present invention effectively optimizes the screening process for new subscribers and provides feedback that takes into account the user's emotional state, thereby shortening the screening period, improving customer satisfaction, and reducing operating costs.

[0781] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0782] Step 1:

[0783] The server collects past review data from a database. It collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) and performs preprocessing, including missing value imputation and data normalization. The input is the past review data in the database, and the output is data converted into a format that can be used for analysis.

[0784] Step 2:

[0785] The terminal provides a function for inputting information about new members through a user interface. When a user inputs information such as name, address, and application documents into the terminal, this information is sent to the server in real time. The input is the information about new members entered by the user, and the output is the information about new members sent to the server.

[0786] Step 3:

[0787] The server performs statistical analysis based on the collected historical data and information on new applicants. The analysis involves calculating the average and standard deviation of the review period, the frequency of each rejection reason, and the rate of application document defects. The input is the preprocessed historical data and information on new applicants, and the output is the analysis results.

[0788] Step 4:

[0789] The server identifies bottlenecks in the review process through data analysis, visualizes them, and sends them to the terminal. For example, it identifies that incomplete documents are the main cause of review delays. The input is the data analysis results, and the output is visualized bottleneck information.

[0790] Step 5:

[0791] The terminal provides an interface that allows the user to set up a new screening flow (for example, the introduction of a pre-document check process). When the user inputs the new settings, the terminal sends the setting information to the server. The input is the information of the new screening flow set by the user, and the output is the information of the new screening flow sent to the server.

[0792] Step 6:

[0793] The server runs a simulation based on the new review flow. It compares it with the conventional review flow and evaluates the efficiency and effectiveness of each (review period reduction rate, cost reduction effect). The simulation results are visualized and sent to the terminal. The input is information about the new review flow, and the output is the simulation results.

[0794] Step 7:

[0795] The server builds a predictive model based on the simulation results and past data. It uses a machine learning algorithm to predict the performance of the new screening process one and two years from now. The inputs are the simulation results and past data, and the output is the results of the predictive model.

[0796] Step 8:

[0797] The server automatically generates detailed reports based on the simulation and prediction results. The reports are formatted in a specific format (e.g., PDF) and sent to the terminal. The input is the result of the prediction model, and the output is the generated report.

[0798] Step 9:

[0799] The terminal displays the generated report to the user and provides an interface for inputting feedback, where the input is the generated report and the output is the user's feedback information.

[0800] Step 10:

[0801] The emotion engine recognizes the user's emotions in real time as they check the report. The recognition results are sent to the server for analysis. The input is the user's voice and facial expression data, and the output is the emotion recognition results.

[0802] Step 11:

[0803] The server analyzes the emotion data and proposes further improvements to the review process based on the user's emotional state. The inputs are emotion recognition results and user feedback information, and the output is improvement suggestions.

[0804] 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.

[0805] 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.

[0806] 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.

[0807] [Third embodiment]

[0808] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0809] 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.

[0810] 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).

[0811] 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.

[0812] 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.

[0813] 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).

[0814] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0815] 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.

[0816] 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.

[0817] 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.

[0818] 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.

[0819] 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."

[0820] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[0821] System Overview

[0822] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[0823] Program processing

[0824] 1. Data Collection

[0825] The server connects to the database and collects the past five years of review data, including application documents, review results, review periods, and reasons for rejection.

[0826] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[0827] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[0828] 2. Data Analysis

[0829] The server analyzes the collected past review data, performs statistical analysis, and calculates the average and standard deviation of the review period, the rate of defects in each document, and the frequency of rejection reasons.

[0830] The server provides the functionality to identify and visualize bottlenecks in the review process through data analysis.

[0831] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[0832] 3. Improvement Simulation

[0833] The terminal provides an interface that allows users to set up new screening flows (for example, introducing pre-document checks).

[0834] The server performs a comparative simulation of the old and new review flows and evaluates their efficiency and effectiveness.

[0835] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[0836] 4. Predictive model generation

[0837] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance of the new review process one and two years from now.

[0838] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[0839] 5. Generate and review reports

[0840] The server generates a report based on the simulation and prediction results, formats it in a specific format, and sends it to the terminal.

[0841] The terminal displays the generated report to the user for the user to review.

[0842] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[0843] In this way, the system of the present invention can effectively optimize the new customer screening process and predict future results, which can be expected to shorten screening times, improve customer satisfaction, and reduce operational costs.

[0844] The processing flow will be explained below.

[0845] Step 1:

[0846] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[0847] Step 2:

[0848] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[0849] Step 3:

[0850] The server performs statistical analysis based on the past review data collected, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects. Next, the analysis results are used to identify bottlenecks in the review process, and the visualized analysis results are sent to the terminal.

[0851] Step 4:

[0852] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[0853] Step 5:

[0854] The server runs a simulation based on the new audit flow that has been set up. It compares the old audit flow with the new one and calculates the efficiency and effectiveness of each (for example, the rate of reduction in audit time and cost reduction). The simulation results are visualized and sent to the terminal.

[0855] Step 6:

[0856] The server uses a machine learning algorithm to build a predictive model based on the simulation results and past data, and uses the built model to predict the performance (e.g., review efficiency, customer satisfaction) of the new screening process one and two years from now.

[0857] Step 7:

[0858] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[0859] Step 8:

[0860] The user can view the generated report via the terminal, which includes the simulation results and the output of the predictive model, providing the user with information to make decisions about future improvements to the review process.

[0861] Step 9:

[0862] Users enter feedback, which is sent from their devices to a server that records it and uses it to improve future simulations and predictive models.

[0863] Example 1

[0864] 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."

[0865] The traditional new customer application screening process involves a large number of documents and information, which can lead to processing delays and bottlenecks. It is also difficult to simulate and predict the effects of an improved screening process in advance. This makes it challenging to improve the efficiency of screening and customer satisfaction.

[0866] 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.

[0867] In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting that data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for providing an interface for a terminal to set a new screening flow, and means for the server to use a machine learning algorithm when analyzing the data. This improves the efficiency of the screening process and enables the design and implementation of a predictable screening flow.

[0868] "Means for collecting past audit data" refers to devices or programs that can collect information related to audits conducted in the past and store it on a server or database.

[0869] The "means for inputting information about new subscribers and collecting that data" refers to an interface and device for inputting personal information and application forms of users who wish to join, and sending them to the server for storage.

[0870] "Means for analyzing collected data and identifying bottlenecks in the review process" refers to analytical tools and algorithms used to analyze collected review data and identify causes of delays and inefficiencies during the review process.

[0871] The "means for setting up a new review flow and simulating it" is a system that provides an interface that allows users to set up a new review flow, and simulates that flow to evaluate its effectiveness.

[0872] "Means for predicting one-year and two-year performance of a configured review flow" refers to a system that runs machine learning algorithms and predictive models to predict future performance of a newly configured review flow.

[0873] The "means for generating a report based on the results of simulation and prediction" refers to a device or program that automatically generates a report based on the results of simulation and prediction, and puts it in a format that can be checked by the user.

[0874] The "means for checking the generated report" refers to an interface and device that displays the generated report to the user and enables the user to check the contents.

[0875] The "means for providing an interface for a terminal to set up a new screening flow" refers to a device and a program for providing an interface for a terminal to allow a user to input settings for a new screening flow.

[0876] "Means for using machine learning algorithms when the server analyzes data" refers to a system for using machine learning algorithms when the server analyzes collected data and for identifying bottlenecks in the review process.

[0877] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[0878] System Overview

[0879] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[0880] Data collection and input

[0881] The server connects to the database and collects screening data from the past five years. This data includes application documents, screening results, screening periods, and reasons for rejection. The terminal provides a function for entering information about applicants through a user interface. The user enters the applicant's information into the terminal, and this information is sent to the server in real time.

[0882] Specific examples

[0883] The user enters Mr. Tanaka's information (name, address, application documents, etc.) into the terminal. This information is sent to the server and stored in the database.

[0884] Data analysis and bottleneck identification

[0885] The server analyzes the collected past review data. Statistical analysis is performed to calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. Through data analysis, the server identifies bottlenecks in the review process and visualizes them.

[0886] Specific examples

[0887] The server analyzes the review data using statistical analysis tools (e.g., Python's Pandas and NumPy). Using the R ggplot2 library, it identifies the main causes of review delays and generates a visualization graph showing that these are incomplete documents.

[0888] Improvement Simulation

[0889] The terminal provides an interface for setting up a new screening flow (for example, introducing a pre-document check). The user sets up the new screening flow, and the server performs a comparative simulation of the old and new screening flows to evaluate their efficiency and effectiveness.

[0890] Specific examples

[0891] The user configures the pre-document check using an HTML form on the terminal. The server runs a Monte Carlo simulation to compare the old and new flows. The new flow shows that the review period is reduced by 20%.

[0892] Predictive Model Generation

[0893] The server builds a predictive model based on the simulation results and past data, using machine learning algorithms such as SciKit-Learn and TensorFlow to predict how the new screening process will perform one and two years from now.

[0894] Specific examples

[0895] The server builds a predictive model and finds that one year later, review efficiency will improve by 15% and customer satisfaction will increase by 10 points.

[0896] Report generation and review

[0897] The server generates a report based on the simulation and prediction results, formats it in a specific format (e.g., PDF, Excel), and sends it to the terminal. The terminal displays the generated report to the user, who can then check the contents.

[0898] Specific examples

[0899] The server generates a PDF report using the ReportLab library and sends it to the terminal, which displays the report to the user using a PDF viewer.The user can then decide to officially implement the new review flow based on the contents of the report.

[0900] Prompt Sentence Examples

[0901] Here is an example of a prompt to use for a generative AI model:

[0902] 1. "This system provides the ability to analyze historical data and identify bottlenecks in the review process. Please explain how data collection and analysis are performed. Include specific tools and methods."

[0903] 2. Please explain in detail the process of comparing the simulation results of the new review flow with the conventional flow. Please also provide specific examples of improvements that result from this.

[0904] 3. "Please describe how the server generates a predictive model and uses it to assess the efficiency of future review processes and customer satisfaction. Please also describe the algorithms and tools used."

[0905] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0906] Step 1:

[0907] The server accesses the database and collects past review data. The server executes SQL queries to retrieve data such as application documents, review results, review periods, and rejection reasons from the past five years. The data is saved in CSV or JSON format and temporarily stored for later analysis. The input to this data collection process is the database connection information to the server, and the output is past review data.

