system

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting financial information such as an account balance, a borrowing status, and a repayment result of a company from a bank; means for collecting industry information such as an industry production volume, a market share, and a ranking; means for collecting word-of-mouth information and evaluation information on the company from various SNSs; means for integrating the financial information, the industry information, and the SNS information and performing internal credit and rating of the company using a machine learning model; and means for providing an evaluation result and a trend prediction using the machine learning model to the company.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In order to develop an appropriate financing strategy, a company needs to comprehensively analyze a wide range of data, including credit ratings from financial institutions, industry trends, and word-of-mouth information on social media. However, collecting this information individually and evaluating it on one's own is extremely time-consuming and requires specialized knowledge. Furthermore, financial institutions need to use up-to-date, multifaceted information to accurately assess a company's creditworthiness and make appropriate lending decisions, but currently they tend to rely on the data they possess. As a result, companies end up with insufficient or excessive financing, and financial institutions face undesirable loan default risks. A solution to these issues is needed. [Means for solving the problem]

[0005] The present invention is a system that includes a means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks, a means for collecting industry information such as industry production volume, market share, and rankings, and a means for collecting word-of-mouth and evaluation information about the company from various social media platforms. The system integrates this information and uses a machine learning model to perform internal credit assessment and rating on the company. Specifically, the system preprocesses the financial information, industry information, and social media information into an integrated dataset, and inputs this integrated dataset into a machine learning model to calculate a company's credit score and assign a rating. Furthermore, the machine learning model is periodically retrained with new data, so that the evaluation results and trend forecasts are always kept up to date. In this way, the present invention makes it easier for companies to predict future funding needs and interest rate trends, and supports more accurate lending decisions for financial institutions, thereby achieving mutually beneficial fundraising and supply.

[0006] "Financial information" refers to information held by the bank relating to the financial status of a company, such as the company's account balance, borrowing status, and repayment history.

[0007] "Industry information" is data that shows market trends, such as production volume, market share, and rankings for a specific industry.

[0008] "SNS information" refers to word-of-mouth and evaluation information about companies posted on various social networking services.

[0009] "Internal credit" is an internal credit assessment process used to evaluate a company's financial soundness and creditworthiness.

[0010] "Rating" refers to ranking a company's creditworthiness based on its credit score, etc.

[0011] A "machine learning model" is a system that learns patterns based on collected data and makes predictions and classifications.

[0012] A "credit score" is an evaluation index that quantifies a company's financial condition and creditworthiness.

[0013] A "dataset" is a collection of multiple pieces of data.

[0014] "Retraining" is the process of retraining a machine learning model using newly collected data to improve its accuracy.

[0015] "Trend forecasting" involves analyzing collected data to predict future market trends and corporate capital needs. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[0038] 1. The server collects your banking information

[0039] First, the server collects financial information such as a company's account balance, loan status, and repayment history from partner financial institutions via API. After going through authentication procedures, the necessary data is periodically retrieved and stored in a database. Security protection using API keys and OAuth tokens is important during this process.

[0040] Examples:

[0041] The server obtains the account balance of Company A and periodically stores updated repayment history data in a database.

[0042] 2. The server collects industry information

[0043] Next, the server collects industry information such as production numbers, market shares, rankings, etc. from relevant research organizations and industry associations, including accessing research organizations' databases and APIs to understand the latest industry trends.

[0044] Examples:

[0045] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[0046] 3. The server collects social media information

[0047] The server uses the API of the social media platform to collect reviews and ratings of companies, and then uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators.

[0048] Examples:

[0049] The server uses the Twitter API to collect tweets about Company A and quantify the negative / positive ratings.

[0050] 4. The server creates the integrated data set

[0051] The server integrates bank information, industry information, and social media information to generate a dataset for input into the machine learning model, and then preprocesses the data, imputing missing values ​​and standardizing them.

[0052] Examples:

[0053] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[0054] 5. The server evaluates the results using a machine learning model.

[0055] The server inputs the integrated data set into a machine learning model to generate internal credit ratings for businesses, which then predict their credit scores and future financing needs.

[0056] Examples:

[0057] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[0058] 6. The user checks the evaluation results

[0059] Corporate personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. The evaluation results are displayed visually, making them easy for users to understand.

[0060] Examples:

[0061] Company A's financial officer accesses the dashboard and views graphs of the company's credit score and the latest trend forecasts.

[0062] 7. Users make funding decisions

[0063] The company's finance department will consider future fundraising strategies based on the provided evaluation results and market trends, thereby achieving efficient and low-risk fundraising.

[0064] Examples:

[0065] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[0066] 8. The server updates the trend forecast

[0067] The server regularly collects new data and retrains the machine learning model to keep the assessment results and trend predictions up to date, ensuring that both companies and financial institutions are always provided with accurate and up-to-date information.

[0068] Examples:

[0069] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[0070] The system of this invention will enable companies to more easily predict future funding needs and interest rate trends, and financial institutions to make more accurate lending decisions. This is expected to strengthen mutual trust and contribute to stable economic growth.

[0071] The processing flow will be explained below.

[0072] Step 1: Gather your banking information

[0073] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. Before this process begins, the server authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[0074] Specific behavior:

[0075] The server sends a request to an API endpoint to retrieve account balances and debit data filtered by company ID.

[0076] The received data is analyzed in JSON format and the necessary data is saved in the database.

[0077] Step 2: Gather industry information

[0078] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[0079] Specific behavior:

[0080] The server sends a request to the research organization's API to retrieve the latest industry data.

[0081] The acquired data is parsed, processed, and formatted before being saved in a database.

[0082] Step 3: Collect social media information

[0083] The server uses the APIs of various social media platforms to collect reviews and ratings related to companies, and then uses text mining technology to analyze and quantify these reviews.

[0084] Specific behavior:

[0085] The server queries social media APIs using specific keywords or hashtags to gather relevant posts.

[0086] The collected data is subjected to text analysis, positive / negative indicators are calculated, and the evaluation information is quantified and stored in a database.

[0087] Step 4: Creating a consolidated dataset

[0088] The server integrates bank information, industry information, and social media information to create an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[0089] Specific behavior:

[0090] The server merges information from different data sources and converts it into a unified format.

[0091] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[0092] Step 5: Evaluation with machine learning models

[0093] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[0094] Specific behavior:

[0095] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[0096] The scoring results are stored in a database and made available for subsequent processes.

[0097] Step 6: Check the evaluation results

[0098] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend forecasts, which are displayed in graphs and charts.

[0099] Specific behavior:

[0100] The user logs in to the dashboard and checks their company's evaluation results.

[0101] Interact with interactive graphs and charts to get more information.

[0102] Step 7: Funding Decision

[0103] The system develops a funding strategy based on the evaluation results and market trends provided by the user. The evaluation results include forecasts of future funding needs and interest rate trends.

[0104] Specific behavior:

[0105] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[0106] If necessary, submit a new loan application through the online form.

[0107] Step 8: Update trend forecasts

[0108] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[0109] Specific behavior:

[0110] The server recollects new financial, industry, and social media information monthly or weekly.

[0111] Retrain the machine learning model on the new dataset and update the evaluation results.

[0112] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions.

[0113] Example 1

[0114] 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."

[0115] In conventional corporate loan service systems, information collection and analysis from individual data sources was fragmented, making it difficult to conduct comprehensive corporate evaluations. Furthermore, data updates were delayed or intermittent, making it impossible to provide the latest evaluation results and trend forecasts. These issues hindered appropriate credit evaluations of companies and accurate lending decisions by financial institutions.

[0116] 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.

[0117] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from social media platforms; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on companies; means for preprocessing the integrated dataset, completing missing values, and standardizing data; means for providing evaluation results and trend forecasts to companies using the machine learning model; and means for visually displaying the evaluation results and trend forecasts using a dedicated dashboard. This makes it possible to always provide accurate and comprehensive corporate evaluations and the latest trend forecasts.

[0118] "Financial information" refers to data on a company's financial status collected from banks, such as the company's account balance, borrowing status, and repayment history.

[0119] "Industry information" refers to trends and statistical data about a specific industry, such as production numbers, market share, and rankings, and is collected from research organizations and industry associations.

[0120] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various SNS platforms, and is converted into evaluation indicators using text mining technology.

[0121] A "machine learning model" is an algorithm that performs internal credit assessments and ratings on companies based on collected data, and is used to provide evaluation results and trend predictions.

[0122] An "integrated dataset" is a single dataset that integrates financial information, industry information, and social media information, and is used as input data for machine learning models.

[0123] "Preprocessing" refers to the process of formatting input data for a machine learning model, and includes tasks such as filling in missing values ​​and standardizing data.

[0124] "Evaluation results" refer to a company's credit score or internal credit calculated by a machine learning model, and are part of the information provided.

[0125] "Trend forecasts" refer to the results of machine learning models predicting a company's future capital needs and market trends.

[0126] The "dedicated dashboard" is an interface that allows users to visually check evaluation results and trend forecasts, and displays information in graph and table format.

[0127] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[0128] Server Roles and Functions

[0129] In this invention, the server collects financial information, industry information, and social media information of companies, and integrates them to generate a dataset to be input into the machine learning model. Specifically, the following hardware and software are used:

[0130] Hardware: High-performance server (e.g., with Xeon processor)

[0131] Software: API access libraries (e.g., requests), database software (e.g., MySQL, PostgreSQL), data processing libraries (e.g., Pandas, NumPy), machine learning libraries (e.g., Scikit-learn, TensorFlow)

[0132] The server processes and collects information as follows:

[0133] 1. Collection of financial information: The server obtains financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via the API, after going through authentication procedures, and stores the information in a database.

[0134] 2. Collecting industry information: The server accesses databases and APIs of research organizations and industry associations, downloads industry information such as production numbers, market shares, and rankings, and stores it in a database.

[0135] 3. Collection of social media information: The server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company, analyzes it using text mining technology, converts it into quantitative evaluation indicators, and stores them in a database.

[0136] 4. Generating an integrated dataset: Integrating bank information, industry information, and social media information, and preprocessing the data. Specifically, we impute missing values, normalize the data, and convert it into a format suitable for machine learning models.

[0137] Specific examples

[0138] For example, the server retrieves Company A's account balance and repayment history data via API and stores it in a MySQL database. Next, it downloads the latest production volume data for the automobile industry from a research institute and imports it into the database in CSV format using Pandas. Furthermore, it retrieves tweets related to Company A using the Twitter API, performs sentiment analysis using NLTK, and stores the results in a PostgreSQL database.

[0139] Roles and functions of user devices

[0140] The user terminal provides an interface that allows company personnel and financial institution analysts to access a dedicated dashboard and check evaluation results and trend forecasts. The following hardware and software are used.

[0141] Hardware: A standard computer or tablet

[0142] Software: Web browser (e.g., Google Chrome, Mozilla Firefox), dashboard software (e.g., Django, Plotly)

[0143] After logging in, users can access the dashboard and visually check the evaluation results and trend forecasts. The results are displayed in graph and table format, making them easy to understand.

[0144] Specific examples

[0145] A financial officer at Company A opens a web browser, accesses the dashboard URL, and enters his or her login information for authentication. The dashboard then retrieves the latest valuation results from the database and uses Plotly to generate and display graphs.

[0146] Prompt Sentence Examples

[0147] For example, if you want to use a generative AI model to predict Company A's credit score and future financing needs, you might use a prompt like this:

[0148] "To predict Company A's latest credit score and future financing needs, use a dataset that combines up-to-date banking, industry, and social media information."

[0149] The system of the present invention provides accurate and up-to-date information to companies and financial institutions, and is an important tool for achieving efficient and low-risk financing. It is implemented using a specific combination of hardware and software, and users can easily check the evaluation results on a dedicated dashboard.

[0150] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0151] Step 1:

[0152] Server collects banking information

[0153] Input: Financial institution API authentication information, company identification information

[0154] Specific operation: The server uses the API key and OAuth token to connect to the financial institution's API and obtain financial information such as the company's account balance, borrowing status, repayment history, etc. The server sends an HTTP request and receives a response in JSON format.

[0155] Data processing: Parse the received JSON data, extract the necessary information, and save the extracted data in the database.

[0156] Output: A database containing financial information about the company.

[0157] Step 2:

[0158] The server collects industry information

[0159] Input: API endpoint and authentication information of research institute or industry association

[0160] Specific operation: The server sends an HTTP request to the research organization's API to obtain industry information such as production numbers, market share, and rankings.

[0161] Data processing: The acquired data is received in CSV format, parsed using the Pandas library, and the necessary information is extracted. The extracted data is then saved in a database.

[0162] Output: A database containing up-to-date industry information.

[0163] Step 3:

[0164] The server collects SNS information

[0165] Input: API endpoints of social media platforms, keywords and hashtags related to your business

[0166] Specific operation: The server connects to the API of the social media platform and collects reviews and ratings information about the company.

[0167] Data processing: The collected text data is analyzed using text mining techniques such as the Natural Language Toolkit (NLTK) and classified into negative and positive evaluations. The classification results are stored in a database.

[0168] Output: A database containing social media reputation data related to companies.

[0169] Step 4:

[0170] The server creates a consolidated dataset

[0171] Input: Database containing financial information, industry information, and social media information

[0172] Specific operation: The server extracts the necessary information from the database and integrates it into a Pandas DataFrame. It imputes missing values ​​using SimpleImputer, normalizes the data using MinMaxScaler, and performs OneHot encoding on categorical data.

[0173] Data processing: Integrate various information and generate a pre-processed dataset.

[0174] Output: A consolidated dataset for machine learning model input.

[0175] Step 5:

[0176] The server evaluates using a machine learning model

[0177] Input: Integrated dataset

[0178] How it works: The server inputs the integrated dataset into the TensorFlow model to perform internal credit assessment and rating for the company. The model processes the data and calculates the company's credit score.

[0179] Data Computing: Predict credit scores and future financing needs based on integrated data sets.

[0180] Output: Company rating results and credit score.

[0181] Step 6:

[0182] The user checks the evaluation results

[0183] Input: Database containing evaluation results, user authentication information

[0184] How it works: Users access a dedicated dashboard and enter their login information to be authenticated. The dashboard retrieves the evaluation results from the database and generates and displays graphs using Plotly.

[0185] Data calculation: Checks access rights through user authentication and visualizes the evaluation results.

[0186] Output: Visualized evaluation results and trend forecasts.

[0187] Step 7:

[0188] Users make funding decisions

[0189] Input: Visualized evaluation results and trend forecasts

[0190] Specific operation: The user (company's finance department) considers fundraising strategies based on the information in the dashboard.

[0191] Data calculation: Based on the provided evaluation results and trend forecasts, companies can make new loan applications.

[0192] Output: New loan application, interest rate terms set.

[0193] Step 8:

[0194] Server updates trend forecast

[0195] Input: API endpoints to collect new information, existing datasets

[0196] Specific operation: The server periodically collects new data (repeating steps 1 to 3). It uses the new data to retrain the machine learning model and generate the latest evaluation results.

[0197] Data computation: Retraining the model with new data and updating the evaluation results.

[0198] Output: A database containing the latest assessment results and trend forecasts.

[0199] (Application example 1)

[0200] 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."

[0201] Traditional corporate internal credit and rating systems are limited to evaluations based on financial, industry, and social media information, and do not address the specific business operations and efficiency of each company's production plans. As a result, when companies seek financing, they face the dual challenges of fundraising and production management, making it difficult to allocate resources effectively. In particular, optimizing capital management and production plans is a key issue for in-factory production management systems.

[0202] 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.

[0203] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media sites; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on the company; means for providing the evaluation results and trend forecasts to an in-factory production management system to support the optimization of production plans and cash management; and means for regularly retraining the machine learning model to ensure the latest evaluation results and trend forecasts. This enables companies to comprehensively utilize financial information, industry information, and social media information to allocate resources economically, thereby optimizing in-factory production plans and streamlining cash management.

[0204] "Financial information" refers to information collected from banks, including a company's account balance, borrowing status, repayment history, etc.

[0205] "Industry information" refers to data on industry trends, such as production volume, market share, and rankings, obtained from research organizations and industry associations.

[0206] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social networking services.

[0207] A "machine learning model" is an algorithm that uses collected data to assign internal credit and ratings to companies.

[0208] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and financial condition.

[0209] A "rating" is a method of quantitatively evaluating and ranking a company's creditworthiness.

[0210] "Evaluation results" refer to a company's credit score and trend predictions calculated using a machine learning model.

[0211] "Trend forecasting" refers to predicting future corporate credit status and market trends based on machine learning models.

[0212] A "production management system" is a system for managing production plans and resource allocation within a factory.

[0213] "Production planning" refers to planning for production schedules and resource allocation within a factory.

[0214] "Cash management" is the process of optimally managing working capital within a factory.

[0215] This invention is a system that collects and integrates bank information, industry information, and social media information, and uses machine learning models to perform internal credit and rating for companies. This system improves the efficiency of corporate cash management and production planning by linking the functions of the server, factory production management system, and user systems.

[0216] Server Roles

[0217] The server first collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. After going through authentication procedures, the necessary data is periodically obtained and stored in a database. Security protection using API keys and OAuth tokens is important during this process. As a concrete example, the server obtains Company A's account balance and periodically updates the repayment history data and stores it in a database.

[0218] Next, the server collects industry information such as production numbers, market shares, and rankings from relevant research organizations and industry associations. This involves accessing the research organizations' databases and APIs to understand the latest industry trends. For example, the server downloads the latest production number data for the automobile industry from a research organization and stores it in a database.

[0219] Furthermore, the server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company. It uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators. For example, the server uses the Twitter API to collect tweets about Company A and quantifies negative / positive evaluations.

[0220] The server combines this bank information, industry information, and social media information to generate a dataset for input into the machine learning model. It also performs data preprocessing, filling in missing values ​​and standardizing them. For example, the server combines information such as Company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[0221] This integrated data set is input into a machine learning model to perform internal credit assessment and rating of the company, thereby predicting the company's credit score and future funding needs. As a specific example, the server uses the integrated data set of Company A to calculate Company A's credit score using the machine learning model.

[0222] The role of production management systems

[0223] The evaluation results and trend forecasts are provided to the production management system to help optimize the company's cash management and production plans. Company personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. This makes the company's production plans more efficient and optimizes resource allocation. In a specific example, a financial officer at Company A accesses the dashboard to view graphs of the company's credit score and the latest trend forecasts.

[0224] Trend forecast updates

[0225] Finally, the server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. For example, the server collects new industry data and social media information monthly and updates the trend predictions for Company A with the latest generative AI model.

[0226] Prompt Sentence Examples

[0227] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[0228] This allows companies to easily optimize their future capital needs and production plans, thereby optimizing economic resource allocation.

[0229] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0230] Step 1:

[0231] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. Specifically, it accesses the financial institution's API using an API key or OAuth token to obtain the company's financial data. This input data is then stored in a database. For example, the server obtains Company A's account balance data and stores it in the database along with regularly updated repayment history data.