[0908] Specific behavior:

[0909] The server uses an SQL query to extract past review data from the database and save it as a file. Example: SELECT FROM review_data WHERE year >= CURRENT_DATE - INTERVAL '5 years';

[0910] Step 2:

[0911] The terminal provides a user interface, and the user enters information about the new member (e.g., name, address, application form). The terminal converts this information into JSON format and sends it to the server. The input is the information about the new member entered by the user, and the output is JSON format data sent to the server.

[0912] Specific behavior:

[0913] When a user enters the information of the prospective subscriber into the form and presses the submit button, the terminal converts the information into JSON format and sends it to the server as an HTTP POST request.

[0914] Step 3:

[0915] The server saves the collected data of new applicants to a database. The server validates the data format and inserts the data into the appropriate fields. The input is the JSON data of new applicants sent from the terminal, and the output is the information of the applicants saved in the database.

[0916] Specific behavior:

[0917] The server parses the received JSON data and inserts the data into the corresponding table. Example: INSERT INTO applicant_data (name, address, documents) VALUES (json_data["name"], json_data["address"], json_data["documents"]);

[0918] Step 4:

[0919] The server analyzes past review data. It uses statistical analysis tools (e.g., Python's Pandas and NumPy) to analyze the data and calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. The input is past review data, and the output is the analysis results.

[0920] Specific behavior:

[0921] The server runs a Python script, creates a data frame using the Pandas library, and performs statistical analysis. For example: import pandas as pd; data = pd.read_csv('review_data.csv'); mean_duration = data['duration'].mean();

[0922] Step 5:

[0923] The server identifies and visualizes bottlenecks in the review process based on the analysis results. It generates graphs using the R ggplot2 library to identify specific bottlenecks. The input is the analysis results, and the output is the visualized graph and the identified bottlenecks.

[0924] Specific behavior:

[0925] The server runs the R script and generates the graph using the ggplot2 library. For example: library(ggplot2); ggplot(data, aes(x=duration, y=freq)) + geom_bar(stat='identity');

[0926] Step 6:

[0927] The terminal provides an interface for setting up a new screening flow (e.g., introducing a pre-document check). The user sets up a new screening flow, which is then sent to the server. The input is the information on the new screening flow set up by the user, and the output is the flow setting data sent to the server.

[0928] Specific behavior:

[0929] When a user enters a new review flow into the form and presses the submit button, the terminal sends the content to the server.

[0930] Step 7:

[0931] The server runs a simulation comparing the old and new screening flows to evaluate their efficiency and effectiveness. The evaluation is performed using Monte Carlo simulation, etc. The input is information about the old and new screening flows, and the output is the simulation results for each flow.

[0932] Specific behavior:

[0933] The server runs the simulation program and evaluates the results. Example: monte_carlo_simulation(old_flow, new_flow);

[0934] Step 8:

[0935] The server builds a predictive model based on the simulation results and past data. It generates the predictive model using machine learning algorithms such as SciKit-Learn and TensorFlow. The input is the simulation results and past review data, and the output is the generated predictive model.

[0936] Specific behavior:

[0937] The server builds a predictive model using machine learning libraries. Example: from sklearn.linear_model import LinearRegression; model = LinearRegression().fit(X, y);

[0938] Step 9:

[0939] The server generates reports based on the simulation and prediction results and formats them in a specific format (e.g. PDF, Excel). The report is created using libraries such as the ReportLab. The input is the prediction model and the simulation results, and the output is the generated report.

[0940] Specific behavior:

[0941] The server generates a PDF report using ReportLab. Example: from reportlab.lib.pagesizes import letter; from reportlab.pdfgen import canvas; c = canvas.Canvas("output.pdf", pagesize=letter); c.drawString(100, 750, "Simulation and Prediction Report"); c.save();

[0942] Step 10:

[0943] The terminal displays the generated report to the user, allowing the user to confirm the contents. The input is the report sent from the server, and the output is the report confirmed by the user.

[0944] Specific behavior:

[0945] The device opens a PDF viewer and displays the report to the user. Example: The device displays the report using the PDF viewer in the browser.

[0946] (Application example 1)

[0947] 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."

[0948] In the conventional screening process, it is difficult to efficiently identify bottlenecks and improve the screening flow using past data.In addition, there is a lack of means to optimize the screening process for new applicants using mobile devices such as smartphones, so it is difficult to sufficiently shorten the screening time or reduce false positives.

[0949] 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.

[0950] In this invention, the server includes a means for collecting past screening data, a means for inputting information on new applicants and collecting that data, and a means for analyzing the collected data and identifying bottlenecks in the screening process. This enables efficient optimization and improvement of the screening process. Furthermore, by including a means for providing an application that is installed on a smartphone and optimizes the screening process for new applicants, screening time can be shortened and customer satisfaction can be improved.

[0951] "Past review data" refers to information related to the review process, such as past application documents, review results, review periods, and reasons for rejection.

[0952] "Information of new applicants" refers to personal information such as the name, address, and application documents of applicants.

[0953] A "bottleneck" refers to a factor or obstruction that delays progress in the review process.

[0954] "Review flow" refers to the series of steps and processes for reviewing an applicant's information and arriving at a result.

[0955] "Simulation" refers to a virtual experiment or trial conducted based on set conditions and flow.

[0956] A "smartphone" refers to a mobile device that has multifunctional data communication capabilities in addition to mobile phone functions.

[0957] An "application" refers to a software program that is installed on a device such as a smartphone and provides specific functions.

[0958] As an embodiment of the present invention, the following system is embodied.

[0959] 1. System Overview

[0960] This system collects past screening data and inputs and collects information on new applicants to efficiently identify bottlenecks in the screening process and simulate and predict new screening flows. In particular, it aims to optimize the screening process through an application installed on smartphones.

[0961] 2. Hardware and Software Used

[0962] Hardware: smartphones, tablets, servers

[0963] software:

[0964] Python and pandas are used for data collection and analysis.

[0965] For simulation and prediction, we use the random forest algorithm from scikit-learn.

[0966] Data visualization will be done using matplotlib and seaborn.

[0967] Python is used to generate reports and output them in PDF format, etc.

[0968] 3. Data processing and calculation flow

[0969] The server retrieves past screening data from the database and analyzes it to identify bottlenecks in the screening process. For example, it can identify that a specific document deficiency is the main cause of screening delays. The device (smartphone or tablet) provides an interface for entering information about new applicants and has the function of sending the collected information to the server.

[0970] The server has the function of setting up a new review flow (e.g., introducing a pre-document check) and runs a simulation to evaluate its effectiveness. This can result in results such as a shortened review period and improved customer satisfaction. The server generates a report based on the simulation results and prediction model and sends it to the terminal in a specific format such as PDF. The user can check the generated report through the terminal and evaluate its effectiveness.

[0971] 4. Specific examples and prompts for generative AI models

[0972] As a concrete example, the following simulation results can be obtained. For example, if a new review flow is introduced, it is predicted that the review period will be shortened by 20% and review efficiency will improve by 15% after one year. These results are provided to the user as a PDF report.

[0973] An example of a prompt to be input to the generative AI model is as follows:

[0974] Please provide a detailed explanation of the correlation between the rate of document deficiencies and the review period based on the review data from the past five years.

[0975] "Please predict the performance improvement if we implement a new review flow."

[0976] In this way, the system of the present invention effectively optimizes the screening process for new subscribers, shortening the screening period and improving customer satisfaction.

[0977] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0978] Step 1:

[0979] The server collects past review data from the database. Specifically, it connects to the review database and extracts information such as application documents, review results, review periods, and rejection reasons over the past five years. The input to this data collection process is the database connection information, and the output is the collected data in a data frame format.

[0980] Step 2:

[0981] The terminal provides an interface for entering information about prospective new subscribers. The user uses the terminal interface to enter their name, address, and application documents. The entered information is sent in real time to the server, where it is stored in a database. The input to this process is the information entered by the user, and the output is the new subscriber's information stored in the server database.

[0982] Step 3:

[0983] The server analyzes the collected past data. As a specific example of the analysis, the server performs statistical analysis to calculate the average and standard deviation of the review period, the defect rate of each document, and the frequency of rejection reasons. The input of this step is the collected past review data, and the output is a report containing the analysis results.

[0984] Step 4:

[0985] Based on the results of the data analysis, the server identifies bottlenecks in the review process and visualizes them. For example, the server may identify that incomplete documents are the main cause of review delays and generate the results in a visualized format such as a graph. The input to this process is the analysis results, and the output is visualized data on the bottlenecks.

[0986] Step 5:

[0987] The terminal provides an interface for setting up a new screening flow. The user sets up a new flow, such as introducing a pre-document check, through this interface. The set information is sent to the server. The input of this step is the information for the new screening flow set up by the user, and the output is the data for the new flow setting sent to the server.

[0988] Step 6:

[0989] The server simulates the old and new review flows and evaluates their efficiency and effectiveness. Specifically, the server uses the collected data and new flow information to run machine learning algorithms such as RandomForestClassifier to simulate the new flow. The inputs to this step are past data and new flow information, and the output is performance evaluation data as the simulation results.

[0990] Step 7:

[0991] The server builds a predictive model based on the simulation results and past data. For example, it uses a machine learning algorithm to predict how a new screening process will perform one and two years from now. The inputs to this process are the simulation results and past data, and the output is the results of the predictive model.

[0992] Step 8:

[0993] The server generates a report based on the prediction results and sends it to the terminal. The report is formatted in a specific format (e.g., PDF) and includes visualization data. The user can view the generated report through the terminal and evaluate the effectiveness. The input of this step is the results from the predictive model, and the output is the generated report.

[0994] The above steps realize a concrete system for optimizing the screening process for new subscribers.

[0995] 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.

[0996] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[0997] System Overview

[0998] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows. Furthermore, by combining it with an emotion engine that recognizes user emotions, the system can further improve the system based on emotion analysis of user feedback.

[0999] Program processing

[1000] 1. Data Collection

[1001] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[1002] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[1003] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[1004] 2. Data Analysis

[1005] The server performs statistical analysis based on the collected past review data, calculating the average and standard deviation of review times, the frequency of each rejection reason, and the rate of application document defects.

[1006] The server identifies bottlenecks in the review process through data analysis and transmits them to the terminal in a visualized form.

[1007] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[1008] 3. Improvement Simulation

[1009] The terminal provides an interface that allows users to set up a new screening flow (for example, introducing a pre-document check process). When the user inputs new settings, the terminal sends the setting information to the server.

[1010] The server runs a simulation based on the new screening flow. It compares the old and new screening flows and evaluates their efficiency and effectiveness (reduced screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[1011] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[1012] 4. Predictive model generation

[1013] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[1014] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[1015] 5. Generate and review reports

[1016] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[1017] The terminal displays the generated report to the user and provides an interface for the user to review the report and enter feedback.

[1018] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[1019] 6. Emotion Recognition by Emotion Engine

[1020] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[1021] Example: While a user is viewing a report, the emotion engine recognizes the user's emotions (e.g., satisfaction, dissatisfaction, excitement, disappointment) via a webcam or microphone and transmits them to the server in real time.

[1022] 7. Emotional Data Analysis and Feedback

[1023] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[1024] Example: The server analyzes sentiment data and user feedback to identify, for example, which specific process steps are causing user frustration. Based on this, further flow improvements are suggested.

[1025] In this way, the system of the present invention not only effectively optimizes the new customer screening process, but also takes into account the user's emotional state to suggest further improvements, thereby further shortening the screening period, improving customer satisfaction, and reducing operational costs.

[1026] The processing flow will be explained below.

[1027] Step 1:

[1028] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[1029] Step 2:

[1030] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[1031] Step 3:

[1032] Statistical analysis is performed based on past review data collected by the server, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects.

[1033] Step 4:

[1034] The server uses the analysis results to identify bottlenecks in the review process and transmits the visualized analysis results to the terminal, using graphs and heat maps.

[1035] Step 5:

[1036] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[1037] Step 6:

[1038] The server runs a simulation based on the new screening flow that has been set up. The old screening flow is compared with the new one, and the efficiency and effectiveness of each (review period reduction rate, cost reduction effect) are evaluated. The simulation results are visualized in graphs and tables and sent to the terminal.

[1039] Step 7:

[1040] The server builds a predictive model based on the simulation results and past data, and uses machine learning algorithms (such as linear regression or random forest) to predict performance (screening efficiency, customer satisfaction) one and two years into the future.

[1041] Step 8:

[1042] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[1043] Step 9:

[1044] The user checks the generated report through the terminal, evaluates the proposed improvements to the review process based on the report contents, and makes a final decision.

[1045] Step 10:

[1046] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[1047] Step 11:

[1048] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[1049] Example 2

[1050] 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."

[1051] In the conventional review process, bottlenecks were identified by analyzing past data and the effectiveness of new flows was simulated, but there were limitations to improving the efficiency of the overall review process or customer satisfaction. Furthermore, it was not possible to reflect user sentiment in real time, and the quality of feedback was insufficient. As a result, the reduction of review time and operational costs was not fully achieved.

[1052] 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.

[1053] A means of collecting past audit data;

[1054] A means for inputting information of new subscribers and collecting the data;

[1055] A means of analyzing the collected data and identifying bottlenecks in the review process;

[1056] A means to set up a new review flow and simulate it,

[1057] A means of predicting performance one and two years into a set review flow;

[1058] a means for generating reports based on the simulation and prediction results;

[1059] A means of reviewing the generated report;

[1060] A means for recognizing user emotions in real time and collecting the results;

[1061] A means to analyze the collected sentiment data and suggest further improvements to the review process; and

[1062] This will make the screening process more efficient, improve customer satisfaction, and reduce operational costs.

[1063] "Review Data" refers to all data related to the review process, including past application documents, review results, review periods, and reasons for rejection.

[1064] "Information of prospective subscribers" refers to all information provided by users who wish to sign up, such as the name, address, and application documents of users who wish to sign up.

[1065] A "bottleneck" refers to an item or step in the review process that tends to slow down processing and reduce efficiency.

[1066] "Review flow" refers to the order and method of each step and procedure in the review process.

[1067] "Simulation" refers to a computer-based analytical method used to virtually implement a new review process and predict its effects.

[1068] "Performance prediction" refers to using machine learning algorithms to predict review efficiency and customer satisfaction one and two years after the introduction of a new review process.

[1069] "Report" refers to a written report summarizing the simulation and prediction results.

[1070] "Emotion recognition" refers to a technology that analyzes a user's facial expressions and voice to identify the user's emotional state in real time.

[1071] "Emotion Data" refers to data regarding a user's emotional state collected through emotion recognition technology.

[1072] "Recommendations for improving the review process" refers to specific changes and improvements proposed to make the review process more efficient, based on analysis of emotional data and user feedback.

[1073] To implement this invention, a system is used that combines a terminal for collecting past screening data and inputting information on new applicants, a server for analyzing the data and setting up and simulating new screening flows, and an emotion engine that recognizes user emotions in real time.

[1074] System Configuration

[1075] This system consists of multiple pieces of hardware and software, as follows:

[1076] 1. Server: Performs large-volume data processing, analysis, and simulation.

[1077] 2. Client Device: Provides an interface for users to enter information about new subscribers and confirm the data.

[1078] 3. Emotion Recognition Engine: A module that analyzes the user's facial expressions and voice to recognize emotions.

[1079] Hardware and Software

[1080] The servers are equipped with high-performance CPUs and GPUs, and feature a comprehensive database management system (e.g., MySQL, PostgreSQL). Data analysis uses Python and related packages (e.g., pandas, numpy, scikit-learn).

[1081] The terminals used are PCs and tablets, which allow users to access the system through a web browser. Web applications are built using HTML, CSS, and JavaScript (e.g., React.js).

[1082] The emotion engine uses input devices such as cameras and microphones, as well as software (e.g., OpenCV, TensorFlow) that controls and analyzes them.

[1083] Program processing flow

[1084] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). This data is then converted into a format that can be used for analysis after missing values ​​are imputed and data normalization is performed.

[1085] The terminal provides a function to input information of new subscribers through a user interface. When a user inputs information of a subscriber, the information is sent to the server in real time and stored in a database.

[1086] The server performs statistical analysis of the collected audit data to identify bottlenecks. The identified bottlenecks are visualized and sent to the terminal. The server then simulates the new audit flow and compares it with the conventional flow to evaluate its efficiency and effectiveness.

[1087] A predictive model is built based on the simulation results and past data, and performance one and two years after the new flow is implemented is predicted. Finally, a report is generated based on the simulation and prediction results and sent to the terminal in PDF format or other format.

[1088] While users are reviewing the report, the emotion engine recognizes their emotions in real time and sends the data to the server, which analyzes the emotion data and provides user feedback to suggest further improvements to the review process.

[1089] Specific examples

[1090] For example, when a user enters information about a prospective new member into a terminal (such as name, address, and application documents), this information is sent to the server and stored in a database. The server analyzes this data and, based on past screening data, identifies incomplete documents as a major bottleneck. The user sets up a new screening flow from the terminal (for example, introducing a pre-document check process), and the server simulates it, showing that the screening period can be reduced by 20%.

[1091] Example prompts from a generative AI model:

[1092] "Please enter the information of the person wishing to sign up. We will propose the optimal screening process based on the analysis of screening data from the past five years."

[1093] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1094] Step 1: Data collection

[1095] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is then filled in with missing values, normalized, and converted into a format that can be used for analysis. Specifically, SQL queries are executed to extract data from the database and preprocessed using Python.

[1096] Input: Database connection information and SQL query

[1097] Output: Preprocessed audit dataset

[1098] Step 2: Enter information about new members

[1099] The terminal provides a function that allows users to input information about prospective subscribers through a user interface. When a user inputs information such as the prospective subscriber's name, address, and application documents into the terminal, this information is sent to the server in real time. The terminal is built using a web application (e.g., React.js), and the input information is sent to the server via an AJAX request.

[1100] Input: Subscriber information entered by the user into the terminal

[1101] Output: Subscriber information sent to the server

[1102] Step 3: Data analysis

[1103] Statistical analysis is performed based on past review data collected by the server. Specifically, data frames are manipulated using Python's pandas and numpy to calculate the average and standard deviation of review periods, the frequency of each reason for rejection, and the rate of incomplete application documents.

[1104] Input: Preprocessed audit dataset

[1105] Output: Calculated statistics (average review period, frequency of rejection reasons, etc.)

[1106] Step 4: Identify bottlenecks

[1107] The server analyzes the data to identify bottlenecks in the review process. Based on this statistical data, the server identifies specific items (e.g., incomplete documents) or specific process steps (e.g., review delays). The results are visualized in graphs and charts and sent to the terminal.

[1108] Input: Calculated statistic value

[1109] Output: Visualized data of bottleneck identification results

[1110] Step 5: Setting up and simulating a new review flow

[1111] The terminal provides the user with an interface that allows them to set up a new screening flow. The user sets up a new flow (for example, the introduction of a pre-document check process), and the information is sent to the server. Based on this information, the server simulates the old and new screening flows and evaluates the efficiency and effectiveness of each flow (reduction in screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[1112] Input: Information about the new screening flow configured by the user

[1113] Output: Visualized data of the simulation results of the old and new screening flows

[1114] Step 6: Generate a predictive model

[1115] The server generates a predictive model based on the simulation results and past data. Specifically, it uses a Python machine learning package (e.g., scikit-learn) to build a model that predicts the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[1116] Input: Simulation results and past audit data

[1117] Output: Prediction model and results (prediction performance after 1 year and 2 years)

[1118] Step 7: Generate and review the report

[1119] The server automatically generates a detailed report based on the simulation and prediction results. The report is formatted in PDF format or similar and sent to the terminal. The terminal displays the generated report to the user and provides an interface where the user can review the report and enter feedback.

[1120] Input: Prediction model and its results

[1121] Output: Report in PDF format

[1122] Step 8: Emotion Recognition with the Emotion Engine

[1123] The emotion engine recognizes the user's emotions in real time as they check the report. Specifically, it analyzes data collected through the camera and microphone using OpenCV and TensorFlow to generate emotion data, which is then sent to the server.

[1124] Input: Real-time data collected from cameras and microphones

[1125] Output: Emotion data

[1126] Step 9: Emotional Data Analysis and Feedback

[1127] The server analyzes the emotion data sent from the emotion engine, evaluates it comprehensively with user feedback, and proposes further improvements to the review process. For analysis, Python analysis libraries (e.g., pandas, numpy) are used.