[0232] Step 2:

[0233] The server collects industry information, such as production numbers, market share, and rankings, from relevant research organizations and industry associations. Specifically, it accesses research organization databases and APIs to obtain the latest industry trend data. This input data is later used as part of an integrated dataset. For example, the server downloads the latest production number data for the automotive industry from a research organization and stores it in a database.

[0234] Step 3:

[0235] The server uses the API of a social media platform to collect word-of-mouth and evaluation information about a company. Specifically, it uses the social media API to obtain text data and analyzes it using text mining technology. The input data are tweets and comments about the company, and the output data is a numerical representation of negative and positive evaluations. For example, the server uses the Twitter API to collect tweets about Company A and uses text mining technology to calculate the numerical value of positive evaluations.

[0236] Step 4:

[0237] The server integrates the bank information, industry information, and social media information mentioned above to generate a dataset to input into the machine learning model. Specifically, it preprocesses the data and performs missing value completion and standardization. The input data is each piece of information mentioned above, and the output data is the integrated dataset. For example, the server integrates information such as Company A's account balance, production volume, and social media ratings and processes it into a single dataset.

[0238] Step 5:

[0239] The server inputs the integrated dataset into a machine learning model to perform internal credit assessment and rating of the company. Specifically, the preprocessed data is input into the machine learning model to predict the company's credit score and future capital needs. The input data is the integrated dataset, and the output data is the company's credit score and trend prediction. For example, the server uses the integrated dataset of Company A to calculate Company A's credit score.

[0240] Step 6:

[0241] The server provides the evaluation results and trend forecasts to the production management system within the factory, supporting the optimization of production plans and financial management. Specifically, it accesses a dedicated dashboard terminal and displays the evaluation results and forecast data in graph and report format. The input data is the evaluation results and trend forecasts, and the output data is information displayed visually. For example, a financial officer at Company A accesses the dashboard to check information on the company's credit score and production plan optimization.

[0242] Step 7:

[0243] The server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. Specifically, new bank information, industry information, and social media information are collected monthly or weekly, and the new data is re-input into the machine learning model to retrain the model. The input data is the newly collected information, and the output data is the latest evaluation results and trend predictions. For example, the server collects the latest industry data and social media information and updates the trend predictions for Company A with a new generative AI model.

[0244] Prompt Sentence Examples

[0245] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[0246] 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.

[0247] This invention is a system for corporate loan services that collects and integrates bank information, industry information, and social media information, performs internal credit assessment and rating for companies using machine learning models, and further combines an emotion engine to recognize user emotions and customize evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising by linking the functions of the server, terminal, and user.

[0248] 1. Collection of banking information

[0249] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. The server authenticates with the financial institution using an API key or OAuth token, periodically retrieves the data, and stores it in a database.

[0250] Examples:

[0251] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[0252] 2. Collecting industry information

[0253] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[0254] Examples:

[0255] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[0256] 3. Collecting social media information

[0257] The server uses the API of the social media platform to collect word-of-mouth and evaluation information about companies, and then uses text mining technology to analyze and quantify the text data.

[0258] Examples:

[0259] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[0260] 4. Creation of an integrated dataset

[0261] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[0262] Examples:

[0263] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[0264] 5. Evaluation with machine learning models

[0265] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and credit score. The machine learning model is pre-trained and calculates the credit score, etc.

[0266] Examples:

[0267] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[0268] 6. Emotion Recognition with Emotion Engine

[0269] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's voice and text inputs and recognizes their emotions to customize the display format of the evaluation results and trend predictions.

[0270] Examples:

[0271] The server uses an emotion engine to analyze the user's voice input and adjusts the display of the evaluation results based on the results.

[0272] 7. Checking the evaluation results

[0273] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The display format of the evaluation results is customized according to the user's emotional state.

[0274] Examples:

[0275] Company A's finance manager accesses a dashboard to view the company's credit score and latest trend forecasts in an emotionally relevant format.

[0276] 8. Funding Decisions

[0277] Users can develop fundraising strategies based on the provided evaluation results and market trends. The display format reflects the user's emotional state, allowing them to make more appropriate decisions.

[0278] Examples:

[0279] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[0280] 9. Trend forecast updates

[0281] The server regularly collects new data and retrains the machine learning models to keep the assessment results and trend predictions up to date, ensuring accurate and up-to-date information.

[0282] Examples:

[0283] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[0284] The system of this invention allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides advice and displays tailored to each user, improving user satisfaction and enabling more effective fundraising.

[0285] The processing flow will be explained below.

[0286] Step 1: Gather your banking information

[0287] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. During this process, the server first authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[0288] Specific behavior:

[0289] The server sends a request to an API endpoint to retrieve account balances and debit data based on the company ID.

[0290] The received data is analyzed in JSON format and the necessary data is saved in the database.

[0291] Step 2: Gather industry information

[0292] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, rankings, etc.

[0293] Specific behavior:

[0294] The server sends a request to the research organization's API to retrieve the latest industry data.

[0295] The acquired data is parsed, processed, and formatted before being saved in a database.

[0296] Step 3: Collect social media information

[0297] The server uses the APIs of various social media platforms to collect word-of-mouth and evaluation information related to companies and analyzes it using text mining technology.

[0298] Specific behavior:

[0299] The server queries the social media API using specific keywords or hashtags to gather relevant posts.

[0300] The collected data is subjected to text analysis, and positive / negative evaluations are quantified and stored in a database.

[0301] Step 4: Creating a consolidated dataset

[0302] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[0303] Specific behavior:

[0304] The server merges information from different data sources and converts it into a unified format.

[0305] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[0306] Step 5: Evaluation with machine learning models

[0307] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[0308] Specific behavior:

[0309] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[0310] The scoring results are stored in a database and made available for subsequent processes.

[0311] Step 6: Emotion Recognition in the Emotion Engine

[0312] The server analyzes the user's voice and text input and uses an emotion engine to recognize emotions, thereby customizing the display and system responses to the user's emotional state.

[0313] Specific behavior:

[0314] The server passes the user's voice and text input to the emotion engine and obtains the analysis results.

[0315] Based on the analysis results, the dashboard display format and response content are adjusted.

[0316] Step 7: Check the evaluation results

[0317] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend predictions, and the display content is customized according to the user's emotional state.

[0318] Specific behavior:

[0319] Users log in to a dashboard to see their company's credit score and the latest trend forecasts.

[0320] Interact with interactive graphs and charts to get more information.

[0321] Step 8: Funding Decision

[0322] The system creates a fundraising strategy based on the evaluation results and market trends provided by the user. The system supports appropriate decisions by displaying the results in a format that responds to the user's emotions.

[0323] Specific behavior:

[0324] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[0325] If necessary, submit a new loan application through the online form.

[0326] Step 9: Update trend forecast

[0327] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[0328] Specific behavior:

[0329] The server recollects new financial, industry, and social media information monthly or weekly.

[0330] Retrain the machine learning model on the new dataset and update the evaluation results.

[0331] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides personalized advice and displays, improving user satisfaction and enabling more effective fundraising.

[0332] Example 2

[0333] 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."

[0334] Corporate lending services collect and integrate bank, industry, and social media information to assess a company's creditworthiness, but they often fail to accurately assess it due to insufficient consideration of data variability and emotions. Furthermore, the methods used to present assessment results and trend forecasts are unable to adapt to the user's emotions, sometimes resulting in insufficient support for user decision-making. This makes it difficult for financial institutions to make reliable lending decisions, making it difficult for companies to raise funds at reasonable interest rates.

[0335] 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.

[0336] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media; means for integrating the financial information, industry information, and SNS information and performing internal credit and rating of the company using a machine learning model; means for recognizing user emotions using an emotion engine and customizing the display format of the evaluation results and trend forecasts; and means for providing the evaluation results and trend forecasts using the machine learning model to the company. This enables internal credit and rating of companies to be performed with high accuracy, and by presenting evaluation results that take user emotions into consideration, it becomes possible to support more appropriate decision-making.

[0337] "Bank information" refers to public financial data obtained from financial institutions, such as a company's account balance, borrowing status, and repayment history.

[0338] "Industry information" refers to data about a particular industry, such as industry production volumes, market shares, and rankings.

[0339] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social media platforms.

[0340] A "machine learning model" is an algorithm used to generate internal credit ratings and ratings for companies based on collected data.

[0341] An "emotion engine" is a system that analyzes a user's voice and text input and recognizes emotions.

[0342] "User" refers to a person who uses this system to check the evaluation results and trend forecasts of a company and make a decision on fundraising.

[0343] A "database" is a data infrastructure for storing and managing collected financial information, industry information, and social media information.

[0344] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and suitability for lending.

[0345] A "rating" is a grade calculated using a machine learning model to indicate a company's creditworthiness and financial stability.

[0346] MODE FOR CARRYING OUT THE INVENTION

[0347] This invention is a system for corporate loan services that links the functions of servers, terminals, and users to collect and integrate corporate financial information, industry information, and social media information, and uses machine learning models and emotion engines to perform internal credit assessments and ratings for companies, providing evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising.

[0348] Collection of banking information

[0349] The server collects financial information about the company from partner financial institutions via API. Specifically, it obtains financial data such as account balances, borrowing status, and repayment history. The server authenticates with the financial institution using an API key or OAuth token, periodically obtains the data, and stores it in a database.

[0350] Examples:

[0351] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[0352] Collecting industry information

[0353] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[0354] Examples:

[0355] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[0356] Collecting SNS information

[0357] The server uses the API of the social media platform to collect reviews and ratings of companies, and analyzes them using text mining technology. The acquired text data is then analyzed and quantified.

[0358] Examples:

[0359] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[0360] Integrating data and using machine learning models

[0361] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data. The preprocessed integrated dataset is then input into the machine learning model to generate a company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[0362] Examples:

[0363] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[0364] Using the Emotion Engine

[0365] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's voice and text inputs and recognizes their emotions to customize the display format of the evaluation results and trend predictions.

[0366] Examples:

[0367] The server uses an emotion engine to analyze the user's voice input and adjusts the display of the evaluation results based on the results.

[0368] Reviewing evaluation results and making funding decisions

[0369] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The display format is customized according to the user's emotional state. Based on the evaluation results and market trends, it becomes possible to develop a fundraising strategy, apply for new loans, and raise funds at appropriate interest rates.

[0370] Examples:

[0371] Company A's finance manager accesses a dashboard to view the company's credit score and latest trend forecasts in an emotionally relevant format.

[0372] Based on the evaluation results, the financial officer will apply for new loans and raise funds at appropriate interest rates.

[0373] Trend forecast updates

[0374] The server periodically collects new data and retrains the machine learning model to keep the assessment results and trend predictions up to date.

[0375] Examples:

[0376] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[0377] This allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions.In addition, the emotion engine provides advice and displays tailored to each user, improving user satisfaction and enabling more effective fundraising.

[0378] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0379] System program processing flow

[0380] Step 1:

[0381] Collection of banking information

[0382] The server collects a company's financial information from affiliated financial institutions via API. It authenticates with the financial institution using an API key or OAuth token as input and obtains data such as account balance, borrowing status, and repayment history. The obtained financial information is received in JSON format, which the server parses and stores in a database.

[0383] Specific behavior:

[0384] 1. The server sends a request to the financial institution's API endpoint.

[0385] 2. Authenticate using an API key or OAuth token.

[0386] 3. Analyze the received JSON format financial information and extract each company's account balance, borrowing status, and repayment history.

[0387] 4. Save the analyzed data in a database.

[0388] Step 2:

[0389] Collecting industry information

[0390] The server collects industry information from relevant research institutes via API or FTP. It authenticates using API keys, usernames, and passwords as input, and obtains industry production figures, market shares, industry rankings, etc. The obtained industry information is analyzed and stored in a database.

[0391] Specific behavior:

[0392] 1. The server sends a request to the research organization's API endpoint.

[0393] 2. Authenticate if necessary (API key or username / password).

[0394] 3. Analyze the received industry information (e.g., production volume data) and extract the necessary information.

[0395] 4. Save the extracted data in the database.

[0396] Step 3:

[0397] Collecting SNS information

[0398] The server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company. It authenticates using the social media API key as input and obtains tweet data containing the company name and related keywords. It then analyzes the obtained tweet data using text mining technology, converting positive / negative evaluations into numerical values ​​and storing them in a database.

[0399] Specific behavior:

[0400] 1. The server sends a request to the Twitter API.

[0401] 2. Obtain tweets containing the company name and related keywords.

[0402] 3. Analyze tweet data using text mining technology and perform sentiment analysis.

[0403] 4. Quantify the negative / positive ratings and store them in a database.

[0404] Step 4:

[0405] Creating a unified dataset

[0406] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. It obtains each of the above information as input from a database, completes missing values, and normalizes the data. It then integrates the preprocessed data to generate a dataset for the machine learning model.

[0407] Specific behavior:

[0408] 1. The server retrieves the collected bank information, industry information, and social media ratings from the database.

[0409] 2. Impute missing values ​​(e.g., impute by the mean) and normalize the data.

[0410] 3. Integrate various information to generate a single data set.

[0411] Step 5:

[0412] Evaluation with machine learning models

[0413] The server inputs the preprocessed integrated dataset into the machine learning model to perform internal credit assessment and rating of the company. Using the integrated dataset as input, the machine learning model outputs the company's credit score and rating results. The calculation results are stored in a database.

[0414] Specific behavior:

[0415] 1. The server reads the consolidated data set.

[0416] 2. Input the integrated dataset into a machine learning model.

[0417] 3. Machine learning models calculate a company's credit score and rating.

[0418] 4. Save the calculation results in the database.

[0419] Step 6:

[0420] Emotion recognition with emotion engine

[0421] The server uses an emotion engine to recognize the user's emotions. It receives the user's voice and text input as input and analyzes the emotions using the emotion engine. It customizes the display format of the evaluation results and trend predictions according to the user's emotional state.

[0422] Specific behavior:

[0423] 1. The server receives the user's voice or text input.

[0424] 2. The emotion engine performs analysis to recognize the user's emotional state.

[0425] 3. Based on the recognition results, adjust the display format of the evaluation results and trend predictions.

[0426] Step 7:

[0427] Checking the evaluation results

[0428] The user visually checks the evaluation results and trend predictions using a dedicated dashboard terminal. The terminal receives the evaluation results provided by the server as input and outputs the content in a customized display format according to the user's emotional state.

[0429] Specific behavior:

[0430] 1. The user logs in to the dashboard terminal.

[0431] 2. The server verifies the user's authentication and displays the dashboard content.

[0432] 3. The dashboard displays the latest assessment results and trend forecasts.

[0433] 4. The display format is customized according to the user's emotional state.

[0434] Step 8:

[0435] Funding decisions

[0436] The user creates a fundraising strategy based on the provided evaluation results and market trends. Using the information confirmed on the dashboard as input, the user submits a new loan application and outputs the details of raising funds at appropriate interest rates.

[0437] Specific behavior:

[0438] 1. The user checks the evaluation results displayed on the dashboard.

[0439] 2. Consider financing scenarios based on market trends.

[0440] 3. If necessary, file a new loan application.

[0441] 4. Appropriate interest rates will be set based on the loan application.

[0442] Step 9:

[0443] Trend forecast updates

[0444] The server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. It collects new financial, industry, and social media information as input and retrains the machine learning model using the updated dataset. It saves the latest prediction results in the database and updates the information provided to users.

[0445] Specific behavior:

[0446] 1. The server periodically collects new financial, industry, and social media information.

[0447] 2. Update the integrated dataset based on the collected data.

[0448] 3. Retrain the machine learning model (update the generative AI model).

[0449] 4. Store the latest evaluation results and trend forecasts in a database.

[0450] (Application example 2)

[0451] 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."

[0452] Conventional corporate lending services collect and analyze financial information, industry information, and social media information separately, making it difficult to provide comprehensive, real-time credit assessments. Furthermore, they do not take into account the user's emotional state, meaning that assessment results and forecast information are not optimized to fit the user's situation. This makes it difficult to make accurate lending decisions and manage transaction risks.

[0453] The specific processing by the specific 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 financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; and means for collecting word-of-mouth and evaluation information about companies from various social networking sites. This enables financial information, industry information, and social networking site information to be integrated and a machine learning model to perform internal credit assessment and rating of companies. Furthermore, by including an emotion engine that recognizes user emotions and customizes the display format of evaluation results and trend forecasts, it is possible to provide information tailored to the user. Furthermore, by having the machine learning model perform credit assessment of each transaction in real time and periodically retrain, the assessment results and trend forecasts are always kept up to date, enabling highly accurate lending decisions and transaction risk management.

[0454] "Bank information" refers to financial information such as a company's account balance, borrowing status, repayment history, etc. This information is necessary to evaluate a company's financial condition.

[0455] "Industry information" is data that includes production volume, market share, rankings, etc. in a specific industry. This information is necessary to understand the overall trends in the industry to which a company belongs.

[0456] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social networking services. This information is used to understand the evaluations and feelings of the general public and customers toward companies.

[0457] A "machine learning model" is an algorithm that performs internal credit assessment and rating of a company based on collected data, making it possible to evaluate a company's credit score and risk.

[0458] The "emotion engine" is a system that recognizes the user's emotions. It analyzes the user's voice and text input and customizes the display format of evaluation results and trend predictions according to the user's emotional state.

[0459] "Evaluation results" are the credit scores and internal credit results of companies obtained using machine learning models, which are used to determine the transaction risk and lending decisions for companies.

[0460] "Trend forecasts" are information used to predict a company's future trends. They use collected data to predict a company's future risks and growth potential.

[0461] "Real-time evaluation" refers to a credit evaluation that is conducted in real time for each transaction, allowing for accurate evaluation based on the latest data at all times.

[0462] "Retraining" is the process of updating a machine learning model based on newly collected data, ensuring that its assessment results and trend predictions are always up-to-date.

[0463] A system for implementing the present invention operates in cooperation with multiple components including a server, a user terminal, a machine learning model, and an emotion engine.

[0464] Data collection

[0465] First, the server collects financial information such as a company's account balance, borrowing status, and repayment history from the bank via API. The server authenticates with the financial institution using an API key or OAuth token, periodically obtains the latest data, and stores it in a database. Industry information, such as production volume, market share, and industry rankings, is also collected from related research institutions via API or FTP. Furthermore, social media information, such as reviews and ratings of companies, is collected using the API of social media platforms, and text data is analyzed and quantified using text mining technology.

[0466] Data Integration and Evaluation

[0467] The server integrates financial, industry, and social media information and preprocesses the data. Specifically, it completes missing values ​​and normalizes them to generate an input dataset for the machine learning model. The generated dataset is then fed into a pre-trained machine learning model to generate a company's internal credit rating and credit rating. The machine learning model is built using machine learning libraries such as TensorFlow and scikit-learn.