[1128] Input: Emotion data and user feedback

[1129] Output: Improvement proposal data and feedback report

[1130] Through the above processing steps, it is possible to improve the efficiency of the new subscriber screening process and customer satisfaction.

[1131] (Application example 2)

[1132] 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."

[1133] In the conventional new subscriber screening process, collecting and analyzing application data and improving the screening flow took time, resulting in problems that reduced user satisfaction. Furthermore, bottlenecks in the screening process were often not clearly identified, preventing effective improvements. Furthermore, the screening process proceeded without taking into consideration the user's feelings, potentially increasing user dissatisfaction and stress. A system that solves these problems and realizes an efficient screening process with high user satisfaction is needed.

[1134] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting the data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for recognizing user emotions in real time using an emotion recognition engine, and means for providing feedback based on the recognized emotion data. This enables an efficient and accurate screening process for new applicants and the provision of feedback that takes user emotions into consideration.

[1135] "Past screening data" refers to historical information such as information on the screening process for past new applicants, application documents, screening results, screening period, and reasons for rejection.

[1136] "Information on new subscribers" refers to personal information such as the name, address, and application documents of users who wish to sign up for a new membership, as well as the contents of that application.

[1137] A "bottleneck" refers to a major obstacle or problem that causes delays or inefficiencies in the new member screening process.

[1138] "New review flow" refers to the newly established review procedures and processes to improve, streamline and optimize the existing review flow.

[1139] "Simulation" refers to a method for virtually simulating a new review flow and evaluating its effectiveness and efficiency.

[1140] "Performance" refers to the efficiency resulting from the operation of the established review flow, the rate at which the review period is shortened, the cost reduction effect, and customer satisfaction, etc.

[1141] "Report" refers to a detailed report prepared based on the simulation and prediction results.

[1142] An "emotion recognition engine" refers to a software module that identifies and analyzes emotions from a user's facial expressions and voice in real time.

[1143] "Feedback" refers to information and advice to suggest further improvements and modifications based on user sentiment and reactions to the actual review process.

[1144] "Data analysis" refers to a method of statistically analyzing collected past screening data and information on new applicants to identify problems in the screening process.

[1145] "Visualization" refers to a method of representing analysis and simulation results in a visual format such as a graph or chart.

[1146] "Format" refers to the rules and style for organizing and formatting reports and data according to a particular format or standard.

[1147] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[1148] The system configuration is as follows:

[1149] 1. Server: This is the central component for data collection, analysis, simulation, prediction, report generation, and sentiment data analysis.

[1150] 2. Terminal: Provides an interface for new subscribers to enter information, check reports, and enter feedback.

[1151] 3. User: A person who wishes to join or an operator in charge of improving the screening process.

[1152] 4. Emotion engine: A software module for recognizing the user's emotions in real time and transmitting them to the server.

[1153] Program processing overview

[1154] server

[1155] The server connects to a database to collect past review data. Specifically, it collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis. The server then analyzes the collected data and identifies bottlenecks in the review process. Furthermore, the server simulates a new review flow and predicts performance one and two years from now based on the results. A report is generated based on the prediction and simulation results. Emotion data sent from the emotion engine is also analyzed and reflected as feedback.

[1156] Terminal

[1157] The terminal provides an interface where users can input information about prospective subscribers, which is then transmitted to the server in real time. The terminal also provides generated reports to the users and provides an interface for inputting feedback.

[1158] User

[1159] The user inputs the information of the new subscriber through the terminal, checks the generated report, and also inputs feedback, which is sent to the server.

[1160] Emotion Engine

[1161] The emotion engine recognizes emotions in real time while the user is reviewing the report and sends the results to a server, such as Google Cloud Natural Language API or Microsoft Azure Text Analytics.

[1162] Hardware and software used

[1163] Hardware: Smartphones (iOS / Android), servers (AWS, GCP, Azure, etc.)

[1164] software:

[1165] Google Cloud Natural Language API or Microsoft Azure Text Analytics: Emotion Recognition Engine

[1166] PostgreSQL or MySQL: Database Management Systems

[1167] Specific examples

[1168] For example, when a prospective new member enters their information using a smartphone, that information is saved in real time on a server. The server then conducts a review based on that information and past data, simulating a new review flow. The results show that adopting a specific flow can shorten the review period. Furthermore, as the user reviews the report, an emotion engine recognizes emotions in real time, and emotions such as dissatisfaction or satisfaction are sent to the server. Based on this, further improvements to the flow are proposed.

[1169] Prompt Sentence Examples

[1170] "Simulate new review flows and evaluate their effectiveness. Identify areas for improvement based on past review data and use simulation results to generate predictive models. Also, recognize user sentiment data in real time and use it to suggest further improvements."

[1171] In this way, the present invention effectively optimizes the screening process for new subscribers and provides feedback that takes into account the user's emotional state, thereby shortening the screening period, improving customer satisfaction, and reducing operating costs.

[1172] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1173] Step 1:

[1174] The server collects past review data from a database. It collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) and performs preprocessing, including missing value imputation and data normalization. The input is the past review data in the database, and the output is data converted into a format that can be used for analysis.

[1175] Step 2:

[1176] The terminal provides a function for inputting information about new members through a user interface. When a user inputs information such as name, address, and application documents into the terminal, this information is sent to the server in real time. The input is the information about new members entered by the user, and the output is the information about new members sent to the server.

[1177] Step 3:

[1178] The server performs statistical analysis based on the collected historical data and information on new applicants. The analysis involves calculating the average and standard deviation of the review period, the frequency of each rejection reason, and the rate of application document defects. The input is the preprocessed historical data and information on new applicants, and the output is the analysis results.

[1179] Step 4:

[1180] The server identifies bottlenecks in the review process through data analysis, visualizes them, and sends them to the terminal. For example, it identifies that incomplete documents are the main cause of review delays. The input is the data analysis results, and the output is visualized bottleneck information.

[1181] Step 5:

[1182] The terminal provides an interface that allows the user to set up a new screening flow (for example, the introduction of a pre-document check process). When the user inputs the new settings, the terminal sends the setting information to the server. The input is the information of the new screening flow set by the user, and the output is the information of the new screening flow sent to the server.

[1183] Step 6:

[1184] The server runs a simulation based on the new review flow. It compares it with the conventional review flow and evaluates the efficiency and effectiveness of each (review period reduction rate, cost reduction effect). The simulation results are visualized and sent to the terminal. The input is information about the new review flow, and the output is the simulation results.

[1185] Step 7:

[1186] The server builds a predictive model based on the simulation results and past data. It uses a machine learning algorithm to predict the performance of the new screening process one and two years from now. The inputs are the simulation results and past data, and the output is the results of the predictive model.

[1187] Step 8:

[1188] The server automatically generates detailed reports based on the simulation and prediction results. The reports are formatted in a specific format (e.g., PDF) and sent to the terminal. The input is the result of the prediction model, and the output is the generated report.

[1189] Step 9:

[1190] The terminal displays the generated report to the user and provides an interface for inputting feedback, where the input is the generated report and the output is the user's feedback information.

[1191] Step 10:

[1192] The emotion engine recognizes the user's emotions in real time as they check the report. The recognition results are sent to the server for analysis. The input is the user's voice and facial expression data, and the output is the emotion recognition results.

[1193] Step 11:

[1194] The server analyzes the emotion data and proposes further improvements to the review process based on the user's emotional state. The inputs are emotion recognition results and user feedback information, and the output is improvement suggestions.

[1195] 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.

[1196] 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.

[1197] 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.

[1198] [Fourth embodiment]

[1199] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1200] 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.

[1201] 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).

[1202] 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.

[1203] 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.

[1204] 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).

[1205] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1206] 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.

[1207] 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.

[1208] 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.

[1209] 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.

[1210] 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.

[1211] 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."

[1212] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[1213] System Overview

[1214] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[1215] Program processing

[1216] 1. Data Collection

[1217] The server connects to the database and collects the past five years of review data, including application documents, review results, review periods, and reasons for rejection.

[1218] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[1219] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[1220] 2. Data Analysis

[1221] The server analyzes the collected past review data, performs statistical analysis, and calculates the average and standard deviation of the review period, the rate of defects in each document, and the frequency of rejection reasons.

[1222] The server provides the functionality to identify and visualize bottlenecks in the review process through data analysis.

[1223] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[1224] 3. Improvement Simulation

[1225] The terminal provides an interface that allows users to set up new screening flows (for example, introducing pre-document checks).

[1226] The server performs a comparative simulation of the old and new review flows and evaluates their efficiency and effectiveness.

[1227] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[1228] 4. Predictive model generation

[1229] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance of the new review process one and two years from now.

[1230] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[1231] 5. Generate and review reports

[1232] The server generates a report based on the simulation and prediction results, formats it in a specific format, and sends it to the terminal.

[1233] The terminal displays the generated report to the user for the user to review.

[1234] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[1235] In this way, the system of the present invention can effectively optimize the new customer screening process and predict future results, which can be expected to shorten screening times, improve customer satisfaction, and reduce operational costs.

[1236] The processing flow will be explained below.

[1237] Step 1:

[1238] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[1239] Step 2:

[1240] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[1241] Step 3:

[1242] The server performs statistical analysis based on the past review data collected, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects. Next, the analysis results are used to identify bottlenecks in the review process, and the visualized analysis results are sent to the terminal.

[1243] Step 4:

[1244] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[1245] Step 5:

[1246] The server runs a simulation based on the new audit flow that has been set up. It compares the old audit flow with the new one and calculates the efficiency and effectiveness of each (for example, the rate of reduction in audit time and cost reduction). The simulation results are visualized and sent to the terminal.

[1247] Step 6:

[1248] The server uses a machine learning algorithm to build a predictive model based on the simulation results and past data, and uses the built model to predict the performance (e.g., review efficiency, customer satisfaction) of the new screening process one and two years from now.

[1249] Step 7:

[1250] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[1251] Step 8:

[1252] The user can view the generated report via the terminal, which includes the simulation results and the output of the predictive model, providing the user with information to make decisions about future improvements to the review process.

[1253] Step 9:

[1254] Users enter feedback, which is sent from their devices to a server that records it and uses it to improve future simulations and predictive models.