[0468] Emotion Recognition and Customization

[0469] The server uses an emotion engine to analyze the user's voice and text input. This emotion engine uses natural language processing libraries such as TextBlob to evaluate the user's emotions with a positive / negative score. The evaluation results and trend predictions are displayed in a customized format according to the user's emotional state.

[0470] Displaying results and making funding decisions

[0471] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The evaluation results are displayed in a format that is customized according to the user's emotional state, allowing the user to make more appropriate decisions. For example, a company's financial officer can look at the evaluation results and apply for new loans, ensuring that funds are raised at appropriate interest rates.

[0472] Specific examples of programs

[0473] A specific example of the embodiment is as follows:

[0474] When a company's financial officer initiates a new transaction, the server collects the company's latest financial, industry, and social media information, and inputs the integrated data into a machine learning model to calculate a credit score. At this point, an emotion engine analyzes the officer's emotional state and displays the evaluation results in an optimal format. This allows the financial officer to make appropriate financing decisions based on the evaluation results, such as avoiding high-risk transactions.

[0475] Example prompt sentence:

[0476] When doing new business, I want to see a company's current credit score and market trends, and if my feelings indicate concern, I want the system to provide feedback on its risk assessment as well.

[0477] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0478] Step 1:

[0479] The server collects financial information such as a company's account balance, borrowing status, and repayment history from financial institutions via API. At this time, the server authenticates with the financial institution using an API key or OAuth token, and periodically retrieves the latest financial information into a database. The input is financial information obtained from the financial institution's API, and the output is financial information stored in the server's database. By collecting financial information, data can be obtained to understand the company's current financial situation.

[0480] Step 2:

[0481] The server collects industry information from related research institutes via API or FTP. Specifically, it obtains data such as production volume, market share, and industry rankings. The input is industry information obtained from the research institute's API, and the output is industry information stored in the server's database. By collecting industry information, data can be obtained to understand trends in the industry to which a company belongs.

[0482] Step 3:

[0483] The server uses the API of the SNS platform to collect word-of-mouth and evaluation information about the company. The acquired text data is analyzed using text mining technology and quantified as positive / negative evaluations. The input is text data obtained from the SNS API, and the output is quantified SNS evaluation information. By collecting and analyzing SNS information, data can be obtained that visualizes the evaluation of the company by the general public and customers.

[0484] Step 4:

[0485] The server integrates the collected financial, industry, and social media information and preprocesses this data. Specifically, it imputes missing values ​​and normalizes the data to generate an input dataset for the machine learning model. The input is financial, industry, and social media information, and the output is a preprocessed integrated dataset. Preprocessing the data prepares the machine learning model for optimal operation.

[0486] Step 5:

[0487] The server inputs the preprocessed integrated dataset into a machine learning model to perform internal credit assessment and rating of the company. The input is the preprocessed integrated dataset, and the output is the company's credit score and risk assessment results. Using a machine learning model for evaluation makes it possible to quantitatively evaluate the company's credit status.

[0488] Step 6:

[0489] The server uses an emotion engine to analyze the user's voice and text input. Specifically, it uses a natural language processing library such as TextBlob to analyze emotions and obtains the results as a score. The input is the user's voice or text data, and the output is an emotion score. Using the emotion engine, it is possible to accurately recognize the user's emotions and provide feedback accordingly.

[0490] Step 7:

[0491] The server provides users with the evaluation results and trend predictions obtained from the machine learning model. At this time, the display format of the evaluation results is customized according to the user's emotional score. The inputs are the credit score, risk evaluation results, and emotional score, and the output is a display of the customized evaluation results. Users can visually check these results using a dashboard terminal, enabling them to make more appropriate decisions.

[0492] Step 8:

[0493] Users determine transaction risks and funding strategies based on the provided evaluation results and trend forecasts. This allows them to make economic decisions such as procuring funds at appropriate interest rates or avoiding high-risk transactions. The input is customized evaluation results, and the output is specific funding decisions. Evaluation displays based on the user's emotional state enable appropriate decisions.

[0494] Step 9:

[0495] The server periodically collects new data and retrains the machine learning model, allowing it to always provide the latest evaluation results and trend forecasts. The input is newly collected financial, industry, and social media information, and the output is an updated machine learning model and evaluation results. Periodic retraining improves the accuracy of the evaluation, enabling the provision of highly reliable information.

[0496] 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.

[0497] 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.

[0498] 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.

[0499] [Second embodiment]

[0500] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0501] 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.

[0502] 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).

[0503] 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.

[0504] 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.

[0505] 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).

[0506] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0507] 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.

[0508] 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.

[0509] 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.

[0510] 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.

[0511] 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."

[0512] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[0513] 1. The server collects your banking information

[0514] First, the server collects financial information such as a company's account balance, loan status, and repayment history from partner financial institutions via API. After going through authentication procedures, the necessary data is periodically retrieved and stored in a database. Security protection using API keys and OAuth tokens is important during this process.

[0515] Examples:

[0516] The server obtains the account balance of Company A and periodically stores updated repayment history data in a database.

[0517] 2. The server collects industry information

[0518] Next, the server collects industry information such as production numbers, market shares, rankings, etc. from relevant research organizations and industry associations, including accessing research organizations' databases and APIs to understand the latest industry trends.

[0519] Examples:

[0520] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[0521] 3. The server collects social media information

[0522] The server uses the API of the social media platform to collect reviews and ratings of companies, and then uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators.

[0523] Examples:

[0524] The server uses the Twitter API to collect tweets about Company A and quantify the negative / positive ratings.

[0525] 4. The server creates the integrated data set

[0526] The server integrates bank information, industry information, and social media information to generate a dataset for input into the machine learning model, and then preprocesses the data, imputing missing values ​​and standardizing them.

[0527] Examples:

[0528] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[0529] 5. The server evaluates the results using a machine learning model.

[0530] The server inputs the integrated data set into a machine learning model to generate internal credit ratings for businesses, which then predict their credit scores and future financing needs.

[0531] Examples:

[0532] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[0533] 6. The user checks the evaluation results

[0534] Corporate personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. The evaluation results are displayed visually, making them easy for users to understand.

[0535] Examples:

[0536] Company A's financial officer accesses the dashboard and views graphs of the company's credit score and the latest trend forecasts.

[0537] 7. Users make funding decisions

[0538] The company's finance department will consider future fundraising strategies based on the provided evaluation results and market trends, thereby achieving efficient and low-risk fundraising.

[0539] Examples:

[0540] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[0541] 8. The server updates the trend forecast

[0542] The server regularly collects new data and retrains the machine learning model to keep the assessment results and trend predictions up to date, ensuring that both companies and financial institutions are always provided with accurate and up-to-date information.

[0543] Examples:

[0544] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[0545] The system of this invention will enable companies to more easily predict future funding needs and interest rate trends, and financial institutions to make more accurate lending decisions. This is expected to strengthen mutual trust and contribute to stable economic growth.

[0546] The processing flow will be explained below.

[0547] Step 1: Gather your banking information

[0548] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. Before this process begins, the server authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[0549] Specific behavior:

[0550] The server sends a request to an API endpoint to retrieve account balances and debit data filtered by company ID.

[0551] The received data is analyzed in JSON format and the necessary data is saved in the database.

[0552] Step 2: Gather industry information

[0553] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[0554] Specific behavior:

[0555] The server sends a request to the research organization's API to retrieve the latest industry data.

[0556] The acquired data is parsed, processed, and formatted before being saved in a database.

[0557] Step 3: Collect social media information

[0558] The server uses the APIs of various social media platforms to collect reviews and ratings related to companies, and then uses text mining technology to analyze and quantify these reviews.

[0559] Specific behavior:

[0560] The server queries social media APIs using specific keywords or hashtags to gather relevant posts.

[0561] The collected data is subjected to text analysis, positive / negative indicators are calculated, and the evaluation information is quantified and stored in a database.

[0562] Step 4: Creating a consolidated dataset

[0563] The server integrates bank information, industry information, and social media information to create an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[0564] Specific behavior:

[0565] The server merges information from different data sources and converts it into a unified format.

[0566] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[0567] Step 5: Evaluation with machine learning models

[0568] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[0569] Specific behavior:

[0570] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[0571] The scoring results are stored in a database and made available for subsequent processes.

[0572] Step 6: Check the evaluation results

[0573] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend forecasts, which are displayed in graphs and charts.

[0574] Specific behavior:

[0575] The user logs in to the dashboard and checks their company's evaluation results.

[0576] Interact with interactive graphs and charts to get more information.

[0577] Step 7: Funding Decision

[0578] The system develops a funding strategy based on the evaluation results and market trends provided by the user. The evaluation results include forecasts of future funding needs and interest rate trends.

[0579] Specific behavior:

[0580] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[0581] If necessary, submit a new loan application through the online form.

[0582] Step 8: Update trend forecasts

[0583] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[0584] Specific behavior:

[0585] The server recollects new financial, industry, and social media information monthly or weekly.

[0586] Retrain the machine learning model on the new dataset and update the evaluation results.

[0587] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions.

[0588] Example 1

[0589] 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."

[0590] In conventional corporate loan service systems, information collection and analysis from individual data sources was fragmented, making it difficult to conduct comprehensive corporate evaluations. Furthermore, data updates were delayed or intermittent, making it impossible to provide the latest evaluation results and trend forecasts. These issues hindered appropriate credit evaluations of companies and accurate lending decisions by financial institutions.

[0591] 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.

[0592] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from social media platforms; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on companies; means for preprocessing the integrated dataset, completing missing values, and standardizing data; means for providing evaluation results and trend forecasts to companies using the machine learning model; and means for visually displaying the evaluation results and trend forecasts using a dedicated dashboard. This makes it possible to always provide accurate and comprehensive corporate evaluations and the latest trend forecasts.

[0593] "Financial information" refers to data on a company's financial status collected from banks, such as the company's account balance, borrowing status, and repayment history.

[0594] "Industry information" refers to trends and statistical data about a specific industry, such as production numbers, market share, and rankings, and is collected from research organizations and industry associations.

[0595] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various SNS platforms, and is converted into evaluation indicators using text mining technology.

[0596] A "machine learning model" is an algorithm that performs internal credit assessments and ratings on companies based on collected data, and is used to provide evaluation results and trend predictions.

[0597] An "integrated dataset" is a single dataset that integrates financial information, industry information, and social media information, and is used as input data for machine learning models.

[0598] "Preprocessing" refers to the process of formatting input data for a machine learning model, and includes tasks such as filling in missing values ​​and standardizing data.

[0599] "Evaluation results" refer to a company's credit score or internal credit calculated by a machine learning model, and are part of the information provided.

[0600] "Trend forecasts" refer to the results of machine learning models predicting a company's future capital needs and market trends.

[0601] The "dedicated dashboard" is an interface that allows users to visually check evaluation results and trend forecasts, and displays information in graph and table format.

[0602] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[0603] Server Roles and Functions

[0604] In this invention, the server collects financial information, industry information, and social media information of companies, and integrates them to generate a dataset to be input into the machine learning model. Specifically, the following hardware and software are used:

[0605] Hardware: High-performance server (e.g., with Xeon processor)

[0606] Software: API access libraries (e.g., requests), database software (e.g., MySQL, PostgreSQL), data processing libraries (e.g., Pandas, NumPy), machine learning libraries (e.g., Scikit-learn, TensorFlow)

[0607] The server processes and collects information as follows:

[0608] 1. Collection of financial information: The server obtains financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via the API, after going through authentication procedures, and stores the information in a database.

[0609] 2. Collecting industry information: The server accesses databases and APIs of research organizations and industry associations, downloads industry information such as production numbers, market shares, and rankings, and stores it in a database.

[0610] 3. Collection of social media information: The server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company, analyzes it using text mining technology, converts it into quantitative evaluation indicators, and stores them in a database.

[0611] 4. Generating an integrated dataset: Integrating bank information, industry information, and social media information, and preprocessing the data. Specifically, we impute missing values, normalize the data, and convert it into a format suitable for machine learning models.

[0612] Specific examples

[0613] For example, the server retrieves Company A's account balance and repayment history data via API and stores it in a MySQL database. Next, it downloads the latest production volume data for the automobile industry from a research institute and imports it into the database in CSV format using Pandas. Furthermore, it retrieves tweets related to Company A using the Twitter API, performs sentiment analysis using NLTK, and stores the results in a PostgreSQL database.

[0614] Roles and functions of user devices

[0615] The user terminal provides an interface that allows company personnel and financial institution analysts to access a dedicated dashboard and check evaluation results and trend forecasts. The following hardware and software are used.

[0616] Hardware: A standard computer or tablet

[0617] Software: Web browser (e.g., Google Chrome, Mozilla Firefox), dashboard software (e.g., Django, Plotly)

[0618] After logging in, users can access the dashboard and visually check the evaluation results and trend forecasts. The results are displayed in graph and table format, making them easy to understand.

[0619] Specific examples

[0620] A financial officer at Company A opens a web browser, accesses the dashboard URL, and enters his or her login information for authentication. The dashboard then retrieves the latest valuation results from the database and uses Plotly to generate and display graphs.

[0621] Prompt Sentence Examples

[0622] For example, if you want to use a generative AI model to predict Company A's credit score and future financing needs, you might use a prompt like this:

[0623] "To predict Company A's latest credit score and future financing needs, use a dataset that combines up-to-date banking, industry, and social media information."

[0624] The system of the present invention provides accurate and up-to-date information to companies and financial institutions, and is an important tool for achieving efficient and low-risk financing. It is implemented using a specific combination of hardware and software, and users can easily check the evaluation results on a dedicated dashboard.

[0625] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0626] Step 1:

[0627] Server collects banking information

[0628] Input: Financial institution API authentication information, company identification information

[0629] Specific operation: The server uses the API key and OAuth token to connect to the financial institution's API and obtain financial information such as the company's account balance, borrowing status, repayment history, etc. The server sends an HTTP request and receives a response in JSON format.

[0630] Data processing: Parse the received JSON data, extract the necessary information, and save the extracted data in the database.

[0631] Output: A database containing financial information about the company.

[0632] Step 2:

[0633] The server collects industry information

[0634] Input: API endpoint and authentication information of research institute or industry association

[0635] Specific operation: The server sends an HTTP request to the research organization's API to obtain industry information such as production numbers, market share, and rankings.

[0636] Data processing: The acquired data is received in CSV format, parsed using the Pandas library, and the necessary information is extracted. The extracted data is then saved in a database.

[0637] Output: A database containing up-to-date industry information.

[0638] Step 3:

[0639] The server collects SNS information

[0640] Input: API endpoints of social media platforms, keywords and hashtags related to your business

[0641] Specific operation: The server connects to the API of the social media platform and collects reviews and ratings information about the company.

[0642] Data processing: The collected text data is analyzed using text mining techniques such as the Natural Language Toolkit (NLTK) and classified into negative and positive evaluations. The classification results are stored in a database.

[0643] Output: A database containing social media reputation data related to companies.

[0644] Step 4:

[0645] The server creates a consolidated dataset

[0646] Input: Database containing financial information, industry information, and social media information

[0647] Specific operation: The server extracts the necessary information from the database and integrates it into a Pandas DataFrame. It imputes missing values ​​using SimpleImputer, normalizes the data using MinMaxScaler, and performs OneHot encoding on categorical data.

[0648] Data processing: Integrate various information and generate a pre-processed dataset.

[0649] Output: A consolidated dataset for machine learning model input.

[0650] Step 5:

[0651] The server evaluates using a machine learning model

[0652] Input: Integrated dataset

[0653] How it works: The server inputs the integrated dataset into the TensorFlow model to perform internal credit assessment and rating for the company. The model processes the data and calculates the company's credit score.

[0654] Data Computing: Predict credit scores and future financing needs based on integrated data sets.

[0655] Output: Company rating results and credit score.

[0656] Step 6:

[0657] The user checks the evaluation results

[0658] Input: Database containing evaluation results, user authentication information

[0659] How it works: Users access a dedicated dashboard and enter their login information to be authenticated. The dashboard retrieves the evaluation results from the database and generates and displays graphs using Plotly.

[0660] Data calculation: Checks access rights through user authentication and visualizes the evaluation results.

[0661] Output: Visualized evaluation results and trend forecasts.

[0662] Step 7:

[0663] Users make funding decisions

[0664] Input: Visualized evaluation results and trend forecasts

[0665] Specific operation: The user (company's finance department) considers fundraising strategies based on the information in the dashboard.

[0666] Data calculation: Based on the provided evaluation results and trend forecasts, companies can make new loan applications.

[0667] Output: New loan application, interest rate terms set.

[0668] Step 8:

[0669] Server updates trend forecast

[0670] Input: API endpoints to collect new information, existing datasets

[0671] Specific operation: The server periodically collects new data (repeating steps 1 to 3). It uses the new data to retrain the machine learning model and generate the latest evaluation results.

[0672] Data computation: Retraining the model with new data and updating the evaluation results.

[0673] Output: A database containing the latest assessment results and trend forecasts.

[0674] (Application example 1)

[0675] 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."

[0676] Traditional corporate internal credit and rating systems are limited to evaluations based on financial, industry, and social media information, and do not address the specific business operations and efficiency of each company's production plans. As a result, when companies seek financing, they face the dual challenges of fundraising and production management, making it difficult to allocate resources effectively. In particular, optimizing capital management and production plans is a key issue for in-factory production management systems.

[0677] 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.

[0678] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media sites; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on the company; means for providing the evaluation results and trend forecasts to an in-factory production management system to support the optimization of production plans and cash management; and means for regularly retraining the machine learning model to ensure the latest evaluation results and trend forecasts. This enables companies to comprehensively utilize financial information, industry information, and social media information to allocate resources economically, thereby optimizing in-factory production plans and streamlining cash management.

[0679] "Financial information" refers to information collected from banks, including a company's account balance, borrowing status, repayment history, etc.

[0680] "Industry information" refers to data on industry trends, such as production volume, market share, and rankings, obtained from research organizations and industry associations.

[0681] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social networking services.

[0682] A "machine learning model" is an algorithm that uses collected data to assign internal credit and ratings to companies.

[0683] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and financial condition.

[0684] A "rating" is a method of quantitatively evaluating and ranking a company's creditworthiness.

[0685] "Evaluation results" refer to a company's credit score and trend predictions calculated using a machine learning model.

[0686] "Trend forecasting" refers to predicting future corporate credit status and market trends based on machine learning models.

[0687] A "production management system" is a system for managing production plans and resource allocation within a factory.

[0688] "Production planning" refers to planning for production schedules and resource allocation within a factory.

[0689] "Cash management" is the process of optimally managing working capital within a factory.

[0690] This invention is a system that collects and integrates bank information, industry information, and social media information, and uses machine learning models to perform internal credit and rating for companies. This system improves the efficiency of corporate cash management and production planning by linking the functions of the server, factory production management system, and user systems.

[0691] Server Roles

[0692] The server first collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. After going through authentication procedures, the necessary data is periodically obtained and stored in a database. Security protection using API keys and OAuth tokens is important during this process. As a concrete example, the server obtains Company A's account balance and periodically updates the repayment history data and stores it in a database.

[0693] Next, the server collects industry information such as production numbers, market shares, and rankings from relevant research organizations and industry associations. This involves accessing the research organizations' databases and APIs to understand the latest industry trends. For example, the server downloads the latest production number data for the automobile industry from a research organization and stores it in a database.

[0694] Furthermore, the server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company. It uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators. For example, the server uses the Twitter API to collect tweets about Company A and quantifies negative / positive evaluations.

[0695] The server combines this bank information, industry information, and social media information to generate a dataset for input into the machine learning model. It also performs data preprocessing, filling in missing values ​​and standardizing them. For example, the server combines information such as Company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[0696] This integrated data set is input into a machine learning model to perform internal credit assessment and rating of the company, thereby predicting the company's credit score and future funding needs. As a specific example, the server uses the integrated data set of Company A to calculate Company A's credit score using the machine learning model.

[0697] The role of production management systems

[0698] The evaluation results and trend forecasts are provided to the production management system to help optimize the company's cash management and production plans. Company personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. This makes the company's production plans more efficient and optimizes resource allocation. In a specific example, a financial officer at Company A accesses the dashboard to view graphs of the company's credit score and the latest trend forecasts.

[0699] Trend forecast updates

[0700] Finally, the server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. For example, the server collects new industry data and social media information monthly and updates the trend predictions for Company A with the latest generative AI model.

[0701] Prompt Sentence Examples

[0702] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[0703] This allows companies to easily optimize their future capital needs and production plans, thereby optimizing economic resource allocation.

[0704] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0705] Step 1:

[0706] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. Specifically, it uses an API key or OAuth token to access the financial institution's API and obtains the company's financial data. This input data is then stored in a database. For example, the server obtains Company A's account balance data and stores it in the database along with regularly updated repayment history data.

[0707] Step 2:

[0708] The server collects industry information, such as production numbers, market share, and rankings, from relevant research organizations and industry associations. Specifically, it accesses research organization databases and APIs to obtain the latest industry trend data. This input data is later used as part of an integrated dataset. For example, the server downloads the latest production number data for the automotive industry from a research organization and stores it in a database.

[0709] Step 3:

[0710] The server uses the API of a social media platform to collect word-of-mouth and evaluation information about a company. Specifically, it uses the social media API to obtain text data and analyzes it using text mining technology. The input data are tweets and comments about the company, and the output data is a numerical representation of negative and positive evaluations. For example, the server uses the Twitter API to collect tweets about Company A and uses text mining technology to calculate the numerical value of positive evaluations.

[0711] Step 4:

[0712] The server integrates the bank information, industry information, and social media information mentioned above to generate a dataset to input into the machine learning model. Specifically, it preprocesses the data and performs missing value completion and standardization. The input data is each piece of information mentioned above, and the output data is the integrated dataset. For example, the server integrates information such as Company A's account balance, production volume, and social media ratings and processes it into a single dataset.

[0713] Step 5:

[0714] The server inputs the integrated dataset into a machine learning model to perform internal credit assessment and rating of the company. Specifically, the preprocessed data is input into the machine learning model to predict the company's credit score and future capital needs. The input data is the integrated dataset, and the output data is the company's credit score and trend prediction. For example, the server uses the integrated dataset of Company A to calculate Company A's credit score.

[0715] Step 6:

[0716] The server provides the evaluation results and trend forecasts to the production management system within the factory, supporting the optimization of production plans and financial management. Specifically, it accesses a dedicated dashboard terminal and displays the evaluation results and forecast data in graph and report format. The input data is the evaluation results and trend forecasts, and the output data is information displayed visually. For example, a financial officer at Company A accesses the dashboard to check information on the company's credit score and production plan optimization.

[0717] Step 7:

[0718] The server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. Specifically, new bank information, industry information, and social media information are collected monthly or weekly, and the new data is re-input into the machine learning model to retrain the model. The input data is the newly collected information, and the output data is the latest evaluation results and trend predictions. For example, the server collects the latest industry data and social media information and updates the trend predictions for Company A with a new generative AI model.

[0719] Prompt Sentence Examples

[0720] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[0721] 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.

[0722] This invention is a system for corporate loan services that collects and integrates bank information, industry information, and social media information, performs internal credit assessment and rating for companies using machine learning models, and further combines an emotion engine to recognize user emotions and customize evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising by linking the functions of the server, terminal, and user.

[0723] 1. Collection of banking information

[0724] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. The server authenticates with the financial institution using an API key or OAuth token, periodically retrieves the data, and stores it in a database.

[0725] Examples:

[0726] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[0727] 2. Collecting industry information

[0728] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[0729] Examples:

[0730] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[0731] 3. Collecting social media information

[0732] The server uses the API of the social media platform to collect word-of-mouth and evaluation information about companies, and then analyzes and quantifies the text data using text mining technology.

[0733] Examples:

[0734] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[0735] 4. Creation of an integrated dataset

[0736] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[0737] Examples:

[0738] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[0739] 5. Evaluation with machine learning models

[0740] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and credit rating. The machine learning model is pre-trained and calculates credit scores, etc.

[0741] Examples:

[0742] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[0743] 6. Emotion Recognition with Emotion Engine

[0744] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's voice and text inputs and recognizes their emotions to customize the display format of the evaluation results and trend predictions.

[0745] Examples:

[0746] The server uses an emotion engine to analyze the user's voice input and adjusts the display of the evaluation results based on the results.

[0747] 7. Checking the evaluation results

[0748] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The display format of the evaluation results is customized according to the user's emotional state.

[0749] Examples:

[0750] Company A's finance manager accesses a dashboard to view the company's credit score and latest trend forecasts in an emotional format.

[0751] 8. Funding Decisions

[0752] Users can develop fundraising strategies based on the provided evaluation results and market trends. The display format reflects the user's emotional state, allowing them to make more appropriate decisions.

[0753] Examples:

[0754] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[0755] 9. Trend forecast updates

[0756] The server regularly collects new data and retrains the machine learning models to keep the assessment results and trend predictions up to date, ensuring accurate and up-to-date information.

[0757] Examples:

[0758] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[0759] The system of this invention allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides advice and displays tailored to each user, improving user satisfaction and enabling more effective fundraising.

[0760] The processing flow will be explained below.

[0761] Step 1: Gather your banking information

[0762] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. During this process, the server first authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[0763] Specific behavior:

[0764] The server sends a request to an API endpoint to retrieve account balances and debit data based on the company ID.

[0765] The received data is analyzed in JSON format and the necessary data is saved in the database.

[0766] Step 2: Gather industry information

[0767] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, rankings, etc.

[0768] Specific behavior:

[0769] The server sends a request to the research organization's API to retrieve the latest industry data.

[0770] The acquired data is parsed, processed, and formatted before being saved in a database.

[0771] Step 3: Collect social media information

[0772] The server uses the APIs of various social media platforms to collect word-of-mouth and evaluation information related to companies and analyzes it using text mining technology.

[0773] Specific behavior:

[0774] The server queries social media APIs using specific keywords or hashtags to gather relevant posts.

[0775] The collected data is subjected to text analysis, and positive / negative evaluations are quantified and stored in a database.

[0776] Step 4: Creating a consolidated dataset

[0777] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[0778] Specific behavior:

[0779] The server merges information from different data sources and converts it into a unified format.

[0780] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[0781] Step 5: Evaluation with machine learning models

[0782] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[0783] Specific behavior:

[0784] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[0785] The scoring results are stored in a database and made available for subsequent processes.

[0786] Step 6: Emotion Recognition in the Emotion Engine

[0787] The server analyzes the user's voice and text input and uses an emotion engine to recognize emotions, thereby customizing the display and system responses to the user's emotional state.

[0788] Specific behavior:

[0789] The server passes the user's voice and text input to the emotion engine and obtains the analysis results.

[0790] Based on the analysis results, the dashboard display format and response content are adjusted.

[0791] Step 7: Check the evaluation results

[0792] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend predictions, and the display content is customized according to the user's emotional state.

[0793] Specific behavior:

[0794] Users log in to a dashboard to see their company's credit score and the latest trend forecasts.

[0795] Interact with interactive graphs and charts to get more information.

[0796] Step 8: Funding Decision

[0797] The system creates a fundraising strategy based on the evaluation results and market trends provided by the user. The system supports appropriate decisions by displaying the results in a format that responds to the user's emotions.

[0798] Specific behavior:

[0799] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[0800] If necessary, submit a new loan application through the online form.

[0801] Step 9: Update trend forecast

[0802] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[0803] Specific behavior:

[0804] The server recollects new financial, industry, and social media information monthly or weekly.

[0805] Retrain the machine learning model on the new dataset and update the evaluation results.

[0806] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides personalized advice and displays, improving user satisfaction and enabling more effective fundraising.

[0807] Example 2

[0808] 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."

[0809] Corporate lending services collect and integrate bank, industry, and social media information to assess a company's creditworthiness, but they often fail to accurately assess it due to insufficient consideration of data variability and emotions. Furthermore, the methods used to present assessment results and trend forecasts are unable to adapt to the user's emotions, sometimes resulting in insufficient support for user decision-making. This makes it difficult for financial institutions to make reliable lending decisions, making it difficult for companies to raise funds at reasonable interest rates.

[0810] 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.

[0811] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media; means for integrating the financial information, industry information, and SNS information and performing internal credit and rating of the company using a machine learning model; means for recognizing user emotions using an emotion engine and customizing the display format of the evaluation results and trend forecasts; and means for providing the evaluation results and trend forecasts using the machine learning model to the company. This enables internal credit and rating of companies to be performed with high accuracy, and by presenting evaluation results that take user emotions into consideration, it becomes possible to support more appropriate decision-making.

[0812] "Bank information" refers to public financial data obtained from financial institutions, such as a company's account balance, borrowing status, and repayment history.

[0813] "Industry information" refers to data about a particular industry, such as industry production volumes, market shares, and rankings.

[0814] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social media platforms.

[0815] A "machine learning model" is an algorithm used to generate internal credit ratings and ratings for companies based on collected data.

[0816] An "emotion engine" is a system that analyzes a user's voice and text input and recognizes emotions.

[0817] "User" refers to a person who uses this system to check the evaluation results and trend forecasts of a company and make a decision on fundraising.

[0818] A "database" is a data infrastructure for storing and managing collected financial information, industry information, and social media information.

[0819] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and suitability for lending.

[0820] A "rating" is a grade calculated using a machine learning model to indicate a company's creditworthiness and financial stability.

[0821] MODE FOR CARRYING OUT THE INVENTION

[0822] This invention is a system for corporate loan services that links the functions of servers, terminals, and users to collect and integrate corporate financial information, industry information, and social media information, and uses machine learning models and emotion engines to perform internal credit assessments and ratings for companies, providing evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising.

[0823] Collection of banking information

[0824] The server collects financial information about the company from partner financial institutions via API. Specifically, it obtains financial data such as account balances, borrowing status, and repayment history. The server authenticates with the financial institution using an API key or OAuth token, periodically obtains the data, and stores it in a database.

[0825] Examples:

[0826] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[0827] Collecting industry information

[0828] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[0829] Examples:

[0830] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[0831] Collecting SNS information

[0832] The server uses the API of the social media platform to collect reviews and ratings of companies, and analyzes them using text mining technology. The acquired text data is then analyzed and quantified.

[0833] Examples:

[0834] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[0835] Integrating data and using machine learning models

[0836] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data. The preprocessed integrated dataset is then input into the machine learning model to generate a company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[0837] Examples:

[0838] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[0839] Using the Emotion Engine

[0840] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's voice and text inputs and recognizes their emotions to customize the display format of the evaluation results and trend predictions.

[0841] Examples:

[0842] The server uses an emotion engine to analyze the user's voice input and adjusts the display of the evaluation results based on the results.

[0843] Reviewing evaluation results and making funding decisions

[0844] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The display format is customized according to the user's emotional state. Based on the evaluation results and market trends, it becomes possible to develop a fundraising strategy, apply for new loans, and raise funds at appropriate interest rates.

[0845] Examples:

[0846] Company A's finance manager accesses a dashboard to view the company's credit score and latest trend forecasts in an emotional format.

[0847] Based on the evaluation results, the financial officer will apply for new loans and raise funds at appropriate interest rates.

[0848] Trend forecast updates

[0849] The server periodically collects new data and retrains the machine learning model to keep the assessment results and trend predictions up to date.

[0850] Examples:

[0851] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[0852] This allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions.In addition, the emotion engine provides advice and displays tailored to each user, improving user satisfaction and enabling more effective fundraising.

[0853] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0854] System program processing flow

[0855] Step 1:

[0856] Collection of banking information

[0857] The server collects a company's financial information from affiliated financial institutions via API. It authenticates with the financial institution using an API key or OAuth token as input and obtains data such as account balance, borrowing status, and repayment history. The obtained financial information is received in JSON format, which the server parses and stores in a database.

[0858] Specific behavior:

[0859] 1. The server sends a request to the financial institution's API endpoint.

[0860] 2. Authenticate using an API key or OAuth token.

[0861] 3. Analyze the received JSON format financial information and extract each company's account balance, borrowing status, and repayment history.

[0862] 4. Save the analyzed data in a database.

[0863] Step 2:

[0864] Collecting industry information

[0865] The server collects industry information from relevant research institutes via API or FTP. It authenticates using API keys, usernames, and passwords as input, and obtains industry production figures, market shares, industry rankings, etc. The obtained industry information is analyzed and stored in a database.

[0866] Specific behavior:

[0867] 1. The server sends a request to the research organization's API endpoint.

[0868] 2. Authenticate if necessary (API key or username / password).

[0869] 3. Analyze the received industry information (e.g., production volume data) and extract the necessary information.

[0870] 4. Save the extracted data in the database.

[0871] Step 3:

[0872] Collecting SNS information

[0873] The server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company. It authenticates using the social media API key as input and obtains tweet data containing the company name and related keywords. It then analyzes the obtained tweet data using text mining technology, converting positive / negative evaluations into numerical values ​​and storing them in a database.

[0874] Specific behavior:

[0875] 1. The server sends a request to the Twitter API.

[0876] 2. Obtain tweets containing the company name and related keywords.

[0877] 3. Analyze tweet data using text mining technology and perform sentiment analysis.

[0878] 4. Quantify the negative / positive ratings and store them in a database.

[0879] Step 4:

[0880] Creating a unified dataset

[0881] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. It obtains each of the above information as input from a database, completes missing values, and normalizes the data. It then integrates the preprocessed data to generate a dataset for the machine learning model.

[0882] Specific behavior:

[0883] 1. The server retrieves the collected bank information, industry information, and social media ratings from the database.

[0884] 2. Impute missing values ​​(e.g., impute by the mean) and normalize the data.

[0885] 3. Integrate various information to generate a single data set.

[0886] Step 5:

[0887] Evaluation with machine learning models

[0888] The server inputs the preprocessed integrated dataset into the machine learning model to perform internal credit assessment and rating of the company. Using the integrated dataset as input, the machine learning model outputs the company's credit score and rating results. The calculation results are stored in a database.

[0889] Specific behavior:

[0890] 1. The server reads the consolidated data set.

[0891] 2. Input the integrated dataset into a machine learning model.

[0892] 3. Machine learning models calculate a company's credit score and rating.

[0893] 4. Save the calculation results in the database.

[0894] Step 6:

[0895] Emotion recognition with emotion engine

[0896] The server uses an emotion engine to recognize the user's emotions. It receives the user's voice and text input as input and analyzes their emotions using the emotion engine. It customizes the display format of the evaluation results and trend predictions according to the user's emotional state.

[0897] Specific behavior:

[0898] 1. The server receives the user's voice or text input.

[0899] 2. The emotion engine performs analysis to recognize the user's emotional state.

[0900] 3. Based on the recognition results, adjust the display format of the evaluation results and trend predictions.

[0901] Step 7:

[0902] Check the evaluation results

[0903] The user visually checks the evaluation results and trend predictions using a dedicated dashboard terminal. The terminal receives the evaluation results provided by the server as input and outputs the content in a customized display format according to the user's emotional state.

[0904] Specific behavior:

[0905] 1. The user logs in to the dashboard terminal.

[0906] 2. The server verifies the user's authentication and displays the dashboard content.

[0907] 3. The dashboard displays the latest assessment results and trend forecasts.

[0908] 4. The display format is customized according to the user's emotional state.

[0909] Step 8:

[0910] Funding decisions

[0911] The user creates a fundraising strategy based on the provided evaluation results and market trends. Using the information confirmed on the dashboard as input, the user submits a new loan application and outputs the details of raising funds at appropriate interest rates.

[0912] Specific behavior:

[0913] 1. The user checks the evaluation results displayed on the dashboard.

[0914] 2. Consider financing scenarios based on market trends.

[0915] 3. If necessary, file a new loan application.

[0916] 4. Appropriate interest rates will be set based on the loan application.

[0917] Step 9:

[0918] Trend forecast updates

[0919] The server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. It collects new financial, industry, and social media information as input and retrains the machine learning model using the updated dataset. It saves the latest prediction results in the database and updates the information provided to users.

[0920] Specific behavior:

[0921] 1. The server periodically collects new financial, industry, and social media information.

[0922] 2. Update the integrated dataset based on the collected data.

[0923] 3. Retrain the machine learning model (update the generative AI model).

[0924] 4. Store the latest evaluation results and trend forecasts in a database.

[0925] (Application example 2)

[0926] 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."

[0927] Conventional corporate lending services collect and analyze financial information, industry information, and social media information separately, making it difficult to provide comprehensive, real-time credit assessments. Furthermore, they do not take into account the user's emotional state, meaning that assessment results and forecast information are not optimized to fit the user's situation. This makes it difficult to make accurate lending decisions and manage transaction risks.

[0928] The specific processing by the specific 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 financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; and means for collecting word-of-mouth and evaluation information about companies from various social networking sites. This enables financial information, industry information, and social networking site information to be integrated and a machine learning model to perform internal credit assessment and rating of companies. Furthermore, by including an emotion engine that recognizes user emotions and customizes the display format of evaluation results and trend forecasts, it is possible to provide information tailored to the user. Furthermore, by having the machine learning model perform credit assessment of each transaction in real time and periodically retrain, the assessment results and trend forecasts are always kept up to date, enabling highly accurate lending decisions and transaction risk management.

[0929] "Bank information" refers to financial information such as a company's account balance, borrowing status, repayment history, etc. This information is necessary to evaluate a company's financial condition.

[0930] "Industry information" is data that includes production volume, market share, rankings, etc. in a specific industry. This information is necessary to understand the overall trends in the industry to which a company belongs.

[0931] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social networking services. This information is used to understand the evaluations and feelings of the general public and customers toward companies.

[0932] A "machine learning model" is an algorithm that performs internal credit assessment and rating of a company based on collected data, making it possible to evaluate a company's credit score and risk.