[1255] Example 1

[1256] 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."

[1257] The traditional new customer application screening process involves a large number of documents and information, which can lead to processing delays and bottlenecks. It is also difficult to simulate and predict the effects of an improved screening process in advance. This makes it challenging to improve the efficiency of screening and customer satisfaction.

[1258] 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.

[1259] In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting that data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for providing an interface for a terminal to set a new screening flow, and means for the server to use a machine learning algorithm when analyzing the data. This improves the efficiency of the screening process and enables the design and implementation of a predictable screening flow.

[1260] "Means for collecting past audit data" refers to devices or programs that can collect information related to audits conducted in the past and store it on a server or database.

[1261] The "means for inputting information about new subscribers and collecting that data" refers to an interface and device for inputting personal information and application forms of users who wish to join, and sending them to the server for storage.

[1262] "Means for analyzing collected data and identifying bottlenecks in the review process" refers to analytical tools and algorithms used to analyze collected review data and identify causes of delays and inefficiencies during the review process.

[1263] The "means for setting up a new review flow and simulating it" is a system that provides an interface that allows users to set up a new review flow, and simulates that flow to evaluate its effectiveness.

[1264] "Means for predicting one-year and two-year performance of a configured review flow" refers to a system that runs machine learning algorithms and predictive models to predict future performance of a newly configured review flow.

[1265] The "means for generating a report based on the results of simulation and prediction" refers to a device or program that automatically generates a report based on the results of simulation and prediction, and puts it in a format that can be checked by the user.

[1266] The "means for checking the generated report" refers to an interface and device that displays the generated report to the user and enables the user to check the contents.

[1267] The "means for providing an interface for a terminal to set up a new screening flow" refers to a device and a program for providing an interface for a terminal to allow a user to input settings for a new screening flow.

[1268] "Means for using machine learning algorithms when the server analyzes data" refers to a system for using machine learning algorithms when the server analyzes collected data and for identifying bottlenecks in the review process.

[1269] As an embodiment of the present invention, specific processes performed by a server, a terminal, and a user will be described below.

[1270] System Overview

[1271] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows, resulting in a streamlined and efficient screening process.

[1272] Data collection and input

[1273] The server connects to the database and collects screening data from the past five years. This data includes application documents, screening results, screening periods, and reasons for rejection. The terminal provides a function for entering information about applicants through a user interface. The user enters the applicant's information into the terminal, and this information is sent to the server in real time.

[1274] Specific examples

[1275] The user enters Mr. Tanaka's information (name, address, application documents, etc.) into the terminal. This information is sent to the server and stored in the database.

[1276] Data analysis and bottleneck identification

[1277] The server analyzes the collected past review data. Statistical analysis is performed to calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. Through data analysis, the server identifies bottlenecks in the review process and visualizes them.

[1278] Specific examples

[1279] The server analyzes the review data using statistical analysis tools (e.g., Python's Pandas and NumPy). Using the R ggplot2 library, it identifies the main causes of review delays and generates a visualization graph showing that these are incomplete documents.

[1280] Improvement Simulation

[1281] The terminal provides an interface for setting up a new screening flow (for example, introducing a pre-document check). The user sets up the new screening flow, and the server performs a comparative simulation of the old and new screening flows to evaluate their efficiency and effectiveness.

[1282] Specific examples

[1283] The user configures the pre-document check using an HTML form on the terminal. The server runs a Monte Carlo simulation to compare the old and new flows. The new flow shows that the review period is reduced by 20%.

[1284] Predictive Model Generation

[1285] The server builds a predictive model based on the simulation results and past data, using machine learning algorithms such as SciKit-Learn and TensorFlow to predict how the new screening process will perform one and two years from now.

[1286] Specific examples

[1287] The server builds a predictive model and finds that one year later, review efficiency will improve by 15% and customer satisfaction will increase by 10 points.

[1288] Report generation and review

[1289] The server generates a report based on the simulation and prediction results, formats it in a specific format (e.g., PDF, Excel), and sends it to the terminal. The terminal displays the generated report to the user, who can then check the contents.

[1290] Specific examples

[1291] The server generates a PDF report using the ReportLab library and sends it to the terminal, which displays the report to the user using a PDF viewer.The user can then decide to officially implement the new review flow based on the contents of the report.

[1292] Prompt Sentence Examples

[1293] Here is an example of a prompt to use for a generative AI model:

[1294] 1. "This system provides the ability to analyze historical data and identify bottlenecks in the review process. Please explain how data collection and analysis are performed. Include specific tools and methods."

[1295] 2. Please explain in detail the process of comparing the simulation results of the new review flow with the conventional flow. Please also provide specific examples of improvements that result from this.

[1296] 3. "Please describe how the server generates a predictive model and uses it to assess the efficiency of future review processes and customer satisfaction. Please also describe the algorithms and tools used."

[1297] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1298] Step 1:

[1299] The server accesses the database and collects past review data. The server executes SQL queries to retrieve data such as application documents, review results, review periods, and rejection reasons from the past five years. The data is saved in CSV or JSON format and temporarily stored for later analysis. The input to this data collection process is the database connection information to the server, and the output is past review data.

[1300] Specific behavior:

[1301] The server uses an SQL query to extract past review data from the database and save it as a file. Example: SELECT FROM review_data WHERE year >= CURRENT_DATE - INTERVAL '5 years';

[1302] Step 2:

[1303] The terminal provides a user interface, and the user enters information about the new member (e.g., name, address, application form). The terminal converts this information into JSON format and sends it to the server. The input is the information about the new member entered by the user, and the output is JSON format data sent to the server.

[1304] Specific behavior:

[1305] When a user enters the information of the prospective subscriber into the form and presses the submit button, the terminal converts the information into JSON format and sends it to the server as an HTTP POST request.

[1306] Step 3:

[1307] The server saves the collected data of new applicants to a database. The server validates the data format and inserts the data into the appropriate fields. The input is the JSON data of new applicants sent from the terminal, and the output is the information of the applicants saved in the database.

[1308] Specific behavior:

[1309] The server parses the received JSON data and inserts the data into the corresponding table. Example: INSERT INTO applicant_data (name, address, documents) VALUES (json_data["name"], json_data["address"], json_data["documents"]);

[1310] Step 4:

[1311] The server analyzes past review data. It uses statistical analysis tools (e.g., Python's Pandas and NumPy) to analyze the data and calculate the average and standard deviation of the review period, the rate of defects for each document, and the frequency of rejection reasons. The input is past review data, and the output is the analysis results.

[1312] Specific behavior:

[1313] The server runs a Python script, creates a data frame using the Pandas library, and performs statistical analysis. For example: import pandas as pd; data = pd.read_csv('review_data.csv'); mean_duration = data['duration'].mean();

[1314] Step 5:

[1315] The server identifies and visualizes bottlenecks in the review process based on the analysis results. It generates graphs using the R ggplot2 library to identify specific bottlenecks. The input is the analysis results, and the output is the visualized graph and the identified bottlenecks.

[1316] Specific behavior:

[1317] The server runs the R script and generates the graph using the ggplot2 library. For example: library(ggplot2); ggplot(data, aes(x=duration, y=freq)) + geom_bar(stat='identity');

[1318] Step 6:

[1319] The terminal provides an interface for setting up a new screening flow (e.g., introducing a pre-document check). The user sets up a new screening flow, which is then sent to the server. The input is the information on the new screening flow set up by the user, and the output is the flow setting data sent to the server.

[1320] Specific behavior:

[1321] When a user enters a new review flow into the form and presses the submit button, the terminal sends the content to the server.

[1322] Step 7:

[1323] The server runs a simulation comparing the old and new screening flows to evaluate their efficiency and effectiveness. The evaluation is performed using Monte Carlo simulation, etc. The input is information about the old and new screening flows, and the output is the simulation results for each flow.

[1324] Specific behavior:

[1325] The server runs the simulation program and evaluates the results. Example: monte_carlo_simulation(old_flow, new_flow);

[1326] Step 8:

[1327] The server builds a predictive model based on the simulation results and past data. It generates the predictive model using machine learning algorithms such as SciKit-Learn and TensorFlow. The input is the simulation results and past review data, and the output is the generated predictive model.

[1328] Specific behavior:

[1329] The server builds a predictive model using machine learning libraries. Example: from sklearn.linear_model import LinearRegression; model = LinearRegression().fit(X, y);

[1330] Step 9:

[1331] The server generates reports based on the simulation and prediction results and formats them in a specific format (e.g. PDF, Excel). The report is created using libraries such as the ReportLab. The input is the prediction model and the simulation results, and the output is the generated report.

[1332] Specific behavior:

[1333] The server generates a PDF report using ReportLab. Example: from reportlab.lib.pagesizes import letter; from reportlab.pdfgen import canvas; c = canvas.Canvas("output.pdf", pagesize=letter); c.drawString(100, 750, "Simulation and Prediction Report"); c.save();

[1334] Step 10:

[1335] The terminal displays the generated report to the user, allowing the user to confirm the contents. The input is the report sent from the server, and the output is the report confirmed by the user.

[1336] Specific behavior:

[1337] The device opens a PDF viewer and displays the report to the user. Example: The device displays the report using the PDF viewer in the browser.

[1338] (Application example 1)

[1339] 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."

[1340] In the conventional screening process, it is difficult to efficiently identify bottlenecks and improve the screening flow using past data.In addition, there is a lack of means to optimize the screening process for new applicants using mobile devices such as smartphones, so it is difficult to sufficiently shorten the screening time or reduce false positives.

[1341] 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.

[1342] In this invention, the server includes a means for collecting past screening data, a means for inputting information on new applicants and collecting that data, and a means for analyzing the collected data and identifying bottlenecks in the screening process. This enables efficient optimization and improvement of the screening process. Furthermore, by including a means for providing an application that is installed on a smartphone and optimizes the screening process for new applicants, screening time can be shortened and customer satisfaction can be improved.

[1343] "Past review data" refers to information related to the review process, such as past application documents, review results, review periods, and reasons for rejection.

[1344] "Information of new applicants" refers to personal information such as the name, address, and application documents of applicants.

[1345] A "bottleneck" refers to a factor or obstruction that delays progress in the review process.

[1346] "Review flow" refers to the series of steps and processes for reviewing an applicant's information and arriving at a result.