[0933] The "emotion engine" is a system that recognizes the user's emotions. It analyzes the user's voice and text input and customizes the display format of evaluation results and trend predictions according to the user's emotional state.

[0934] "Evaluation results" are the credit scores and internal credit results of companies obtained using machine learning models, which are used to determine the transaction risk and lending decisions for companies.

[0935] "Trend forecasts" are information used to predict a company's future trends. They use collected data to predict a company's future risks and growth potential.

[0936] "Real-time evaluation" refers to a credit evaluation that is conducted in real time for each transaction, allowing for accurate evaluation based on the latest data at all times.

[0937] "Retraining" is the process of updating a machine learning model based on newly collected data, ensuring that its assessment results and trend predictions are always up-to-date.

[0938] A system for implementing the present invention operates in cooperation with multiple components including a server, a user terminal, a machine learning model, and an emotion engine.

[0939] Data collection

[0940] First, the server collects financial information such as a company's account balance, borrowing status, and repayment history from the bank via API. The server authenticates with the financial institution using an API key or OAuth token, periodically obtains the latest data, and stores it in a database. Industry information, such as production volume, market share, and industry rankings, is also collected from related research institutions via API or FTP. Furthermore, social media information, such as reviews and ratings of companies, is collected using the API of social media platforms, and text data is analyzed and quantified using text mining technology.

[0941] Data Integration and Evaluation

[0942] The server integrates financial, industry, and social media information and preprocesses the data. Specifically, it completes missing values ​​and normalizes them to generate an input dataset for the machine learning model. The generated dataset is then fed into a pre-trained machine learning model to generate a company's internal credit rating and credit rating. The machine learning model is built using machine learning libraries such as TensorFlow and scikit-learn.

[0943] Emotion Recognition and Customization

[0944] The server uses an emotion engine to analyze the user's voice and text input. This emotion engine uses natural language processing libraries such as TextBlob to evaluate the user's emotions with a positive / negative score. The evaluation results and trend predictions are displayed in a customized format according to the user's emotional state.

[0945] Displaying results and making funding decisions

[0946] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The evaluation results are displayed in a format that is customized according to the user's emotional state, allowing the user to make more appropriate decisions. For example, a company's financial officer can look at the evaluation results and apply for new loans, ensuring that funds are raised at appropriate interest rates.

[0947] Specific examples of programs

[0948] A specific example of the embodiment is as follows:

[0949] When a company's financial officer initiates a new transaction, the server collects the company's latest financial, industry, and social media information, and inputs the integrated data into a machine learning model to calculate a credit score. At this point, an emotion engine analyzes the officer's emotional state and displays the evaluation results in an optimal format. This allows the financial officer to make appropriate financing decisions based on the evaluation results, such as avoiding high-risk transactions.

[0950] Example prompt sentence:

[0951] When doing new business, I want to see a company's current credit score and market trends, and if my feelings indicate concern, I want the system to provide feedback on its risk assessment as well.

[0952] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0953] Step 1:

[0954] The server collects financial information such as a company's account balance, borrowing status, and repayment history from financial institutions via API. At this time, the server authenticates with the financial institution using an API key or OAuth token, and periodically retrieves the latest financial information into a database. The input is financial information obtained from the financial institution's API, and the output is financial information stored in the server's database. By collecting financial information, data can be obtained to understand the company's current financial situation.

[0955] Step 2:

[0956] The server collects industry information from related research institutes via API or FTP. Specifically, it obtains data such as production volume, market share, and industry rankings. The input is industry information obtained from the research institute's API, and the output is industry information stored in the server's database. By collecting industry information, data can be obtained to understand trends in the industry to which a company belongs.

[0957] Step 3:

[0958] The server uses the API of the SNS platform to collect word-of-mouth and evaluation information about the company. The acquired text data is analyzed using text mining technology and quantified as positive / negative evaluations. The input is text data obtained from the SNS API, and the output is quantified SNS evaluation information. By collecting and analyzing SNS information, data can be obtained that visualizes the evaluation of the company by the general public and customers.

[0959] Step 4:

[0960] The server integrates the collected financial, industry, and social media information and preprocesses this data. Specifically, it imputes missing values ​​and normalizes the data to generate an input dataset for the machine learning model. The input is financial, industry, and social media information, and the output is a preprocessed integrated dataset. Preprocessing the data prepares the machine learning model for optimal operation.

[0961] Step 5:

[0962] The server inputs the preprocessed integrated dataset into a machine learning model to perform internal credit assessment and rating of the company. The input is the preprocessed integrated dataset, and the output is the company's credit score and risk assessment results. Using a machine learning model for evaluation makes it possible to quantitatively evaluate the company's credit status.

[0963] Step 6:

[0964] The server uses an emotion engine to analyze the user's voice and text input. Specifically, it uses a natural language processing library such as TextBlob to analyze emotions and obtains the results as a score. The input is the user's voice or text data, and the output is an emotion score. Using the emotion engine, it is possible to accurately recognize the user's emotions and provide feedback accordingly.

[0965] Step 7:

[0966] The server provides users with the evaluation results and trend predictions obtained from the machine learning model. At this time, the display format of the evaluation results is customized according to the user's emotional score. The inputs are the credit score, risk evaluation results, and emotional score, and the output is a display of the customized evaluation results. Users can visually check these results using a dashboard terminal, enabling them to make more appropriate decisions.

[0967] Step 8:

[0968] Users determine transaction risks and funding strategies based on the provided evaluation results and trend forecasts. This allows them to make economic decisions such as procuring funds at appropriate interest rates or avoiding high-risk transactions. The input is customized evaluation results, and the output is specific funding decisions. Evaluation displays based on the user's emotional state enable appropriate decisions.

[0969] Step 9:

[0970] The server periodically collects new data and retrains the machine learning model, allowing it to always provide the latest evaluation results and trend forecasts. The input is newly collected financial, industry, and social media information, and the output is an updated machine learning model and evaluation results. Periodic retraining improves the accuracy of the evaluation, enabling the provision of highly reliable information.

[0971] 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.

[0972] 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.

[0973] 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.

[0974] [Third embodiment]

[0975] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0976] 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.

[0977] 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).

[0978] 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.

[0979] 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.

[0980] 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).

[0981] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0982] 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.

[0983] 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.

[0984] 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.

[0985] 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.

[0986] 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."

[0987] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[0988] 1. The server collects your banking information

[0989] First, the server collects financial information such as a company's account balance, loan status, and repayment history from partner financial institutions via API. After going through authentication procedures, the necessary data is periodically retrieved and stored in a database. Security protection using API keys and OAuth tokens is important during this process.

[0990] Examples:

[0991] The server obtains the account balance of Company A and periodically stores updated repayment history data in a database.

[0992] 2. The server collects industry information

[0993] Next, the server collects industry information such as production numbers, market shares, rankings, etc. from relevant research organizations and industry associations, including accessing research organizations' databases and APIs to understand the latest industry trends.

[0994] Examples:

[0995] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[0996] 3. The server collects social media information

[0997] The server uses the API of the social media platform to collect reviews and ratings of companies, and then uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators.

[0998] Examples:

[0999] The server uses the Twitter API to collect tweets about Company A and quantify the negative / positive ratings.

[1000] 4. The server creates the integrated data set

[1001] The server integrates bank information, industry information, and social media information to generate a dataset for input into the machine learning model, and then preprocesses the data, imputing missing values ​​and standardizing them.

[1002] Examples:

[1003] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[1004] 5. The server evaluates the results using a machine learning model.

[1005] The server inputs the integrated data set into a machine learning model to generate internal credit ratings for businesses, which then predict their credit scores and future financing needs.

[1006] Examples:

[1007] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[1008] 6. The user checks the evaluation results

[1009] Corporate personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. The evaluation results are displayed visually, making them easy for users to understand.

[1010] Examples:

[1011] Company A's financial officer accesses the dashboard and views graphs of the company's credit score and the latest trend forecasts.

[1012] 7. Users make funding decisions

[1013] The company's finance department will consider future fundraising strategies based on the provided evaluation results and market trends, thereby achieving efficient and low-risk fundraising.

[1014] Examples:

[1015] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[1016] 8. The server updates the trend forecast

[1017] The server regularly collects new data and retrains the machine learning model to keep the assessment results and trend predictions up to date, ensuring that both companies and financial institutions are always provided with accurate and up-to-date information.

[1018] Examples:

[1019] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[1020] The system of this invention will enable companies to more easily predict future funding needs and interest rate trends, and financial institutions to make more accurate lending decisions. This is expected to strengthen mutual trust and contribute to stable economic growth.

[1021] The processing flow will be explained below.

[1022] Step 1: Gather your banking information

[1023] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. Before this process begins, the server authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[1024] Specific behavior:

[1025] The server sends a request to an API endpoint to retrieve account balances and debit data filtered by company ID.

[1026] The received data is analyzed in JSON format and the necessary data is saved in the database.

[1027] Step 2: Gather industry information

[1028] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[1029] Specific behavior:

[1030] The server sends a request to the research organization's API to retrieve the latest industry data.

[1031] The acquired data is parsed, processed, and formatted before being saved in a database.

[1032] Step 3: Collect social media information

[1033] The server uses the APIs of various social media platforms to collect reviews and ratings related to companies, and then uses text mining technology to analyze and quantify these reviews.

[1034] Specific behavior:

[1035] The server queries social media APIs using specific keywords or hashtags to gather relevant posts.

[1036] The collected data is subjected to text analysis, positive / negative indicators are calculated, and the evaluation information is quantified and stored in a database.

[1037] Step 4: Creating a consolidated dataset

[1038] The server integrates bank information, industry information, and social media information to create an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[1039] Specific behavior:

[1040] The server merges information from different data sources and converts it into a unified format.

[1041] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[1042] Step 5: Evaluation with machine learning models

[1043] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1044] Specific behavior:

[1045] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[1046] The scoring results are stored in a database and made available for subsequent processes.

[1047] Step 6: Check the evaluation results

[1048] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend forecasts, which are displayed in graphs and charts.

[1049] Specific behavior:

[1050] The user logs in to the dashboard and checks their company's evaluation results.

[1051] Interact with interactive graphs and charts to get more information.

[1052] Step 7: Funding Decision

[1053] The system develops a funding strategy based on the evaluation results and market trends provided by the user. The evaluation results include forecasts of future funding needs and interest rate trends.

[1054] Specific behavior:

[1055] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[1056] If necessary, submit a new loan application through the online form.

[1057] Step 8: Update trend forecasts

[1058] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[1059] Specific behavior:

[1060] The server recollects new financial, industry, and social media information monthly or weekly.

[1061] Retrain the machine learning model on the new dataset and update the evaluation results.

[1062] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions.

[1063] Example 1

[1064] 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."

[1065] In conventional corporate loan service systems, information collection and analysis from individual data sources was fragmented, making it difficult to conduct comprehensive corporate evaluations. Furthermore, data updates were delayed or intermittent, making it impossible to provide the latest evaluation results and trend forecasts. These issues hindered appropriate credit evaluations of companies and accurate lending decisions by financial institutions.

[1066] 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.

[1067] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from social media platforms; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on companies; means for preprocessing the integrated dataset, completing missing values, and standardizing data; means for providing evaluation results and trend forecasts to companies using the machine learning model; and means for visually displaying the evaluation results and trend forecasts using a dedicated dashboard. This makes it possible to always provide accurate and comprehensive corporate evaluations and the latest trend forecasts.

[1068] "Financial information" refers to data on a company's financial status collected from banks, such as the company's account balance, borrowing status, and repayment history.

[1069] "Industry information" refers to trends and statistical data about a specific industry, such as production numbers, market share, and rankings, and is collected from research organizations and industry associations.

[1070] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various SNS platforms, and is converted into evaluation indicators using text mining technology.

[1071] A "machine learning model" is an algorithm that performs internal credit assessments and ratings on companies based on collected data, and is used to provide evaluation results and trend predictions.

[1072] An "integrated dataset" is a single dataset that integrates financial information, industry information, and social media information, and is used as input data for machine learning models.

[1073] "Preprocessing" refers to the process of formatting input data for a machine learning model, and includes tasks such as filling in missing values ​​and standardizing data.

[1074] "Evaluation results" refer to a company's credit score or internal credit calculated by a machine learning model, and are part of the information provided.

[1075] "Trend forecasts" refer to the results of machine learning models predicting a company's future capital needs and market trends.

[1076] The "dedicated dashboard" is an interface that allows users to visually check evaluation results and trend forecasts, and displays information in graph and table format.

[1077] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[1078] Server Roles and Functions

[1079] In this invention, the server collects financial information, industry information, and social media information of companies, and integrates them to generate a dataset to be input into the machine learning model. Specifically, the following hardware and software are used:

[1080] Hardware: High-performance server (e.g., with Xeon processor)

[1081] Software: API access libraries (e.g., requests), database software (e.g., MySQL, PostgreSQL), data processing libraries (e.g., Pandas, NumPy), machine learning libraries (e.g., Scikit-learn, TensorFlow)

[1082] The server processes and collects information as follows:

[1083] 1. Collection of financial information: The server obtains financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via the API, after going through authentication procedures, and stores the information in a database.

[1084] 2. Collecting industry information: The server accesses databases and APIs of research organizations and industry associations, downloads industry information such as production numbers, market shares, and rankings, and stores it in a database.

[1085] 3. Collection of social media information: The server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company, analyzes it using text mining technology, converts it into quantitative evaluation indicators, and stores them in a database.

[1086] 4. Generating an integrated dataset: Integrating bank information, industry information, and social media information, and preprocessing the data. Specifically, we impute missing values, normalize the data, and convert it into a format suitable for machine learning models.

[1087] Specific examples

[1088] For example, the server retrieves Company A's account balance and repayment history data via API and stores it in a MySQL database. Next, it downloads the latest production volume data for the automobile industry from a research institute and imports it into the database in CSV format using Pandas. Furthermore, it retrieves tweets related to Company A using the Twitter API, performs sentiment analysis using NLTK, and stores the results in a PostgreSQL database.

[1089] Roles and functions of user devices

[1090] The user terminal provides an interface that allows company personnel and financial institution analysts to access a dedicated dashboard and check evaluation results and trend forecasts. The following hardware and software are used.

[1091] Hardware: A standard computer or tablet

[1092] Software: Web browser (e.g., Google Chrome, Mozilla Firefox), dashboard software (e.g., Django, Plotly)

[1093] After logging in, users can access the dashboard and visually check the evaluation results and trend forecasts. The results are displayed in graph and table format, making them easy to understand.

[1094] Specific examples

[1095] A financial officer at Company A opens a web browser, accesses the dashboard URL, and enters his or her login information for authentication. The dashboard then retrieves the latest valuation results from the database and uses Plotly to generate and display graphs.

[1096] Prompt Sentence Examples

[1097] For example, if you want to use a generative AI model to predict Company A's credit score and future financing needs, you might use a prompt like this:

[1098] "To predict Company A's latest credit score and future financing needs, use a dataset that combines up-to-date banking, industry, and social media information."

[1099] The system of the present invention provides accurate and up-to-date information to companies and financial institutions, and is an important tool for achieving efficient and low-risk financing. It is implemented using a specific combination of hardware and software, and users can easily check the evaluation results on a dedicated dashboard.

[1100] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1101] Step 1:

[1102] Server collects banking information

[1103] Input: Financial institution API authentication information, company identification information

[1104] Specific operation: The server uses the API key and OAuth token to connect to the financial institution's API and obtain financial information such as the company's account balance, borrowing status, repayment history, etc. The server sends an HTTP request and receives a response in JSON format.

[1105] Data processing: Parse the received JSON data, extract the necessary information, and save the extracted data in the database.

[1106] Output: A database containing financial information about the company.

[1107] Step 2:

[1108] The server collects industry information

[1109] Input: API endpoint and authentication information of research institute or industry association

[1110] Specific operation: The server sends an HTTP request to the research organization's API to obtain industry information such as production numbers, market share, and rankings.

[1111] Data processing: The acquired data is received in CSV format, parsed using the Pandas library, and the necessary information is extracted. The extracted data is then saved in a database.

[1112] Output: A database containing up-to-date industry information.

[1113] Step 3:

[1114] The server collects SNS information

[1115] Input: API endpoints of social media platforms, keywords and hashtags related to your business

[1116] Specific operation: The server connects to the API of the social media platform and collects reviews and ratings information about the company.

[1117] Data processing: The collected text data is analyzed using text mining techniques such as the Natural Language Toolkit (NLTK) and classified into negative and positive evaluations. The classification results are stored in a database.

[1118] Output: A database containing social media reputation data related to companies.

[1119] Step 4:

[1120] The server creates a consolidated dataset

[1121] Input: Database containing financial information, industry information, and social media information

[1122] Specific operation: The server extracts the necessary information from the database and integrates it into a Pandas DataFrame. It imputes missing values ​​using SimpleImputer, normalizes the data using MinMaxScaler, and performs OneHot encoding on categorical data.

[1123] Data processing: Integrate various information and generate a pre-processed dataset.

[1124] Output: A consolidated dataset for machine learning model input.

[1125] Step 5:

[1126] The server evaluates using a machine learning model

[1127] Input: Integrated dataset

[1128] How it works: The server inputs the integrated dataset into the TensorFlow model to perform internal credit assessment and rating for the company. The model processes the data and calculates the company's credit score.

[1129] Data Computing: Predict credit scores and future financing needs based on integrated data sets.

[1130] Output: Company rating results and credit score.

[1131] Step 6:

[1132] The user checks the evaluation results

[1133] Input: Database containing evaluation results, user authentication information

[1134] How it works: Users access a dedicated dashboard and enter their login information to be authenticated. The dashboard retrieves the evaluation results from the database and generates and displays graphs using Plotly.

[1135] Data calculation: Checks access rights through user authentication and visualizes the evaluation results.

[1136] Output: Visualized evaluation results and trend forecasts.

[1137] Step 7:

[1138] Users make funding decisions

[1139] Input: Visualized evaluation results and trend forecasts

[1140] Specific operation: The user (company's finance department) considers fundraising strategies based on the information in the dashboard.

[1141] Data calculation: Based on the provided evaluation results and trend forecasts, companies can make new loan applications.

[1142] Output: New loan application, interest rate terms set.

[1143] Step 8:

[1144] Server updates trend forecast

[1145] Input: API endpoints to collect new information, existing datasets

[1146] Specific operation: The server periodically collects new data (repeating steps 1 to 3). It uses the new data to retrain the machine learning model and generate the latest evaluation results.

[1147] Data computation: Retraining the model with new data and updating the evaluation results.

[1148] Output: A database containing the latest assessment results and trend forecasts.

[1149] (Application example 1)

[1150] 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."

[1151] Traditional corporate internal credit and rating systems are limited to evaluations based on financial, industry, and social media information, and do not address the specific business operations and efficiency of each company's production plans. As a result, when companies seek financing, they face the dual challenges of fundraising and production management, making it difficult to allocate resources effectively. In particular, optimizing capital management and production plans is a key issue for in-factory production management systems.

[1152] 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.

[1153] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media sites; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on the company; means for providing the evaluation results and trend forecasts to an in-factory production management system to support the optimization of production plans and cash management; and means for regularly retraining the machine learning model to ensure the latest evaluation results and trend forecasts. This enables companies to comprehensively utilize financial information, industry information, and social media information to allocate resources economically, thereby optimizing in-factory production plans and streamlining cash management.

[1154] "Financial information" refers to information collected from banks, including a company's account balance, borrowing status, repayment history, etc.

[1155] "Industry information" refers to data on industry trends, such as production volume, market share, and rankings, obtained from research organizations and industry associations.

[1156] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social networking services.

[1157] A "machine learning model" is an algorithm that uses collected data to assign internal credit and ratings to companies.

[1158] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and financial condition.

[1159] A "rating" is a method of quantitatively evaluating and ranking a company's creditworthiness.

[1160] "Evaluation results" refer to a company's credit score and trend predictions calculated using a machine learning model.

[1161] "Trend forecasting" refers to predicting future corporate credit status and market trends based on machine learning models.

[1162] A "production management system" is a system for managing production plans and resource allocation within a factory.

[1163] "Production planning" refers to planning for production schedules and resource allocation within a factory.

[1164] "Cash management" is the process of optimally managing working capital within a factory.

[1165] This invention is a system that collects and integrates bank information, industry information, and social media information, and uses machine learning models to perform internal credit and rating for companies. This system improves the efficiency of corporate cash management and production planning by linking the functions of the server, factory production management system, and user systems.

[1166] Server Roles

[1167] The server first collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. After going through authentication procedures, the necessary data is periodically obtained and stored in a database. Security protection using API keys and OAuth tokens is important during this process. As a concrete example, the server obtains Company A's account balance and periodically updates the repayment history data and stores it in a database.

[1168] Next, the server collects industry information such as production numbers, market shares, and rankings from relevant research organizations and industry associations. This involves accessing the research organizations' databases and APIs to understand the latest industry trends. For example, the server downloads the latest production number data for the automobile industry from a research organization and stores it in a database.

[1169] Furthermore, the server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company. It uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators. For example, the server uses the Twitter API to collect tweets about Company A and quantifies negative / positive evaluations.

[1170] The server combines this bank information, industry information, and social media information to generate a dataset for input into the machine learning model. It also performs data preprocessing, filling in missing values ​​and standardizing them. For example, the server combines information such as Company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[1171] This integrated data set is input into a machine learning model to perform internal credit assessment and rating of the company, thereby predicting the company's credit score and future funding needs. As a specific example, the server uses the integrated data set of Company A to calculate Company A's credit score using the machine learning model.

[1172] The role of production management systems

[1173] The evaluation results and trend forecasts are provided to the production management system to help optimize the company's cash management and production plans. Company personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. This makes the company's production plans more efficient and optimizes resource allocation. In a specific example, a financial officer at Company A accesses the dashboard to view graphs of the company's credit score and the latest trend forecasts.

[1174] Trend forecast updates

[1175] Finally, the server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. For example, the server collects new industry data and social media information monthly and updates the trend predictions for Company A with the latest generative AI model.

[1176] Prompt Sentence Examples

[1177] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[1178] This allows companies to easily optimize their future capital needs and production plans, thereby optimizing economic resource allocation.

[1179] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1180] Step 1:

[1181] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. Specifically, it uses an API key or OAuth token to access the financial institution's API and obtains the company's financial data. This input data is then stored in a database. For example, the server obtains Company A's account balance data and stores it in the database along with regularly updated repayment history data.

[1182] Step 2:

[1183] The server collects industry information, such as production numbers, market share, and rankings, from relevant research organizations and industry associations. Specifically, it accesses research organization databases and APIs to obtain the latest industry trend data. This input data is later used as part of an integrated dataset. For example, the server downloads the latest production number data for the automotive industry from a research organization and stores it in a database.

[1184] Step 3:

[1185] The server uses the API of a social media platform to collect word-of-mouth and evaluation information about a company. Specifically, it uses the social media API to obtain text data and analyzes it using text mining technology. The input data are tweets and comments about the company, and the output data is a numerical representation of negative and positive evaluations. For example, the server uses the Twitter API to collect tweets about Company A and uses text mining technology to calculate the numerical value of positive evaluations.

[1186] Step 4:

[1187] The server integrates the bank information, industry information, and social media information mentioned above to generate a dataset to input into the machine learning model. Specifically, it preprocesses the data and performs missing value completion and standardization. The input data is each piece of information mentioned above, and the output data is the integrated dataset. For example, the server integrates information such as Company A's account balance, production volume, and social media ratings and processes it into a single dataset.

[1188] Step 5:

[1189] The server inputs the integrated dataset into a machine learning model to perform internal credit assessment and rating of the company. Specifically, the preprocessed data is input into the machine learning model to predict the company's credit score and future capital needs. The input data is the integrated dataset, and the output data is the company's credit score and trend prediction. For example, the server uses the integrated dataset of Company A to calculate Company A's credit score.

[1190] Step 6:

[1191] The server provides the evaluation results and trend forecasts to the production management system within the factory, supporting the optimization of production plans and financial management. Specifically, it accesses a dedicated dashboard terminal and displays the evaluation results and forecast data in graph and report format. The input data is the evaluation results and trend forecasts, and the output data is information displayed visually. For example, a financial officer at Company A accesses the dashboard to check information on the company's credit score and production plan optimization.

[1192] Step 7:

[1193] The server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. Specifically, new bank information, industry information, and social media information are collected monthly or weekly, and the new data is re-input into the machine learning model to retrain the model. The input data is the newly collected information, and the output data is the latest evaluation results and trend predictions. For example, the server collects the latest industry data and social media information and updates the trend predictions for Company A with a new generative AI model.

[1194] Prompt Sentence Examples

[1195] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[1196] 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.

[1197] This invention is a system for corporate loan services that collects and integrates bank information, industry information, and social media information, performs internal credit assessment and rating for companies using machine learning models, and further combines an emotion engine to recognize user emotions and customize evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising by linking the functions of the server, terminal, and user.

[1198] 1. Collection of banking information

[1199] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. The server authenticates with the financial institution using an API key or OAuth token, periodically retrieves the data, and stores it in a database.

[1200] Examples:

[1201] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[1202] 2. Collecting industry information

[1203] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[1204] Examples:

[1205] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[1206] 3. Collecting social media information

[1207] The server uses the API of the social media platform to collect word-of-mouth and evaluation information about companies, and then analyzes and quantifies the text data using text mining technology.

[1208] Examples:

[1209] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[1210] 4. Creation of an integrated dataset

[1211] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[1212] Examples:

[1213] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[1214] 5. Evaluation with machine learning models

[1215] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and credit rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1216] Examples:

[1217] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[1218] 6. Emotion Recognition with Emotion Engine

[1219] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's voice and text inputs and recognizes their emotions to customize the display format of the evaluation results and trend predictions.

[1220] Examples:

[1221] The server uses an emotion engine to analyze the user's voice input and adjusts the display of the evaluation results based on the results.

[1222] 7. Checking the evaluation results

[1223] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The display format of the evaluation results is customized according to the user's emotional state.

[1224] Examples:

[1225] Company A's finance manager accesses a dashboard to view the company's credit score and latest trend forecasts in an emotional format.

[1226] 8. Funding Decisions

[1227] Users can develop fundraising strategies based on the provided evaluation results and market trends. The display format reflects the user's emotional state, allowing them to make more appropriate decisions.

[1228] Examples:

[1229] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[1230] 9. Trend forecast updates

[1231] The server regularly collects new data and retrains the machine learning models to keep the assessment results and trend predictions up to date, ensuring accurate and up-to-date information.

[1232] Examples:

[1233] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[1234] The system of this invention allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides advice and displays tailored to each user, improving user satisfaction and enabling more effective fundraising.

[1235] The processing flow will be explained below.

[1236] Step 1: Gather your banking information

[1237] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. During this process, the server first authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[1238] Specific behavior:

[1239] The server sends a request to an API endpoint to retrieve account balances and debit data based on the company ID.

[1240] The received data is analyzed in JSON format and the necessary data is saved in the database.

[1241] Step 2: Gather industry information

[1242] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, rankings, etc.

[1243] Specific behavior:

[1244] The server sends a request to the research organization's API to retrieve the latest industry data.

[1245] The acquired data is parsed, processed, and formatted before being saved in a database.

[1246] Step 3: Collect social media information

[1247] The server uses the APIs of various social media platforms to collect word-of-mouth and evaluation information related to companies and analyzes it using text mining technology.

[1248] Specific behavior:

[1249] The server queries social media APIs using specific keywords or hashtags to gather relevant posts.

[1250] The collected data is subjected to text analysis, and positive / negative evaluations are quantified and stored in a database.

[1251] Step 4: Creating a consolidated dataset

[1252] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[1253] Specific behavior:

[1254] The server merges information from different data sources and converts it into a unified format.

[1255] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[1256] Step 5: Evaluation with machine learning models

[1257] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1258] Specific behavior:

[1259] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[1260] The scoring results are stored in a database and made available for subsequent processes.

[1261] Step 6: Emotion Recognition in the Emotion Engine

[1262] The server analyzes the user's voice and text input and uses an emotion engine to recognize emotions, thereby customizing the display and system responses to the user's emotional state.

[1263] Specific behavior:

[1264] The server passes the user's voice and text input to the emotion engine and obtains the analysis results.

[1265] Based on the analysis results, the dashboard display format and response content are adjusted.

[1266] Step 7: Check the evaluation results

[1267] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend predictions, and the display content is customized according to the user's emotional state.

[1268] Specific behavior:

[1269] Users log in to a dashboard to see their company's credit score and the latest trend forecasts.

[1270] Interact with interactive graphs and charts to get more information.

[1271] Step 8: Funding Decision

[1272] The system creates a fundraising strategy based on the evaluation results and market trends provided by the user. The system supports appropriate decisions by displaying the results in a format that responds to the user's emotions.

[1273] Specific behavior:

[1274] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[1275] If necessary, submit a new loan application through the online form.

[1276] Step 9: Update trend forecast

[1277] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[1278] Specific behavior:

[1279] The server recollects new financial, industry, and social media information monthly or weekly.

[1280] Retrain the machine learning model on the new dataset and update the evaluation results.

[1281] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides personalized advice and displays, improving user satisfaction and enabling more effective fundraising.

[1282] Example 2

[1283] 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."

[1284] Corporate lending services collect and integrate bank, industry, and social media information to assess a company's creditworthiness, but they often fail to accurately assess it due to insufficient consideration of data variability and emotions. Furthermore, the methods used to present assessment results and trend forecasts are unable to adapt to the user's emotions, sometimes resulting in insufficient support for user decision-making. This makes it difficult for financial institutions to make reliable lending decisions, making it difficult for companies to raise funds at reasonable interest rates.

[1285] 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.

[1286] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media; means for integrating the financial information, industry information, and SNS information and performing internal credit and rating of the company using a machine learning model; means for recognizing user emotions using an emotion engine and customizing the display format of the evaluation results and trend forecasts; and means for providing the evaluation results and trend forecasts using the machine learning model to the company. This enables internal credit and rating of companies to be performed with high accuracy, and by presenting evaluation results that take user emotions into consideration, it becomes possible to support more appropriate decision-making.

[1287] "Bank information" refers to public financial data obtained from financial institutions, such as a company's account balance, borrowing status, and repayment history.

[1288] "Industry information" refers to data about a particular industry, such as industry production volumes, market shares, and rankings.

[1289] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social media platforms.

[1290] A "machine learning model" is an algorithm used to generate internal credit ratings and ratings for companies based on collected data.

[1291] An "emotion engine" is a system that analyzes a user's voice and text input and recognizes emotions.

[1292] "User" refers to a person who uses this system to check the evaluation results and trend forecasts of a company and make a decision on fundraising.

[1293] A "database" is a data infrastructure for storing and managing collected financial information, industry information, and social media information.

[1294] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and suitability for lending.

[1295] A "rating" is a grade calculated using a machine learning model to indicate a company's creditworthiness and financial stability.

[1296] MODE FOR CARRYING OUT THE INVENTION

[1297] This invention is a system for corporate loan services that links the functions of servers, terminals, and users to collect and integrate corporate financial information, industry information, and social media information, and uses machine learning models and emotion engines to perform internal credit assessments and ratings for companies, providing evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising.

[1298] Collection of banking information

[1299] The server collects financial information about the company from partner financial institutions via API. Specifically, it obtains financial data such as account balances, borrowing status, and repayment history. The server authenticates with the financial institution using an API key or OAuth token, periodically obtains the data, and stores it in a database.

[1300] Examples:

[1301] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[1302] Collecting industry information

[1303] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[1304] Examples:

[1305] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[1306] Collecting SNS information

[1307] The server uses the API of the social media platform to collect reviews and ratings of companies, and analyzes them using text mining technology. The acquired text data is then analyzed and quantified.

[1308] Examples:

[1309] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[1310] Integrating data and using machine learning models

[1311] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data. The preprocessed integrated dataset is then input into the machine learning model to generate a company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1312] Examples:

[1313] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[1314] Using the Emotion Engine

[1315] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's voice and text inputs and recognizes their emotions to customize the display format of the evaluation results and trend predictions.

[1316] Examples:

[1317] The server uses an emotion engine to analyze the user's voice input and adjusts the display of the evaluation results based on the results.

[1318] Reviewing evaluation results and making funding decisions

[1319] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The display format is customized according to the user's emotional state. Based on the evaluation results and market trends, it becomes possible to develop a fundraising strategy, apply for new loans, and raise funds at appropriate interest rates.

[1320] Examples:

[1321] Company A's finance manager accesses a dashboard to view the company's credit score and latest trend forecasts in an emotional format.

[1322] Based on the evaluation results, the financial officer will apply for new loans and raise funds at appropriate interest rates.

[1323] Trend forecast updates

[1324] The server periodically collects new data and retrains the machine learning model to keep the assessment results and trend predictions up to date.

[1325] Examples:

[1326] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[1327] This allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions.In addition, the emotion engine provides advice and displays tailored to each user, improving user satisfaction and enabling more effective fundraising.

[1328] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1329] System program processing flow

[1330] Step 1:

[1331] Collection of banking information

[1332] The server collects a company's financial information from affiliated financial institutions via API. It authenticates with the financial institution using an API key or OAuth token as input and obtains data such as account balance, borrowing status, and repayment history. The obtained financial information is received in JSON format, which the server parses and stores in a database.

[1333] Specific behavior:

[1334] 1. The server sends a request to the financial institution's API endpoint.

[1335] 2. Authenticate using an API key or OAuth token.

[1336] 3. Analyze the received JSON format financial information and extract each company's account balance, borrowing status, and repayment history.

[1337] 4. Save the analyzed data in a database.

[1338] Step 2:

[1339] Collecting industry information

[1340] The server collects industry information from relevant research institutes via API or FTP. It authenticates using API keys, usernames, and passwords as input, and obtains industry production figures, market shares, industry rankings, etc. The obtained industry information is analyzed and stored in a database.

[1341] Specific behavior:

[1342] 1. The server sends a request to the research organization's API endpoint.

[1343] 2. Authenticate if necessary (API key or username / password).

[1344] 3. Analyze the received industry information (e.g., production volume data) and extract the necessary information.

[1345] 4. Save the extracted data in the database.

[1346] Step 3:

[1347] Collecting SNS information

[1348] The server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company. It authenticates using the social media API key as input and obtains tweet data containing the company name and related keywords. It then analyzes the obtained tweet data using text mining technology, converting positive / negative evaluations into numerical values ​​and storing them in a database.

[1349] Specific behavior:

[1350] 1. The server sends a request to the Twitter API.

[1351] 2. Obtain tweets containing the company name and related keywords.

[1352] 3. Analyze tweet data using text mining technology and perform sentiment analysis.

[1353] 4. Quantify the negative / positive ratings and store them in a database.

[1354] Step 4:

[1355] Creating a unified dataset

[1356] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. It obtains each of the above information as input from a database, completes missing values, and normalizes the data. It then integrates the preprocessed data to generate a dataset for the machine learning model.

[1357] Specific behavior:

[1358] 1. The server retrieves the collected bank information, industry information, and social media ratings from the database.

[1359] 2. Impute missing values ​​(e.g., impute by the mean) and normalize the data.

[1360] 3. Integrate various information to generate a single data set.

[1361] Step 5:

[1362] Evaluation with machine learning models

[1363] The server inputs the preprocessed integrated dataset into the machine learning model to perform internal credit assessment and rating of the company. Using the integrated dataset as input, the machine learning model outputs the company's credit score and rating results. The calculation results are stored in a database.

[1364] Specific behavior:

[1365] 1. The server reads the consolidated data set.

[1366] 2. Input the integrated dataset into a machine learning model.

[1367] 3. Machine learning models calculate a company's credit score and rating.

[1368] 4. Save the calculation results in the database.

[1369] Step 6:

[1370] Emotion recognition with emotion engine

[1371] The server uses an emotion engine to recognize the user's emotions. It receives the user's voice and text input as input and analyzes their emotions using the emotion engine. It customizes the display format of the evaluation results and trend predictions according to the user's emotional state.

[1372] Specific behavior:

[1373] 1. The server receives the user's voice or text input.

[1374] 2. The emotion engine performs analysis to recognize the user's emotional state.

[1375] 3. Based on the recognition results, adjust the display format of the evaluation results and trend predictions.

[1376] Step 7:

[1377] Check the evaluation results

[1378] The user visually checks the evaluation results and trend predictions using a dedicated dashboard terminal. The terminal receives the evaluation results provided by the server as input and outputs the content in a customized display format according to the user's emotional state.

[1379] Specific behavior:

[1380] 1. The user logs in to the dashboard terminal.

[1381] 2. The server verifies the user's authentication and displays the dashboard content.

[1382] 3. The dashboard displays the latest assessment results and trend forecasts.

[1383] 4. The display format is customized according to the user's emotional state.

[1384] Step 8:

[1385] Funding decisions

[1386] The user creates a fundraising strategy based on the provided evaluation results and market trends. Using the information confirmed on the dashboard as input, the user submits a new loan application and outputs the details of raising funds at appropriate interest rates.

[1387] Specific behavior:

[1388] 1. The user checks the evaluation results displayed on the dashboard.

[1389] 2. Consider financing scenarios based on market trends.

[1390] 3. If necessary, file a new loan application.

[1391] 4. Appropriate interest rates will be set based on the loan application.

[1392] Step 9:

[1393] Trend forecast updates

[1394] The server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. It collects new financial, industry, and social media information as input and retrains the machine learning model using the updated dataset. It saves the latest prediction results in the database and updates the information provided to users.