[1347] "Simulation" refers to a virtual experiment or trial conducted based on set conditions and flow.

[1348] A "smartphone" refers to a mobile device that has multifunctional data communication capabilities in addition to mobile phone functions.

[1349] An "application" refers to a software program that is installed on a device such as a smartphone and provides specific functions.

[1350] As an embodiment of the present invention, the following system is embodied.

[1351] 1. System Overview

[1352] This system collects past screening data and inputs and collects information on new applicants to efficiently identify bottlenecks in the screening process and simulate and predict new screening flows. In particular, it aims to optimize the screening process through an application installed on smartphones.

[1353] 2. Hardware and Software Used

[1354] Hardware: smartphones, tablets, servers

[1355] software:

[1356] Python and pandas are used for data collection and analysis.

[1357] For simulation and prediction, we use the random forest algorithm from scikit-learn.

[1358] Data visualization will be done using matplotlib and seaborn.

[1359] Python is used to generate reports and output them in PDF format, etc.

[1360] 3. Data processing and calculation flow

[1361] The server retrieves past screening data from the database and analyzes it to identify bottlenecks in the screening process. For example, it can identify that a specific document deficiency is the main cause of screening delays. The device (smartphone or tablet) provides an interface for entering information about new applicants and has the function of sending the collected information to the server.

[1362] The server has the function of setting up a new review flow (e.g., introducing a pre-document check) and runs a simulation to evaluate its effectiveness. This can result in results such as a shortened review period and improved customer satisfaction. The server generates a report based on the simulation results and prediction model and sends it to the terminal in a specific format such as PDF. The user can check the generated report through the terminal and evaluate its effectiveness.

[1363] 4. Specific examples and prompts for generative AI models

[1364] As a concrete example, the following simulation results can be obtained. For example, if a new review flow is introduced, it is predicted that the review period will be shortened by 20% and review efficiency will improve by 15% after one year. These results are provided to the user as a PDF report.

[1365] An example of a prompt to be input to the generative AI model is as follows:

[1366] Please provide a detailed explanation of the correlation between the rate of document deficiencies and the review period based on the review data from the past five years.

[1367] "Please predict the performance improvement if we implement a new review flow."

[1368] In this way, the system of the present invention effectively optimizes the screening process for new subscribers, shortening the screening period and improving customer satisfaction.

[1369] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1370] Step 1:

[1371] The server collects past review data from the database. Specifically, it connects to the review database and extracts information such as application documents, review results, review periods, and rejection reasons over the past five years. The input to this data collection process is the database connection information, and the output is the collected data in a data frame format.

[1372] Step 2:

[1373] The terminal provides an interface for entering information about prospective new subscribers. The user uses the terminal interface to enter their name, address, and application documents. The entered information is sent in real time to the server, where it is stored in a database. The input to this process is the information entered by the user, and the output is the new subscriber's information stored in the server database.

[1374] Step 3:

[1375] The server analyzes the collected past data. As a specific example of the analysis, the server performs statistical analysis to calculate the average and standard deviation of the review period, the defect rate of each document, and the frequency of rejection reasons. The input of this step is the collected past review data, and the output is a report containing the analysis results.

[1376] Step 4:

[1377] Based on the results of the data analysis, the server identifies bottlenecks in the review process and visualizes them. For example, the server may identify that incomplete documents are the main cause of review delays and generate the results in a visualized format such as a graph. The input to this process is the analysis results, and the output is visualized data on the bottlenecks.

[1378] Step 5:

[1379] The terminal provides an interface for setting up a new screening flow. The user sets up a new flow, such as introducing a pre-document check, through this interface. The set information is sent to the server. The input of this step is the information for the new screening flow set up by the user, and the output is the data for the new flow setting sent to the server.

[1380] Step 6:

[1381] The server simulates the old and new review flows and evaluates their efficiency and effectiveness. Specifically, the server uses the collected data and new flow information to run machine learning algorithms such as RandomForestClassifier to simulate the new flow. The inputs to this step are past data and new flow information, and the output is performance evaluation data as the simulation results.

[1382] Step 7:

[1383] The server builds a predictive model based on the simulation results and past data. For example, it uses a machine learning algorithm to predict how a new screening process will perform one and two years from now. The inputs to this process are the simulation results and past data, and the output is the results of the predictive model.

[1384] Step 8:

[1385] The server generates a report based on the prediction results and sends it to the terminal. The report is formatted in a specific format (e.g., PDF) and includes visualization data. The user can view the generated report through the terminal and evaluate the effectiveness. The input of this step is the results from the predictive model, and the output is the generated report.

[1386] The above steps realize a concrete system for optimizing the screening process for new subscribers.

[1387] 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.

[1388] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[1389] System Overview

[1390] The system optimizes the screening process for new subscribers and simulates the effects and impact of improvements to screening standards and procedures. In particular, it uses past data to identify bottlenecks in the screening process and simulates and predicts the effects of new screening flows. Furthermore, by combining it with an emotion engine that recognizes user emotions, the system can further improve the system based on emotion analysis of user feedback.

[1391] Program processing

[1392] 1. Data Collection

[1393] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[1394] The terminal provides a function to input information about the person who wants to join through a user interface. When a user inputs information about the person who wants to join (for example, Mr. Tanaka) into the terminal, this information is sent to the server in real time.

[1395] Example: Mr. Tanaka wishes to sign up as a new member and enters his name, address, and application documents into the terminal. This information is sent to the server and stored in the database.

[1396] 2. Data Analysis

[1397] The server performs statistical analysis based on the collected past review data, calculating the average and standard deviation of review times, the frequency of each rejection reason, and the rate of application document defects.

[1398] The server identifies bottlenecks in the review process through data analysis and transmits them to the terminal in a visualized form.

[1399] Example: The server performs an analysis and identifies that in past reviews, 10% of cases had incomplete documentation, which is the main cause of review delays.

[1400] 3. Improvement Simulation

[1401] The terminal provides an interface that allows users to set up a new screening flow (for example, introducing a pre-document check process). When the user inputs new settings, the terminal sends the setting information to the server.

[1402] The server runs a simulation based on the new screening flow. It compares the old and new screening flows and evaluates their efficiency and effectiveness (reduced screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[1403] Example: A user configures the introduction of a pre-document check, and the server runs a simulation, showing that the new flow reduces the review period by 20%.

[1404] 4. Predictive model generation

[1405] The server builds a predictive model based on the simulation results and past data, and uses a machine learning algorithm to predict the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[1406] Example: The server builds a predictive model that predicts that review efficiency will improve by 15% and customer satisfaction will increase by 10 points one year from now.

[1407] 5. Generate and review reports

[1408] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[1409] The terminal displays the generated report to the user and provides an interface for the user to review the report and enter feedback.

[1410] Example: The server generates a PDF report and sends it to the terminal, where the user can review it. Based on the report, the user can decide to officially implement a new review flow.

[1411] 6. Emotion Recognition by Emotion Engine

[1412] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[1413] Example: While a user is viewing a report, the emotion engine recognizes the user's emotions (e.g., satisfaction, dissatisfaction, excitement, disappointment) via a webcam or microphone and transmits them to the server in real time.

[1414] 7. Emotional Data Analysis and Feedback

[1415] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[1416] Example: The server analyzes sentiment data and user feedback to identify, for example, which specific process steps are causing user frustration. Based on this, further flow improvements are suggested.

[1417] In this way, the system of the present invention not only effectively optimizes the new customer screening process, but also takes into account the user's emotional state to suggest further improvements, thereby further shortening the screening period, improving customer satisfaction, and reducing operational costs.

[1418] The processing flow will be explained below.

[1419] Step 1:

[1420] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis.

[1421] Step 2:

[1422] The user uses a terminal to enter information about the new subscriber (name, address, desired plan, attachment of required documents, etc.) into an online form. The terminal then transmits the entered information to the server in real time.

[1423] Step 3:

[1424] Statistical analysis is performed based on past review data collected by the server, calculating the average and standard deviation of review time, the frequency of each rejection reason, and the rate of application document defects.

[1425] Step 4:

[1426] The server uses the analysis results to identify bottlenecks in the review process and transmits the visualized analysis results to the terminal, using graphs and heat maps.

[1427] Step 5:

[1428] The user sets up a new screening process (for example, introducing a pre-document check process) using the terminal. The terminal sends the setting information to the server.

[1429] Step 6:

[1430] The server runs a simulation based on the new screening flow that has been set up. The old screening flow is compared with the new one, and the efficiency and effectiveness of each (review period reduction rate, cost reduction effect) are evaluated. The simulation results are visualized in graphs and tables and sent to the terminal.

[1431] Step 7:

[1432] The server builds a predictive model based on the simulation results and past data, and uses machine learning algorithms (such as linear regression or random forest) to predict performance (screening efficiency, customer satisfaction) one and two years into the future.

[1433] Step 8:

[1434] The server automatically generates a detailed report based on the simulation and prediction results, which is then formatted into a specific format (e.g., PDF) and sent to the terminal.

[1435] Step 9:

[1436] The user checks the generated report through the terminal, evaluates the proposed improvements to the review process based on the report contents, and makes a final decision.

[1437] Step 10:

[1438] The emotion engine recognizes the user's emotions in real time as they review the report, and the results of the emotion recognition are sent to the server for analysis.

[1439] Step 11:

[1440] The server analyzes the emotional data from the emotion engine and suggests further improvements to the review process based on the user's emotional state. The server comprehensively evaluates the feedback entered by the user and the results of the emotional data analysis to help improve the process in the future.

[1441] Example 2

[1442] 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."

[1443] In the conventional review process, bottlenecks were identified by analyzing past data and the effectiveness of new flows was simulated, but there were limitations to improving the efficiency of the overall review process or customer satisfaction. Furthermore, it was not possible to reflect user sentiment in real time, and the quality of feedback was insufficient. As a result, the reduction of review time and operational costs was not fully achieved.

[1444] 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.

[1445] A means of collecting past audit data;

[1446] A means for inputting information of new subscribers and collecting the data;

[1447] A means of analyzing the collected data and identifying bottlenecks in the review process;

[1448] A means to set up a new review flow and simulate it,

[1449] A means of predicting performance one and two years into a set review flow;

[1450] a means for generating reports based on the simulation and prediction results;

[1451] A means of reviewing the generated report;

[1452] A means for recognizing user emotions in real time and collecting the results;

[1453] A means to analyze the collected sentiment data and suggest further improvements to the review process; and

[1454] This will make the screening process more efficient, improve customer satisfaction, and reduce operational costs.