[1395] Specific behavior:

[1396] 1. The server periodically collects new financial, industry, and social media information.

[1397] 2. Update the integrated dataset based on the collected data.

[1398] 3. Retrain the machine learning model (update the generative AI model).

[1399] 4. Store the latest evaluation results and trend forecasts in a database.

[1400] (Application example 2)

[1401] 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."

[1402] Conventional corporate lending services collect and analyze financial information, industry information, and social media information separately, making it difficult to provide comprehensive, real-time credit assessments. Furthermore, they do not take into account the user's emotional state, meaning that assessment results and forecast information are not optimized to fit the user's situation. This makes it difficult to make accurate lending decisions and manage transaction risks.

[1403] The specific processing by the specific 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 financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; and means for collecting word-of-mouth and evaluation information about companies from various social networking sites. This enables financial information, industry information, and social networking site information to be integrated and a machine learning model to perform internal credit assessment and rating of companies. Furthermore, by including an emotion engine that recognizes user emotions and customizes the display format of evaluation results and trend forecasts, it is possible to provide information tailored to the user. Furthermore, by having the machine learning model perform credit assessment of each transaction in real time and periodically retrain, the assessment results and trend forecasts are always kept up to date, enabling highly accurate lending decisions and transaction risk management.

[1404] "Bank information" refers to financial information such as a company's account balance, borrowing status, repayment history, etc. This information is necessary to evaluate a company's financial condition.

[1405] "Industry information" is data that includes production volume, market share, rankings, etc. in a specific industry. This information is necessary to understand the overall trends in the industry to which a company belongs.

[1406] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social networking services. This information is used to understand the evaluations and feelings of the general public and customers toward companies.

[1407] A "machine learning model" is an algorithm that performs internal credit assessment and rating of a company based on collected data, making it possible to evaluate a company's credit score and risk.

[1408] The "emotion engine" is a system that recognizes the user's emotions. It analyzes the user's voice and text input and customizes the display format of evaluation results and trend predictions according to the user's emotional state.

[1409] "Evaluation results" are the credit scores and internal credit results of companies obtained using machine learning models, which are used to determine the transaction risk and lending decisions for companies.

[1410] "Trend forecasts" are information used to predict a company's future trends. They use collected data to predict a company's future risks and growth potential.

[1411] "Real-time evaluation" refers to a credit evaluation that is conducted in real time for each transaction, allowing for accurate evaluation based on the latest data at all times.

[1412] "Retraining" is the process of updating a machine learning model based on newly collected data, ensuring that its assessment results and trend predictions are always up-to-date.

[1413] A system for implementing the present invention operates in cooperation with multiple components including a server, a user terminal, a machine learning model, and an emotion engine.

[1414] Data collection

[1415] First, the server collects financial information such as a company's account balance, borrowing status, and repayment history from the bank via API. The server authenticates with the financial institution using an API key or OAuth token, periodically obtains the latest data, and stores it in a database. Industry information, such as production volume, market share, and industry rankings, is also collected from related research institutions via API or FTP. Furthermore, social media information, such as reviews and ratings of companies, is collected using the API of social media platforms, and text data is analyzed and quantified using text mining technology.

[1416] Data Integration and Evaluation

[1417] The server integrates financial, industry, and social media information and preprocesses the data. Specifically, it completes missing values ​​and normalizes them to generate an input dataset for the machine learning model. The generated dataset is then fed into a pre-trained machine learning model to generate a company's internal credit rating and credit rating. The machine learning model is built using machine learning libraries such as TensorFlow and scikit-learn.

[1418] Emotion Recognition and Customization

[1419] The server uses an emotion engine to analyze the user's voice and text input. This emotion engine uses natural language processing libraries such as TextBlob to evaluate the user's emotions with a positive / negative score. The evaluation results and trend predictions are displayed in a customized format according to the user's emotional state.

[1420] Displaying results and making funding decisions

[1421] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The evaluation results are displayed in a format that is customized according to the user's emotional state, allowing the user to make more appropriate decisions. For example, a company's financial officer can look at the evaluation results and apply for new loans, ensuring that funds are raised at appropriate interest rates.

[1422] Specific examples of programs

[1423] A specific example of the embodiment is as follows:

[1424] When a company's financial officer initiates a new transaction, the server collects the company's latest financial, industry, and social media information, and inputs the integrated data into a machine learning model to calculate a credit score. At this point, an emotion engine analyzes the officer's emotional state and displays the evaluation results in an optimal format. This allows the financial officer to make appropriate financing decisions based on the evaluation results, such as avoiding high-risk transactions.

[1425] Example prompt sentence:

[1426] When doing new business, I want to see a company's current credit score and market trends, and if my feelings indicate concern, I want the system to provide feedback on its risk assessment as well.

[1427] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1428] Step 1:

[1429] The server collects financial information such as a company's account balance, borrowing status, and repayment history from financial institutions via API. At this time, the server authenticates with the financial institution using an API key or OAuth token, and periodically retrieves the latest financial information into a database. The input is financial information obtained from the financial institution's API, and the output is financial information stored in the server's database. By collecting financial information, data can be obtained to understand the company's current financial situation.

[1430] Step 2:

[1431] The server collects industry information from related research institutes via API or FTP. Specifically, it obtains data such as production volume, market share, and industry rankings. The input is industry information obtained from the research institute's API, and the output is industry information stored in the server's database. By collecting industry information, data can be obtained to understand trends in the industry to which a company belongs.

[1432] Step 3:

[1433] The server uses the API of the SNS platform to collect word-of-mouth and evaluation information about the company. The acquired text data is analyzed using text mining technology and quantified as positive / negative evaluations. The input is text data obtained from the SNS API, and the output is quantified SNS evaluation information. By collecting and analyzing SNS information, data can be obtained that visualizes the evaluation of the company by the general public and customers.

[1434] Step 4:

[1435] The server integrates the collected financial, industry, and social media information and preprocesses this data. Specifically, it imputes missing values ​​and normalizes the data to generate an input dataset for the machine learning model. The input is financial, industry, and social media information, and the output is a preprocessed integrated dataset. Preprocessing the data prepares the machine learning model for optimal operation.

[1436] Step 5:

[1437] The server inputs the preprocessed integrated dataset into a machine learning model to perform internal credit assessment and rating of the company. The input is the preprocessed integrated dataset, and the output is the company's credit score and risk assessment results. Using a machine learning model for evaluation makes it possible to quantitatively evaluate the company's credit status.

[1438] Step 6:

[1439] The server uses an emotion engine to analyze the user's voice and text input. Specifically, it uses a natural language processing library such as TextBlob to analyze emotions and obtains the results as a score. The input is the user's voice or text data, and the output is an emotion score. Using the emotion engine, it is possible to accurately recognize the user's emotions and provide feedback accordingly.

[1440] Step 7:

[1441] The server provides users with the evaluation results and trend predictions obtained from the machine learning model. At this time, the display format of the evaluation results is customized according to the user's emotional score. The inputs are the credit score, risk evaluation results, and emotional score, and the output is a display of the customized evaluation results. Users can visually check these results using a dashboard terminal, enabling them to make more appropriate decisions.

[1442] Step 8:

[1443] Users determine transaction risks and funding strategies based on the provided evaluation results and trend forecasts. This allows them to make economic decisions such as procuring funds at appropriate interest rates or avoiding high-risk transactions. The input is customized evaluation results, and the output is specific funding decisions. Evaluation displays based on the user's emotional state enable appropriate decisions.

[1444] Step 9:

[1445] The server periodically collects new data and retrains the machine learning model, allowing it to always provide the latest evaluation results and trend forecasts. The input is newly collected financial, industry, and social media information, and the output is an updated machine learning model and evaluation results. Periodic retraining improves the accuracy of the evaluation, enabling the provision of highly reliable information.

[1446] 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.

[1447] 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.

[1448] 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.

[1449] [Fourth embodiment]

[1450] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1451] 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.

[1452] 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).

[1453] 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.

[1454] 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.

[1455] 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).

[1456] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1457] 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.

[1458] 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.

[1459] 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.

[1460] 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.

[1461] 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.

[1462] 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."

[1463] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[1464] 1. The server collects your banking information

[1465] First, the server collects financial information such as a company's account balance, loan status, and repayment history from partner financial institutions via API. After going through authentication procedures, the necessary data is periodically retrieved and stored in a database. Security protection using API keys and OAuth tokens is important during this process.

[1466] Examples:

[1467] The server obtains the account balance of Company A and periodically stores updated repayment history data in a database.

[1468] 2. The server collects industry information

[1469] Next, the server collects industry information such as production numbers, market shares, rankings, etc. from relevant research organizations and industry associations, including accessing research organizations' databases and APIs to understand the latest industry trends.

[1470] Examples:

[1471] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[1472] 3. The server collects social media information

[1473] The server uses the API of the social media platform to collect reviews and ratings of companies, and then uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators.

[1474] Examples:

[1475] The server uses the Twitter API to collect tweets about Company A and quantify the negative / positive ratings.

[1476] 4. The server creates the integrated data set

[1477] The server integrates bank information, industry information, and social media information to generate a dataset for input into the machine learning model, and then preprocesses the data, imputing missing values ​​and standardizing them.

[1478] Examples:

[1479] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[1480] 5. The server evaluates the results using a machine learning model.

[1481] The server inputs the integrated data set into a machine learning model to generate internal credit ratings for businesses, which then predict their credit scores and future financing needs.

[1482] Examples:

[1483] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[1484] 6. The user checks the evaluation results

[1485] Corporate personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. The evaluation results are displayed visually, making them easy for users to understand.

[1486] Examples:

[1487] Company A's financial officer accesses the dashboard and views graphs of the company's credit score and the latest trend forecasts.

[1488] 7. Users make funding decisions

[1489] The company's finance department will consider future fundraising strategies based on the provided evaluation results and market trends, thereby achieving efficient and low-risk fundraising.

[1490] Examples:

[1491] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[1492] 8. The server updates the trend forecast

[1493] The server regularly collects new data and retrains the machine learning model to keep the assessment results and trend predictions up to date, ensuring that both companies and financial institutions are always provided with accurate and up-to-date information.

[1494] Examples:

[1495] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[1496] The system of this invention will enable companies to more easily predict future funding needs and interest rate trends, and financial institutions to make more accurate lending decisions. This is expected to strengthen mutual trust and contribute to stable economic growth.

[1497] The processing flow will be explained below.

[1498] Step 1: Gather your banking information

[1499] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. Before this process begins, the server authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[1500] Specific behavior:

[1501] The server sends a request to an API endpoint to retrieve account balances and debit data filtered by company ID.

[1502] The received data is analyzed in JSON format and the necessary data is saved in the database.

[1503] Step 2: Gather industry information

[1504] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[1505] Specific behavior:

[1506] The server sends a request to the research organization's API to retrieve the latest industry data.

[1507] The acquired data is parsed, processed, and formatted before being saved in a database.

[1508] Step 3: Collect social media information

[1509] The server uses the APIs of various social media platforms to collect reviews and ratings related to companies, and then uses text mining technology to analyze and quantify these reviews.

[1510] Specific behavior:

[1511] The server queries social media APIs using specific keywords or hashtags to gather relevant posts.

[1512] The collected data is subjected to text analysis, positive / negative indicators are calculated, and the evaluation information is quantified and stored in a database.

[1513] Step 4: Creating a consolidated dataset

[1514] The server integrates bank information, industry information, and social media information to create an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[1515] Specific behavior:

[1516] The server merges information from different data sources and converts it into a unified format.

[1517] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[1518] Step 5: Evaluation with machine learning models

[1519] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1520] Specific behavior:

[1521] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[1522] The scoring results are stored in a database and made available for subsequent processes.

[1523] Step 6: Check the evaluation results

[1524] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend forecasts, which are displayed in graphs and charts.

[1525] Specific behavior:

[1526] The user logs in to the dashboard and checks their company's evaluation results.

[1527] Interact with interactive graphs and charts to get more information.

[1528] Step 7: Funding Decision

[1529] The system develops a funding strategy based on the evaluation results and market trends provided by the user. The evaluation results include forecasts of future funding needs and interest rate trends.

[1530] Specific behavior:

[1531] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[1532] If necessary, submit a new loan application through the online form.

[1533] Step 8: Update trend forecasts

[1534] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[1535] Specific behavior:

[1536] The server recollects new financial, industry, and social media information monthly or weekly.

[1537] Retrain the machine learning model on the new dataset and update the evaluation results.

[1538] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions.

[1539] Example 1

[1540] 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."

[1541] In conventional corporate loan service systems, information collection and analysis from individual data sources was fragmented, making it difficult to conduct comprehensive corporate evaluations. Furthermore, data updates were delayed or intermittent, making it impossible to provide the latest evaluation results and trend forecasts. These issues hindered appropriate credit evaluations of companies and accurate lending decisions by financial institutions.

[1542] 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.

[1543] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from social media platforms; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on companies; means for preprocessing the integrated dataset, completing missing values, and standardizing data; means for providing evaluation results and trend forecasts to companies using the machine learning model; and means for visually displaying the evaluation results and trend forecasts using a dedicated dashboard. This makes it possible to always provide accurate and comprehensive corporate evaluations and the latest trend forecasts.

[1544] "Financial information" refers to data on a company's financial status collected from banks, such as the company's account balance, borrowing status, and repayment history.

[1545] "Industry information" refers to trends and statistical data about a specific industry, such as production numbers, market share, and rankings, and is collected from research organizations and industry associations.

[1546] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various SNS platforms, and is converted into evaluation indicators using text mining technology.

[1547] A "machine learning model" is an algorithm that performs internal credit assessments and ratings on companies based on collected data, and is used to provide evaluation results and trend predictions.

[1548] An "integrated dataset" is a single dataset that integrates financial information, industry information, and social media information, and is used as input data for machine learning models.

[1549] "Preprocessing" refers to the process of formatting input data for a machine learning model, and includes tasks such as filling in missing values ​​and standardizing data.

[1550] "Evaluation results" refer to a company's credit score or internal credit calculated by a machine learning model, and are part of the information provided.

[1551] "Trend forecasts" refer to the results of machine learning models predicting a company's future capital needs and market trends.

[1552] The "dedicated dashboard" is an interface that allows users to visually check evaluation results and trend forecasts, and displays information in graph and table format.

[1553] This invention is a system that provides corporate loan services, and collects and integrates bank information, industry information, and social media information, and performs internal credit and rating for companies using machine learning models. This system realizes fundraising and supply that is beneficial to both companies and financial institutions by linking the functions of the server, terminal, and user.

[1554] Server Roles and Functions

[1555] In this invention, the server collects financial information, industry information, and social media information of companies, and integrates them to generate a dataset to be input into the machine learning model. Specifically, the following hardware and software are used:

[1556] Hardware: High-performance server (e.g., with Xeon processor)

[1557] Software: API access libraries (e.g., requests), database software (e.g., MySQL, PostgreSQL), data processing libraries (e.g., Pandas, NumPy), machine learning libraries (e.g., Scikit-learn, TensorFlow)

[1558] The server processes and collects information as follows:

[1559] 1. Collection of financial information: The server obtains financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via the API, after going through authentication procedures, and stores the information in a database.

[1560] 2. Collecting industry information: The server accesses databases and APIs of research organizations and industry associations, downloads industry information such as production numbers, market shares, and rankings, and stores it in a database.

[1561] 3. Collection of social media information: The server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company, analyzes it using text mining technology, converts it into quantitative evaluation indicators, and stores them in a database.

[1562] 4. Generating an integrated dataset: Integrating bank information, industry information, and social media information, and preprocessing the data. Specifically, we impute missing values, normalize the data, and convert it into a format suitable for machine learning models.

[1563] Specific examples

[1564] For example, the server retrieves Company A's account balance and repayment history data via API and stores it in a MySQL database. Next, it downloads the latest production volume data for the automobile industry from a research institute and imports it into the database in CSV format using Pandas. Furthermore, it retrieves tweets related to Company A using the Twitter API, performs sentiment analysis using NLTK, and stores the results in a PostgreSQL database.

[1565] Roles and functions of user devices

[1566] The user terminal provides an interface that allows company personnel and financial institution analysts to access a dedicated dashboard and check evaluation results and trend forecasts. The following hardware and software are used.

[1567] Hardware: A standard computer or tablet

[1568] Software: Web browser (e.g., Google Chrome, Mozilla Firefox), dashboard software (e.g., Django, Plotly)

[1569] After logging in, users can access the dashboard and visually check the evaluation results and trend forecasts. The results are displayed in graph and table format, making them easy to understand.

[1570] Specific examples

[1571] A financial officer at Company A opens a web browser, accesses the dashboard URL, and enters his or her login information for authentication. The dashboard then retrieves the latest valuation results from the database and uses Plotly to generate and display graphs.

[1572] Prompt Sentence Examples

[1573] For example, if you want to use a generative AI model to predict Company A's credit score and future financing needs, you might use a prompt like this:

[1574] "To predict Company A's latest credit score and future financing needs, use a dataset that combines up-to-date banking, industry, and social media information."

[1575] The system of the present invention provides accurate and up-to-date information to companies and financial institutions, and is an important tool for achieving efficient and low-risk financing. It is implemented using a specific combination of hardware and software, and users can easily check the evaluation results on a dedicated dashboard.

[1576] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1577] Step 1:

[1578] Server collects banking information

[1579] Input: Financial institution API authentication information, company identification information

[1580] Specific operation: The server uses the API key and OAuth token to connect to the financial institution's API and obtain financial information such as the company's account balance, borrowing status, repayment history, etc. The server sends an HTTP request and receives a response in JSON format.

[1581] Data processing: Parse the received JSON data, extract the necessary information, and save the extracted data in the database.

[1582] Output: A database containing financial information about the company.

[1583] Step 2:

[1584] The server collects industry information

[1585] Input: API endpoint and authentication information of research institute or industry association

[1586] Specific operation: The server sends an HTTP request to the research organization's API to obtain industry information such as production numbers, market share, and rankings.

[1587] Data processing: The acquired data is received in CSV format, parsed using the Pandas library, and the necessary information is extracted. The extracted data is then saved in a database.

[1588] Output: A database containing up-to-date industry information.

[1589] Step 3:

[1590] The server collects SNS information

[1591] Input: API endpoints of social media platforms, keywords and hashtags related to your business

[1592] Specific operation: The server connects to the API of the social media platform and collects reviews and ratings information about the company.

[1593] Data processing: The collected text data is analyzed using text mining techniques such as the Natural Language Toolkit (NLTK) and classified into negative and positive evaluations. The classification results are stored in a database.

[1594] Output: A database containing social media reputation data related to companies.

[1595] Step 4:

[1596] The server creates a consolidated dataset

[1597] Input: Database containing financial information, industry information, and social media information

[1598] Specific operation: The server extracts the necessary information from the database and integrates it into a Pandas DataFrame. It imputes missing values ​​using SimpleImputer, normalizes the data using MinMaxScaler, and performs OneHot encoding on categorical data.

[1599] Data processing: Integrate various information and generate a pre-processed dataset.

[1600] Output: A consolidated dataset for machine learning model input.

[1601] Step 5:

[1602] The server evaluates using a machine learning model

[1603] Input: Integrated dataset

[1604] How it works: The server inputs the integrated dataset into the TensorFlow model to perform internal credit assessment and rating for the company. The model processes the data and calculates the company's credit score.

[1605] Data Computing: Predict credit scores and future financing needs based on integrated data sets.

[1606] Output: Company rating results and credit score.

[1607] Step 6:

[1608] The user checks the evaluation results

[1609] Input: Database containing evaluation results, user authentication information

[1610] How it works: Users access a dedicated dashboard and enter their login information to be authenticated. The dashboard retrieves the evaluation results from the database and generates and displays graphs using Plotly.

[1611] Data calculation: Checks access rights through user authentication and visualizes the evaluation results.

[1612] Output: Visualized evaluation results and trend forecasts.

[1613] Step 7:

[1614] Users make funding decisions

[1615] Input: Visualized evaluation results and trend forecasts

[1616] Specific operation: The user (company's finance department) considers fundraising strategies based on the information in the dashboard.

[1617] Data calculation: Based on the provided evaluation results and trend forecasts, companies can make new loan applications.

[1618] Output: New loan application, interest rate terms set.

[1619] Step 8:

[1620] Server updates trend forecast

[1621] Input: API endpoints to collect new information, existing datasets

[1622] Specific operation: The server periodically collects new data (repeating steps 1 to 3). It uses the new data to retrain the machine learning model and generate the latest evaluation results.

[1623] Data computation: Retraining the model with new data and updating the evaluation results.

[1624] Output: A database containing the latest assessment results and trend forecasts.

[1625] (Application example 1)

[1626] 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."

[1627] Traditional corporate internal credit and rating systems are limited to evaluations based on financial, industry, and social media information, and do not address the specific business operations and efficiency of each company's production plans. As a result, when companies seek financing, they face the dual challenges of fundraising and production management, making it difficult to allocate resources effectively. In particular, optimizing capital management and production plans is a key issue for in-factory production management systems.

[1628] 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.

[1629] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media sites; means for integrating the financial information, industry information, and social media information and using a machine learning model to perform internal credit and rating on the company; means for providing the evaluation results and trend forecasts to an in-factory production management system to support the optimization of production plans and cash management; and means for regularly retraining the machine learning model to ensure the latest evaluation results and trend forecasts. This enables companies to comprehensively utilize financial information, industry information, and social media information to allocate resources economically, thereby optimizing in-factory production plans and streamlining cash management.

[1630] "Financial information" refers to information collected from banks, including a company's account balance, borrowing status, repayment history, etc.

[1631] "Industry information" refers to data on industry trends, such as production volume, market share, and rankings, obtained from research organizations and industry associations.

[1632] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social networking services.

[1633] A "machine learning model" is an algorithm that uses collected data to assign internal credit and ratings to companies.

[1634] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and financial condition.

[1635] A "rating" is a method of quantitatively evaluating and ranking a company's creditworthiness.

[1636] "Evaluation results" refer to a company's credit score and trend predictions calculated using a machine learning model.

[1637] "Trend forecasting" refers to predicting future corporate credit status and market trends based on machine learning models.

[1638] A "production management system" is a system for managing production plans and resource allocation within a factory.

[1639] "Production planning" refers to planning for production schedules and resource allocation within a factory.

[1640] "Cash management" is the process of optimally managing working capital within a factory.

[1641] This invention is a system that collects and integrates bank information, industry information, and social media information, and uses machine learning models to perform internal credit and rating for companies. This system improves the efficiency of corporate cash management and production planning by linking the functions of the server, factory production management system, and user systems.

[1642] Server Roles

[1643] The server first collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. After going through authentication procedures, the necessary data is periodically obtained and stored in a database. Security protection using API keys and OAuth tokens is important during this process. As a concrete example, the server obtains Company A's account balance and periodically updates the repayment history data and stores it in a database.

[1644] Next, the server collects industry information such as production numbers, market shares, and rankings from relevant research organizations and industry associations. This involves accessing the research organizations' databases and APIs to understand the latest industry trends. For example, the server downloads the latest production number data for the automobile industry from a research organization and stores it in a database.

[1645] Furthermore, the server uses the API of the social media platform to collect word-of-mouth and evaluation information about the company. It uses text mining technology to analyze the text data and convert it into quantitative evaluation indicators. For example, the server uses the Twitter API to collect tweets about Company A and quantifies negative / positive evaluations.

[1646] The server combines this bank information, industry information, and social media information to generate a dataset for input into the machine learning model. It also performs data preprocessing, filling in missing values ​​and standardizing them. For example, the server combines information such as Company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[1647] This integrated data set is input into a machine learning model to perform internal credit assessment and rating of the company, thereby predicting the company's credit score and future funding needs. As a specific example, the server uses the integrated data set of Company A to calculate Company A's credit score using the machine learning model.

[1648] The role of production management systems

[1649] The evaluation results and trend forecasts are provided to the production management system to help optimize the company's cash management and production plans. Company personnel and analysts at financial institutions can use a dedicated dashboard terminal to check the evaluation results and trend forecasts. This makes the company's production plans more efficient and optimizes resource allocation. In a specific example, a financial officer at Company A accesses the dashboard to view graphs of the company's credit score and the latest trend forecasts.

[1650] Trend forecast updates

[1651] Finally, the server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. For example, the server collects new industry data and social media information monthly and updates the trend predictions for Company A with the latest generative AI model.

[1652] Prompt Sentence Examples

[1653] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[1654] This allows companies to easily optimize their future capital needs and production plans, thereby optimizing economic resource allocation.

[1655] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1656] Step 1:

[1657] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. Specifically, it uses an API key or OAuth token to access the financial institution's API and obtains the company's financial data. This input data is then stored in a database. For example, the server obtains Company A's account balance data and stores it in the database along with regularly updated repayment history data.

[1658] Step 2:

[1659] The server collects industry information, such as production numbers, market share, and rankings, from relevant research organizations and industry associations. Specifically, it accesses research organization databases and APIs to obtain the latest industry trend data. This input data is later used as part of an integrated dataset. For example, the server downloads the latest production number data for the automotive industry from a research organization and stores it in a database.

[1660] Step 3:

[1661] The server uses the API of a social media platform to collect word-of-mouth and evaluation information about a company. Specifically, it uses the social media API to obtain text data and analyzes it using text mining technology. The input data are tweets and comments about the company, and the output data is a numerical representation of negative and positive evaluations. For example, the server uses the Twitter API to collect tweets about Company A and uses text mining technology to calculate the numerical value of positive evaluations.

[1662] Step 4:

[1663] The server integrates the bank information, industry information, and social media information mentioned above to generate a dataset to input into the machine learning model. Specifically, it preprocesses the data and performs missing value completion and standardization. The input data is each piece of information mentioned above, and the output data is the integrated dataset. For example, the server integrates information such as Company A's account balance, production volume, and social media ratings and processes it into a single dataset.

[1664] Step 5:

[1665] The server inputs the integrated dataset into a machine learning model to perform internal credit assessment and rating of the company. Specifically, the preprocessed data is input into the machine learning model to predict the company's credit score and future capital needs. The input data is the integrated dataset, and the output data is the company's credit score and trend prediction. For example, the server uses the integrated dataset of Company A to calculate Company A's credit score.

[1666] Step 6:

[1667] The server provides the evaluation results and trend forecasts to the production management system within the factory, supporting the optimization of production plans and financial management. Specifically, it accesses a dedicated dashboard terminal and displays the evaluation results and forecast data in graph and report format. The input data is the evaluation results and trend forecasts, and the output data is information displayed visually. For example, a financial officer at Company A accesses the dashboard to check information on the company's credit score and production plan optimization.

[1668] Step 7:

[1669] The server periodically collects new data and retrains the machine learning model to keep the evaluation results and trend predictions up to date. Specifically, new bank information, industry information, and social media information are collected monthly or weekly, and the new data is re-input into the machine learning model to retrain the model. The input data is the newly collected information, and the output data is the latest evaluation results and trend predictions. For example, the server collects the latest industry data and social media information and updates the trend predictions for Company A with a new generative AI model.

[1670] Prompt Sentence Examples

[1671] For example, a prompt might read, "Develop an application to evaluate the best financing method for installing a new production line and optimize the production schedule. You will need to integrate banking data, industry data, and social media data to make predictions using a machine learning model. Please also provide details on API keys and database integration methods."

[1672] 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.

[1673] This invention is a system for corporate loan services that collects and integrates bank information, industry information, and social media information, performs internal credit assessment and rating for companies using machine learning models, and further combines an emotion engine to recognize user emotions and customize evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising by linking the functions of the server, terminal, and user.

[1674] 1. Collection of banking information

[1675] The server collects financial information such as a company's account balance, borrowing status, and repayment history from affiliated financial institutions via API. The server authenticates with the financial institution using an API key or OAuth token, periodically retrieves the data, and stores it in a database.

[1676] Examples:

[1677] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[1678] 2. Collecting industry information

[1679] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[1680] Examples:

[1681] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[1682] 3. Collecting social media information

[1683] The server uses the API of the social media platform to collect word-of-mouth and evaluation information about companies, and then analyzes and quantifies the text data using text mining technology.

[1684] Examples:

[1685] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[1686] 4. Creation of an integrated dataset

[1687] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[1688] Examples:

[1689] The server combines information such as company A's account balance, production volume, and social media ratings into a single dataset and handles missing values ​​appropriately.

[1690] 5. Evaluation with machine learning models

[1691] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and credit rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1692] Examples:

[1693] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machine learning model.

[1694] 6. Emotion Recognition with Emotion Engine

[1695] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's voice and text inputs and recognizes their emotions to customize the display format of the evaluation results and trend predictions.

[1696] Examples:

[1697] The server uses an emotion engine to analyze the user's voice input and adjusts the display of the evaluation results based on the results.

[1698] 7. Checking the evaluation results

[1699] Users can visually check the evaluation results and trend forecasts using a dedicated dashboard terminal. The display format of the evaluation results is customized according to the user's emotional state.

[1700] Examples:

[1701] Company A's finance manager accesses a dashboard to view the company's credit score and latest trend forecasts in an emotional format.

[1702] 8. Funding Decisions

[1703] Users can develop fundraising strategies based on the provided evaluation results and market trends. The display format reflects the user's emotional state, allowing them to make more appropriate decisions.

[1704] Examples:

[1705] Based on the evaluation results, Company A's financial officer will apply for a new loan and raise funds at an appropriate interest rate.

[1706] 9. Trend forecast updates

[1707] The server regularly collects new data and retrains the machine learning models to keep the assessment results and trend predictions up to date, ensuring accurate and up-to-date information.

[1708] Examples:

[1709] The server collects new industry data and social media information monthly and updates trend forecasts for Company A using the latest generative AI model.

[1710] The system of this invention allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides advice and displays tailored to each user, improving user satisfaction and enabling more effective fundraising.

[1711] The processing flow will be explained below.

[1712] Step 1: Gather your banking information

[1713] The server uses the financial institution's API to obtain financial information such as the company's account balance, borrowing status, repayment history, etc. During this process, the server first authenticates with the financial institution using an API key or OAuth token to obtain the appropriate access rights.

[1714] Specific behavior:

[1715] The server sends a request to an API endpoint to retrieve account balances and debit data based on the company ID.

[1716] The received data is analyzed in JSON format and the necessary data is saved in the database.

[1717] Step 2: Gather industry information

[1718] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, rankings, etc.

[1719] Specific behavior:

[1720] The server sends a request to the research organization's API to retrieve the latest industry data.

[1721] The acquired data is parsed, processed, and formatted before being saved in a database.

[1722] Step 3: Collect social media information

[1723] The server uses the APIs of various social media platforms to collect word-of-mouth and evaluation information related to companies and analyzes it using text mining technology.

[1724] Specific behavior:

[1725] The server queries social media APIs using specific keywords or hashtags to gather relevant posts.

[1726] The collected data is subjected to text analysis, and positive / negative evaluations are quantified and stored in a database.

[1727] Step 4: Creating a consolidated dataset

[1728] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data.

[1729] Specific behavior:

[1730] The server merges information from different data sources and converts it into a unified format.

[1731] If there is missing data, it is filled in with appropriate values ​​and the data is scaled and normalized.

[1732] Step 5: Evaluation with machine learning models

[1733] The server inputs the pre-processed integrated data set into a machine learning model to calculate the company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1734] Specific behavior:

[1735] The server inputs the integrated data set into a generative AI model to calculate a company's credit score.

[1736] The scoring results are stored in a database and made available for subsequent processes.

[1737] Step 6: Emotion Recognition in the Emotion Engine

[1738] The server analyzes the user's voice and text input and uses an emotion engine to recognize emotions, thereby customizing the display and system responses to the user's emotional state.

[1739] Specific behavior:

[1740] The server passes the user's voice and text input to the emotion engine and obtains the analysis results.

[1741] Based on the analysis results, the dashboard display format and response content are adjusted.

[1742] Step 7: Check the evaluation results

[1743] Users can use a dedicated dashboard terminal to visually check the evaluation results and trend predictions, and the display content is customized according to the user's emotional state.

[1744] Specific behavior:

[1745] Users log in to a dashboard to see their company's credit score and the latest trend forecasts.

[1746] Interact with interactive graphs and charts to get more information.

[1747] Step 8: Funding Decision

[1748] The system creates a fundraising strategy based on the evaluation results and market trends provided by the user. The system supports appropriate decisions by displaying the results in a format that responds to the user's emotions.

[1749] Specific behavior:

[1750] The user reviews the evaluation results and conducts meetings and plans to formulate the optimal financing plan.

[1751] If necessary, submit a new loan application through the online form.

[1752] Step 9: Update trend forecast

[1753] The server periodically collects new data and retrains the machine learning model, ensuring that the assessment results and trend predictions are always up-to-date.

[1754] Specific behavior:

[1755] The server recollects new financial, industry, and social media information monthly or weekly.

[1756] Retrain the machine learning model on the new dataset and update the evaluation results.

[1757] This system allows companies to more accurately predict future funding needs and interest rate trends, and financial institutions to make more reliable lending decisions. Furthermore, the emotion engine provides personalized advice and displays, improving user satisfaction and enabling more effective fundraising.

[1758] Example 2

[1759] 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."

[1760] Corporate lending services collect and integrate bank, industry, and social media information to assess a company's creditworthiness, but they often fail to accurately assess it due to insufficient consideration of data variability and emotions. Furthermore, the methods used to present assessment results and trend forecasts are unable to adapt to the user's emotions, sometimes resulting in insufficient support for user decision-making. This makes it difficult for financial institutions to make reliable lending decisions, making it difficult for companies to raise funds at reasonable interest rates.

[1761] 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.

[1762] In this invention, the server includes: means for collecting financial information such as a company's account balance, borrowing status, and repayment history from banks; means for collecting industry information such as industry production volume, market share, and rankings; means for collecting word-of-mouth and evaluation information about companies from various social media; means for integrating the financial information, industry information, and SNS information and performing internal credit and rating of the company using a machine learning model; means for recognizing user emotions using an emotion engine and customizing the display format of the evaluation results and trend forecasts; and means for providing the evaluation results and trend forecasts using the machine learning model to the company. This enables internal credit and rating of companies to be performed with high accuracy, and by presenting evaluation results that take user emotions into consideration, it becomes possible to support more appropriate decision-making.

[1763] "Bank information" refers to public financial data obtained from financial institutions, such as a company's account balance, borrowing status, and repayment history.

[1764] "Industry information" refers to data about a particular industry, such as industry production volumes, market shares, and rankings.

[1765] "SNS information" refers to word-of-mouth and evaluation information about companies collected from various social media platforms.

[1766] A "machine learning model" is an algorithm used to generate internal credit ratings and ratings for companies based on collected data.

[1767] An "emotion engine" is a system that analyzes a user's voice and text input and recognizes emotions.

[1768] "User" refers to a person who uses this system to check the evaluation results and trend forecasts of a company and make a decision on fundraising.

[1769] A "database" is a data infrastructure for storing and managing collected financial information, industry information, and social media information.

[1770] "Internal credit" refers to an internal credit assessment used to evaluate a company's creditworthiness and suitability for lending.

[1771] A "rating" is a grade calculated using a machine learning model to indicate a company's creditworthiness and financial stability.

[1772] MODE FOR CARRYING OUT THE INVENTION

[1773] This invention is a system for corporate loan services that links the functions of servers, terminals, and users to collect and integrate corporate financial information, industry information, and social media information, and uses machine learning models and emotion engines to perform internal credit assessments and ratings for companies, providing evaluation results and trend forecasts. This system supports more accurate loan decisions and fundraising.

[1774] Collection of banking information

[1775] The server collects financial information about the company from partner financial institutions via API. Specifically, it obtains financial data such as account balances, borrowing status, and repayment history. The server authenticates with the financial institution using an API key or OAuth token, periodically obtains the data, and stores it in a database.

[1776] Examples:

[1777] The server obtains the account balance of Company A and periodically saves the repayment history data in a database.

[1778] Collecting industry information

[1779] The server collects industry information from relevant research institutes via API or FTP, including production numbers, market shares, industry rankings, etc.

[1780] Examples:

[1781] The server downloads the latest production data for the automotive industry from research agencies and stores it in a database.

[1782] Collecting SNS information

[1783] The server uses the API of the social media platform to collect reviews and ratings of companies, and analyzes them using text mining technology. The acquired text data is then analyzed and quantified.

[1784] Examples:

[1785] The server uses the Twitter API to collect tweets about Company A, converts negative / positive ratings into numerical values, and stores them in a database.

[1786] Integrating data and using machine learning models

[1787] The server integrates bank information, industry information, and social media information to generate an input dataset for the machine learning model. Data preprocessing involves filling in missing values ​​and normalizing the data. The preprocessed integrated dataset is then input into the machine learning model to generate a company's internal credit rating and rating. The machine learning model is pre-trained and calculates credit scores, etc.

[1788] Examples:

[1789] The server uses the integrated dataset of Company A to calculate Company A's credit score using a machi...

Claims

1. A means of collecting financial information from banks, such as company account balances, borrowing status, and repayment history, and A means of collecting industry information such as industry production numbers, market shares, and rankings; A means of collecting word-of-mouth and evaluation information about companies from various social media platforms, A means for integrating the financial information, industry information, and SNS information and using a machine learning model to perform internal credit and rating of a company; A means for providing companies with evaluation results and trend predictions using the machine learning model; A system including:

2. 10. The system of claim 1, wherein the machine learning model calculates a credit score for a business based on the collected data.

3. 10. The system of claim 1, wherein the machine learning model is retrained based on newly collected data, thereby periodically updating the evaluation results and trend predictions.

Images