[1455] "Review Data" refers to all data related to the review process, including past application documents, review results, review periods, and reasons for rejection.

[1456] "Information of prospective subscribers" refers to all information provided by users who wish to sign up, such as the name, address, and application documents of users who wish to sign up.

[1457] A "bottleneck" refers to an item or step in the review process that tends to slow down processing and reduce efficiency.

[1458] "Review flow" refers to the order and method of each step and procedure in the review process.

[1459] "Simulation" refers to a computer-based analytical method used to virtually implement a new review process and predict its effects.

[1460] "Performance prediction" refers to using machine learning algorithms to predict review efficiency and customer satisfaction one and two years after the introduction of a new review process.

[1461] "Report" refers to a written report summarizing the simulation and prediction results.

[1462] "Emotion recognition" refers to a technology that analyzes a user's facial expressions and voice to identify the user's emotional state in real time.

[1463] "Emotion Data" refers to data regarding a user's emotional state collected through emotion recognition technology.

[1464] "Recommendations for improving the review process" refers to specific changes and improvements proposed to make the review process more efficient, based on analysis of emotional data and user feedback.

[1465] To implement this invention, a system is used that combines a terminal for collecting past screening data and inputting information on new applicants, a server for analyzing the data and setting up and simulating new screening flows, and an emotion engine that recognizes user emotions in real time.

[1466] System Configuration

[1467] This system consists of multiple pieces of hardware and software, as follows:

[1468] 1. Server: Performs large-volume data processing, analysis, and simulation.

[1469] 2. Client Device: Provides an interface for users to enter information about new subscribers and confirm the data.

[1470] 3. Emotion Recognition Engine: A module that analyzes the user's facial expressions and voice to recognize emotions.

[1471] Hardware and Software

[1472] The servers are equipped with high-performance CPUs and GPUs, and feature a comprehensive database management system (e.g., MySQL, PostgreSQL). Data analysis uses Python and related packages (e.g., pandas, numpy, scikit-learn).

[1473] The terminals used are PCs and tablets, which allow users to access the system through a web browser. Web applications are built using HTML, CSS, and JavaScript (e.g., React.js).

[1474] The emotion engine uses input devices such as cameras and microphones, as well as software (e.g., OpenCV, TensorFlow) that controls and analyzes them.

[1475] Program processing flow

[1476] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). This data is then converted into a format that can be used for analysis after missing values ​​are imputed and data normalization is performed.

[1477] The terminal provides a function to input information of new subscribers through a user interface. When a user inputs information of a subscriber, the information is sent to the server in real time and stored in a database.

[1478] The server performs statistical analysis of the collected audit data to identify bottlenecks. The identified bottlenecks are visualized and sent to the terminal. The server then simulates the new audit flow and compares it with the conventional flow to evaluate its efficiency and effectiveness.

[1479] A predictive model is built based on the simulation results and past data, and performance one and two years after the new flow is implemented is predicted. Finally, a report is generated based on the simulation and prediction results and sent to the terminal in PDF format or other format.

[1480] While users are reviewing the report, the emotion engine recognizes their emotions in real time and sends the data to the server, which analyzes the emotion data and provides user feedback to suggest further improvements to the review process.

[1481] Specific examples

[1482] For example, when a user enters information about a prospective new member into a terminal (such as name, address, and application documents), this information is sent to the server and stored in a database. The server analyzes this data and, based on past screening data, identifies incomplete documents as a major bottleneck. The user sets up a new screening flow from the terminal (for example, introducing a pre-document check process), and the server simulates it, showing that the screening period can be reduced by 20%.

[1483] Example prompts from a generative AI model:

[1484] "Please enter the information of the person wishing to sign up. We will propose the optimal screening process based on the analysis of screening data from the past five years."

[1485] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1486] Step 1: Data collection

[1487] The server connects to the database and collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is then filled in with missing values, normalized, and converted into a format that can be used for analysis. Specifically, SQL queries are executed to extract data from the database and preprocessed using Python.

[1488] Input: Database connection information and SQL query

[1489] Output: Preprocessed audit dataset

[1490] Step 2: Enter information about new members

[1491] The terminal provides a function that allows users to input information about prospective subscribers through a user interface. When a user inputs information such as the prospective subscriber's name, address, and application documents into the terminal, this information is sent to the server in real time. The terminal is built using a web application (e.g., React.js), and the input information is sent to the server via an AJAX request.

[1492] Input: Subscriber information entered by the user into the terminal

[1493] Output: Subscriber information sent to the server

[1494] Step 3: Data analysis

[1495] Statistical analysis is performed based on past review data collected by the server. Specifically, data frames are manipulated using Python's pandas and numpy to calculate the average and standard deviation of review periods, the frequency of each reason for rejection, and the rate of incomplete application documents.

[1496] Input: Preprocessed audit dataset

[1497] Output: Calculated statistics (average review period, frequency of rejection reasons, etc.)

[1498] Step 4: Identify bottlenecks

[1499] The server analyzes the data to identify bottlenecks in the review process. Based on this statistical data, the server identifies specific items (e.g., incomplete documents) or specific process steps (e.g., review delays). The results are visualized in graphs and charts and sent to the terminal.

[1500] Input: Calculated statistic value

[1501] Output: Visualized data of bottleneck identification results

[1502] Step 5: Setting up and simulating a new review flow

[1503] The terminal provides the user with an interface that allows them to set up a new screening flow. The user sets up a new flow (for example, the introduction of a pre-document check process), and the information is sent to the server. Based on this information, the server simulates the old and new screening flows and evaluates the efficiency and effectiveness of each flow (reduction in screening time, cost reduction). The simulation results are visualized and sent to the terminal.

[1504] Input: Information about the new screening flow configured by the user

[1505] Output: Visualized data of the simulation results of the old and new screening flows

[1506] Step 6: Generate a predictive model

[1507] The server generates a predictive model based on the simulation results and past data. Specifically, it uses a Python machine learning package (e.g., scikit-learn) to build a model that predicts the performance (screening efficiency, customer satisfaction) of the new screening process one and two years from now.

[1508] Input: Simulation results and past audit data

[1509] Output: Prediction model and results (prediction performance after 1 year and 2 years)

[1510] Step 7: Generate and review the report

[1511] The server automatically generates a detailed report based on the simulation and prediction results. The report is formatted in PDF format or similar and sent to the terminal. The terminal displays the generated report to the user and provides an interface where the user can review the report and enter feedback.

[1512] Input: Prediction model and its results

[1513] Output: Report in PDF format

[1514] Step 8: Emotion Recognition with the Emotion Engine

[1515] The emotion engine recognizes the user's emotions in real time as they check the report. Specifically, it analyzes data collected through the camera and microphone using OpenCV and TensorFlow to generate emotion data, which is then sent to the server.

[1516] Input: Real-time data collected from cameras and microphones

[1517] Output: Emotion data

[1518] Step 9: Emotional Data Analysis and Feedback

[1519] The server analyzes the emotion data sent from the emotion engine, evaluates it comprehensively with user feedback, and proposes further improvements to the review process. For analysis, Python analysis libraries (e.g., pandas, numpy) are used.

[1520] Input: Emotion data and user feedback

[1521] Output: Improvement proposal data and feedback report

[1522] Through the above processing steps, it is possible to improve the efficiency of the new subscriber screening process and customer satisfaction.

[1523] (Application example 2)

[1524] 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."

[1525] In the conventional new subscriber screening process, collecting and analyzing application data and improving the screening flow took time, resulting in problems that reduced user satisfaction. Furthermore, bottlenecks in the screening process were often not clearly identified, preventing effective improvements. Furthermore, the screening process proceeded without taking into consideration the user's feelings, potentially increasing user dissatisfaction and stress. A system that solves these problems and realizes an efficient screening process with high user satisfaction is needed.

[1526] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting past screening data, means for inputting information on new applicants and collecting the data, means for analyzing the collected data and identifying bottlenecks in the screening process, means for setting a new screening flow and simulating it, means for predicting performance one year and two years from now using the set screening flow, means for generating a report based on the simulation and prediction results, means for checking the generated report, means for recognizing user emotions in real time using an emotion recognition engine, and means for providing feedback based on the recognized emotion data. This enables an efficient and accurate screening process for new applicants and the provision of feedback that takes user emotions into consideration.

[1527] "Past screening data" refers to historical information such as information on the screening process for past new applicants, application documents, screening results, screening period, and reasons for rejection.

[1528] "Information on new subscribers" refers to personal information such as the name, address, and application documents of users who wish to sign up for a new membership, as well as the contents of that application.

[1529] A "bottleneck" refers to a major obstacle or problem that causes delays or inefficiencies in the new member screening process.

[1530] "New review flow" refers to the newly established review procedures and processes to improve, streamline and optimize the existing review flow.

[1531] "Simulation" refers to a method for virtually simulating a new review flow and evaluating its effectiveness and efficiency.

[1532] "Performance" refers to the efficiency resulting from the operation of the established review flow, the rate at which the review period is shortened, the cost reduction effect, and customer satisfaction, etc.

[1533] "Report" refers to a detailed report prepared based on the simulation and prediction results.

[1534] An "emotion recognition engine" refers to a software module that identifies and analyzes emotions from a user's facial expressions and voice in real time.

[1535] "Feedback" refers to information and advice to suggest further improvements and modifications based on user sentiment and reactions to the actual review process.

[1536] "Data analysis" refers to a method of statistically analyzing collected past screening data and information on new applicants to identify problems in the screening process.

[1537] "Visualization" refers to a method of representing analysis and simulation results in a visual format such as a graph or chart.

[1538] "Format" refers to the rules and style for organizing and formatting reports and data according to a particular format or standard.

[1539] As an embodiment of the present invention, specific processing by a server, a terminal, a user, and an emotion engine will be described below.

[1540] The system configuration is as follows:

[1541] 1. Server: This is the central component for data collection, analysis, simulation, prediction, report generation, and sentiment data analysis.

[1542] 2. Terminal: Provides an interface for new subscribers to enter information, check reports, and enter feedback.

[1543] 3. User: A person who wishes to join or an operator in charge of improving the screening process.

[1544] 4. Emotion engine: A software module for recognizing the user's emotions in real time and transmitting them to the server.

[1545] Program processing overview

[1546] server

[1547] The server connects to a database to collect past review data. Specifically, it collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.). The collected data is preprocessed (missing value completion, data normalization, etc.) and converted into a format that can be used for analysis. The server then analyzes the collected data and identifies bottlenecks in the review process. Furthermore, the server simulates a new review flow and predicts performance one and two years from now based on the results. A report is generated based on the prediction and simulation results. Emotion data sent from the emotion engine is also analyzed and reflected as feedback.

[1548] Terminal

[1549] The terminal provides an interface where users can input information about prospective subscribers, which is then transmitted to the server in real time. The terminal also provides generated reports to the users and provides an interface for inputting feedback.

[1550] User

[1551] The user inputs the information of the new subscriber through the terminal, checks the generated report, and also inputs feedback, which is sent to the server.

[1552] Emotion Engine

[1553] The emotion engine recognizes emotions in real time while the user is reviewing the report and sends the results to a server, such as Google Cloud Natural Language API or Microsoft Azure Text Analytics.

[1554] Hardware and software used

[1555] Hardware: Smartphones (iOS / Android), servers (AWS, GCP, Azure, etc.)

[1556] software:

[1557] Google Cloud Natural Language API or Microsoft Azure Text Analytics: Emotion Recognition Engine

[1558] PostgreSQL or MySQL: Database Management Systems

[1559] Specific examples

[1560] For example, when a prospective new member enters their information using a smartphone, that information is saved in real time on a server. The server then conducts a review based on that information and past data, simulating a new review flow. The results show that adopting a specific flow can shorten the review period. Furthermore, as the user reviews the report, an emotion engine recognizes emotions in real time, and emotions such as dissatisfaction or satisfaction are sent to the server. Based on this, further improvements to the flow are proposed.

[1561] Prompt Sentence Examples

[1562] "Simulate new review flows and evaluate their effectiveness. Identify areas for improvement based on past review data and use simulation results to generate predictive models. Also, recognize user sentiment data in real time and use it to suggest further improvements."

[1563] In this way, the present invention effectively optimizes the screening process for new subscribers and provides feedback that takes into account the user's emotional state, thereby shortening the screening period, improving customer satisfaction, and reducing operating costs.

[1564] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1565] Step 1:

[1566] The server collects past review data from a database. It collects all review data from the past five years (application documents, review results, review period, reasons for rejection, etc.) and performs preprocessing, including missing value imputation and data normalization. The input is the past review data in the database, and the output is data converted into a format that can be used for analysis.

[1567] Step 2:

[1568] The terminal provides a function for inputting information about new members through a user interface. When a user inputs information such as name, address, and application documents into the terminal, this information is sent to the server in real time. The input is the information about new members entered by the user, and the output is the information about new members sent to the server.

[1569] Step 3:

[1570] The server performs statistical analysis based on the collected historical data and information on new applicants. The analysis involves calculating the average and standard deviation of the review period, the frequency of each rejection reason, and the rate of application document defects. The input is the preprocessed historical data and information on new applicants, and the output is the analysis results.

[1571] Step 4:

[1572] The server identifies bottlenecks in the review process through data analysis, visualizes them, and sends them to the terminal. For example, it identifies that incomplete documents are the main cause of review delays. The input is the data analysis results, and the output is visualized bottleneck information.

[1573] Step 5:

[1574] The terminal provides an interface that allows the user to set up a new screening flow (for example, the introduction of a pre-document check process). When the user inputs the new settings, the terminal sends the setting information to the server. The input is the information of the new screening flow set by the user, and the output is the information of the new screening flow sent to the server.

[1575] Step 6:

[1576] The server runs a simulation based on the new review flow. It compares it with the conventional review flow and evaluates the efficiency and effectiveness of each (review period reduction rate, cost reduction effect). The simulation results are visualized and sent to the terminal. The input is information about the new review flow, and the output is the simulation results.

[1577] Step 7:

[1578] The server builds a predictive model based on the simulation results and past data. It uses a machine learning algorithm to predict the performance of the new screening process one and two years from now. The inputs are the simulation results and past data, and the output is the results of the predictive model.

[1579] Step 8:

[1580] The server automatically generates detailed reports based on the simulation and prediction results. The reports are formatted in a specific format (e.g., PDF) and sent to the terminal. The input is the result of the prediction model, and the output is the generated report.

[1581] Step 9:

[1582] The terminal displays the generated report to the user and provides an interface for inputting feedback, where the input is the generated report and the output is the user's feedback information.

[1583] Step 10:

[1584] The emotion engine recognizes the user's emotions in real time as they check the report. The recognition results are sent to the server for analysis. The input is the user's voice and facial expression data, and the output is the emotion recognition results.

[1585] Step 11:

[1586] The server analyzes the emotion data and proposes further improvements to the review process based on the user's emotional state. The inputs are emotion recognition results and user feedback information, and the output is improvement suggestions.

[1587] 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.

[1588] 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.

[1589] 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.

[1590] 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.

[1591] FIG. 9 illustrates 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 behaviors 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.

[1592] 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.

[1593] 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).

[1594] 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.

[1595] 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."

[1596] 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.

[1597] 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).

[1598] 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.

[1599] 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.

[1600] 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.

[1601] 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.

[1602] 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.

[1603] 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.

[1604] 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.

[1605] 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.

[1606] 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.

[1607] 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.

[1608] The following is further disclosed regarding the above embodiment.

[1609] (Claim 1)

[1610] A means of collecting past audit data;

[1611] A means for inputting information of new subscribers and collecting the data;

[1612] A means of analyzing the collected data and identifying bottlenecks in the review process;

[1613] A means to set up a new review flow and simulate it,

[1614] A means of predicting performance one and two years into a set review flow;

[1615] a means for generating reports based on the simulation and prediction results;

[1616] A means of reviewing the generated report;

[1617] A system including:

[1618] (Claim 2)

[1619] 10. The system of claim 1, further comprising means for visualizing bottlenecks in the review process based on the analyzed data.

[1620] (Claim 3)

[1621] 10. The system of claim 1, further comprising means for formatting and transmitting the report in a specific format.

[1622] "Example 1"

[1623] (Claim 1)

[1624] A means of collecting past audit data;

[1625] A means for inputting information of new subscribers and collecting the data;

[1626] A means of analyzing the collected data and identifying bottlenecks in the review process;

[1627] A means to set up a new review flow and simulate it,

[1628] A means of predicting performance one and two years into a set review flow;

[1629] a means for generating reports based on the simulation and prediction results;

[1630] A means of reviewing the generated report;

[1631] A means for providing an interface for the terminal to set up a new screening flow;

[1632] a means for the server to use machine learning algorithms in analyzing the data;

[1633] A system including:

[1634] (Claim 2)

[1635] 10. The system of claim 1, further comprising means for visualizing bottlenecks in the review process based on the analyzed data.

[1636] (Claim 3)

[1637] 10. The system of claim 1, further comprising means for formatting and transmitting the report in a specific format.

[1638] "Application Example 1"

[1639] (Claim 1)

[1640] A means of collecting past audit data;

[1641] A means for inputting information of new subscribers and collecting the data;

[1642] A means of analyzing the collected data and identifying bottlenecks in the review process;

[1643] A means to set up a new review flow and simulate it,

[1644] A means of predicting performance one and two years into a set review flow;

[1645] a means for generating reports based on the simulation and prediction results;

[1646] A means of reviewing the generated report;

[1647] means of providing an application that can be installed on a smartphone to optimize the screening process for new subscribers;

[1648] A system including:

[1649] (Claim 2)

[1650] 10. The system of claim 1, further comprising means for visualizing bottlenecks in the review process based on the analyzed data.

[1651] (Claim 3)

[1652] 10. The system of claim 1, further comprising means for formatting and transmitting the report in a specific format.

[1653] "Example 2: Combining Emotion Engines"

[1654] (Claim 1)

[1655] A means of collecting past audit data;

[1656] A means for inputting information of new subscribers and collecting the data;

[1657] A means of analyzing the collected data and identifying bottlenecks in the review process;

[1658] A means to set up a new review flow and simulate it,

[1659] A means of predicting performance one and two years into a set review flow;

[1660] a means for generating reports based on the simulation and prediction results;

[1661] A means of reviewing the generated report;

[1662] A means for recognizing user emotions in real time and collecting the results;

[1663] A means to analyze the collected sentiment data and suggest further improvements to the review process; and

[1664] A system including:

[1665] (Claim 2)

[1666] 10. The system of claim 1, further comprising means for visualizing bottlenecks in the review process based on the analyzed data.

[1667] (Claim 3)

[1668] 10. The system of claim 1, further comprising means for formatting and transmitting the report in a specific format.

[1669] "Application example 2 when combining emotion engines"

[1670] (Claim 1)

[1671] A means of collecting past audit data;

[1672] A means for inputting information of new subscribers and collecting the data;

[1673] A means of analyzing the collected data and identifying bottlenecks in the review process;

[1674] A means to set up a new review flow and simulate it,

[1675] A means of predicting performance one and two years into a set review flow;

[1676] a means for generating reports based on the simulation and prediction results;

[1677] A means of reviewing the generated report;

[1678] means for recognizing a user's emotions in real time using an emotion recognition engine;

[1679] means for providing feedback based on the recognized emotion data;

[1680] A system including:

[1681] (Claim 2)

[1682] A means of visualizing bottlenecks in the review process based on the analyzed data;

[1683] It also includes a means for analyzing emotion data and, based on that, proposing ideas for further improving the review process.

[1684] 10. The system of claim 1.

[1685] (Claim 3)

[1686] and means for formatting and transmitting the report in a specific format.

[1687] 10. The system of claim 1. [Explanation of symbols]

[1688] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting past audit data; A means for inputting information of new subscribers and collecting the data; A means of analyzing the collected data and identifying bottlenecks in the review process; A means to set up a new review flow and simulate it, A means of predicting performance one and two years into a set review flow; a means for generating reports based on the simulation and prediction results; A means of reviewing the generated report; A system including:

2. The system of claim 1 , further comprising means for visualizing bottlenecks in the screening process based on the analyzed data.

3. 10. The system of claim 1, further comprising means for formatting and transmitting said report in a particular format.

Images